rahular commited on
Commit
300a419
1 Parent(s): 8dbb7fc

Upload ShukaPipeline

Browse files
README.md ADDED
@@ -0,0 +1,199 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: transformers
3
+ tags: []
4
+ ---
5
+
6
+ # Model Card for Model ID
7
+
8
+ <!-- Provide a quick summary of what the model is/does. -->
9
+
10
+
11
+
12
+ ## Model Details
13
+
14
+ ### Model Description
15
+
16
+ <!-- Provide a longer summary of what this model is. -->
17
+
18
+ This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated.
19
+
20
+ - **Developed by:** [More Information Needed]
21
+ - **Funded by [optional]:** [More Information Needed]
22
+ - **Shared by [optional]:** [More Information Needed]
23
+ - **Model type:** [More Information Needed]
24
+ - **Language(s) (NLP):** [More Information Needed]
25
+ - **License:** [More Information Needed]
26
+ - **Finetuned from model [optional]:** [More Information Needed]
27
+
28
+ ### Model Sources [optional]
29
+
30
+ <!-- Provide the basic links for the model. -->
31
+
32
+ - **Repository:** [More Information Needed]
33
+ - **Paper [optional]:** [More Information Needed]
34
+ - **Demo [optional]:** [More Information Needed]
35
+
36
+ ## Uses
37
+
38
+ <!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
39
+
40
+ ### Direct Use
41
+
42
+ <!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
43
+
44
+ [More Information Needed]
45
+
46
+ ### Downstream Use [optional]
47
+
48
+ <!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
49
+
50
+ [More Information Needed]
51
+
52
+ ### Out-of-Scope Use
53
+
54
+ <!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
55
+
56
+ [More Information Needed]
57
+
58
+ ## Bias, Risks, and Limitations
59
+
60
+ <!-- This section is meant to convey both technical and sociotechnical limitations. -->
61
+
62
+ [More Information Needed]
63
+
64
+ ### Recommendations
65
+
66
+ <!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
67
+
68
+ Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
69
+
70
+ ## How to Get Started with the Model
71
+
72
+ Use the code below to get started with the model.
73
+
74
+ [More Information Needed]
75
+
76
+ ## Training Details
77
+
78
+ ### Training Data
79
+
80
+ <!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
81
+
82
+ [More Information Needed]
83
+
84
+ ### Training Procedure
85
+
86
+ <!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
87
+
88
+ #### Preprocessing [optional]
89
+
90
+ [More Information Needed]
91
+
92
+
93
+ #### Training Hyperparameters
94
+
95
+ - **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
96
+
97
+ #### Speeds, Sizes, Times [optional]
98
+
99
+ <!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
100
+
101
+ [More Information Needed]
102
+
103
+ ## Evaluation
104
+
105
+ <!-- This section describes the evaluation protocols and provides the results. -->
106
+
107
+ ### Testing Data, Factors & Metrics
108
+
109
+ #### Testing Data
110
+
111
+ <!-- This should link to a Dataset Card if possible. -->
112
+
113
+ [More Information Needed]
114
+
115
+ #### Factors
116
+
117
+ <!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
118
+
119
+ [More Information Needed]
120
+
121
+ #### Metrics
122
+
123
+ <!-- These are the evaluation metrics being used, ideally with a description of why. -->
124
+
125
+ [More Information Needed]
126
+
127
+ ### Results
128
+
129
+ [More Information Needed]
130
+
131
+ #### Summary
132
+
133
+
134
+
135
+ ## Model Examination [optional]
136
+
137
+ <!-- Relevant interpretability work for the model goes here -->
138
+
139
+ [More Information Needed]
140
+
141
+ ## Environmental Impact
142
+
143
+ <!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
144
+
145
+ Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
146
+
147
+ - **Hardware Type:** [More Information Needed]
148
+ - **Hours used:** [More Information Needed]
149
+ - **Cloud Provider:** [More Information Needed]
150
+ - **Compute Region:** [More Information Needed]
151
+ - **Carbon Emitted:** [More Information Needed]
152
+
153
+ ## Technical Specifications [optional]
154
+
155
+ ### Model Architecture and Objective
156
+
157
+ [More Information Needed]
158
+
159
+ ### Compute Infrastructure
160
+
161
+ [More Information Needed]
162
+
163
+ #### Hardware
164
+
165
+ [More Information Needed]
166
+
167
+ #### Software
168
+
169
+ [More Information Needed]
170
+
171
+ ## Citation [optional]
172
+
173
+ <!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
174
+
175
+ **BibTeX:**
176
+
177
+ [More Information Needed]
178
+
179
+ **APA:**
180
+
181
+ [More Information Needed]
182
+
183
+ ## Glossary [optional]
184
+
185
+ <!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
186
+
187
+ [More Information Needed]
188
+
189
+ ## More Information [optional]
190
+
191
+ [More Information Needed]
192
+
193
+ ## Model Card Authors [optional]
194
+
195
+ [More Information Needed]
196
+
197
+ ## Model Card Contact
198
+
199
+ [More Information Needed]
config.json ADDED
@@ -0,0 +1,195 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "runs/llama3_saaras_enhi_v0.9.4",
3
+ "architectures": [
4
+ "ShukaModel"
5
+ ],
6
+ "audio_config": {
7
+ "_name_or_path": "sarvam/saaras_0_11",
8
+ "activation_dropout": 0.0,
9
+ "activation_function": "gelu",
10
+ "apply_spec_augment": true,
11
+ "architectures": [
12
+ "WhisperForConditionalGeneration"
13
+ ],
14
+ "attention_dropout": 0.0,
15
+ "begin_suppress_tokens": [
16
+ 220,
17
+ 50257
18
+ ],
19
+ "bos_token_id": 50257,
20
+ "d_model": 1280,
21
+ "decoder_attention_heads": 20,
22
+ "decoder_ffn_dim": 5120,
23
+ "decoder_layerdrop": 0.0,
24
+ "decoder_layers": 32,
25
+ "decoder_start_token_id": 50258,
26
+ "dropout": 0.0,
27
+ "encoder_attention_heads": 20,
28
+ "encoder_ffn_dim": 5120,
29
+ "encoder_layerdrop": 0.0,
30
+ "encoder_layers": 32,
31
+ "eos_token_id": 50257,
32
+ "forced_decoder_ids": [
33
+ [
34
+ 1,
35
+ 50259
36
+ ],
37
+ [
38
+ 2,
39
+ 50359
40
+ ],
41
+ [
42
+ 3,
43
+ 50363
44
+ ]
45
+ ],
46
+ "init_std": 0.02,
47
+ "is_encoder_decoder": true,
48
+ "max_length": 448,
49
+ "max_source_positions": 1500,
50
+ "max_target_positions": 448,
51
+ "median_filter_width": 7,
52
+ "model_type": "whisper",
53
+ "num_hidden_layers": 32,
54
+ "num_mel_bins": 80,
55
+ "pad_token_id": 50257,
56
+ "scale_embedding": false,
57
+ "suppress_tokens": [
58
+ 1,
59
+ 2,
60
+ 7,
61
+ 8,
62
+ 9,
63
+ 10,
64
+ 14,
65
+ 25,
66
+ 26,
67
+ 27,
68
+ 28,
69
+ 29,
70
+ 31,
71
+ 58,
72
+ 59,
73
+ 60,
74
+ 61,
75
+ 62,
76
+ 63,
77
+ 90,
78
+ 91,
79
+ 92,
80
+ 93,
81
+ 359,
82
+ 503,
83
+ 522,
84
+ 542,
85
+ 873,
86
+ 893,
87
+ 902,
88
+ 918,
89
+ 922,
90
+ 931,
91
+ 1350,
92
+ 1853,
93
+ 1982,
94
+ 2460,
95
+ 2627,
96
+ 3246,
97
+ 3253,
98
+ 3268,
99
+ 3536,
100
+ 3846,
101
+ 3961,
102
+ 4183,
103
+ 4667,
104
+ 6585,
105
+ 6647,
106
+ 7273,
107
+ 9061,
108
+ 9383,
109
+ 10428,
110
+ 10929,
111
+ 11938,
112
+ 12033,
113
+ 12331,
114
+ 12562,
115
+ 13793,
116
+ 14157,
117
+ 14635,
118
+ 15265,
119
+ 15618,
120
+ 16553,
121
+ 16604,
122
+ 18362,
123
+ 18956,
124
+ 20075,
125
+ 21675,
126
+ 22520,
127
+ 26130,
128
+ 26161,
129
+ 26435,
130
+ 28279,
131
+ 29464,
132
+ 31650,
133
+ 32302,
134
+ 32470,
135
+ 36865,
136
+ 42863,
137
+ 47425,
138
+ 49870,
139
+ 50254,
140
+ 50258,
141
+ 50358,
142
+ 50359,
143
+ 50360,
144
+ 50361,
145
+ 50362
146
+ ],
147
+ "torch_dtype": "float32",
148
+ "use_cache": true,
149
+ "vocab_size": 51865
150
+ },
151
+ "audio_model_id": null,
152
+ "audio_token_index": 32000,
153
+ "auto_map": {
154
+ "AutoConfig": "shuka_config.ShukaConfig",
155
+ "AutoModel": "shuka_model.ShukaModel"
156
+ },
157
+ "custom_pipelines": {
158
+ "shuka-pipeline": {
159
+ "impl": "shuka_pipeline.ShukaPipeline",
160
+ "pt": [
161
+ "AutoModel"
162
+ ],
163
+ "tf": [],
164
+ "type": "multimodal"
165
+ }
166
+ },
167
+ "encoder_ds_factor": 160,
168
+ "hidden_size": 4096,
169
+ "ignore_index": -100,
170
+ "initializer_range": 0.02,
171
+ "model_type": "shuka",
172
+ "norm_init": 0.4,
173
+ "projector_act": "silu",
174
+ "stack_factor": 8,
175
+ "text_config": {
176
+ "_name_or_path": "meta-llama/Meta-Llama-3-8B-Instruct",
177
+ "architectures": [
178
+ "LlamaForCausalLM"
179
+ ],
180
+ "bos_token_id": 128000,
181
+ "eos_token_id": 128009,
182
+ "intermediate_size": 14336,
183
+ "max_position_embeddings": 8192,
184
+ "model_type": "llama",
185
+ "num_key_value_heads": 8,
186
+ "rms_norm_eps": 1e-05,
187
+ "rope_theta": 500000.0,
188
+ "torch_dtype": "bfloat16",
189
+ "vocab_size": 128256
190
+ },
191
+ "text_model_id": null,
192
+ "torch_dtype": "bfloat16",
193
+ "transformers_version": "4.41.2",
194
+ "vocab_size": 128256
195
+ }
generation_config.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "bos_token_id": 128000,
4
+ "eos_token_id": 128009,
5
+ "pad_token_id": 128009,
6
+ "transformers_version": "4.41.2"
7
+ }
model-00001-of-00004.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:579097641f55dab4a03be00c73fcab08ab599e82ac1bbd96ff11cd08b4e0e95b
3
+ size 4941662064
model-00002-of-00004.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:15f03a7aa67f4aefd27c3fab1684b8a49c0636f2d572f6316371e043366b0d96
3
+ size 4999820896
model-00003-of-00004.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2e97a7b690d34b2901edd649da811c7bf7f388e56d865dad266e91a32e515620
3
+ size 4915917680
model-00004-of-00004.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dfbb9312a19aca5dc65ceefe16b6b64b6a408ad639e5346a2a763be8cff30d44
3
+ size 2594255080
model.safetensors.index.json ADDED
@@ -0,0 +1,789 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metadata": {
3
+ "total_size": 17451560960
4
+ },
5
+ "weight_map": {
6
+ "audio_tower.conv1.bias": "model-00001-of-00004.safetensors",
7
+ "audio_tower.conv1.weight": "model-00001-of-00004.safetensors",
8
+ "audio_tower.conv2.bias": "model-00001-of-00004.safetensors",
9
+ "audio_tower.conv2.weight": "model-00001-of-00004.safetensors",
10
+ "audio_tower.embed_positions.weight": "model-00001-of-00004.safetensors",
11
+ "audio_tower.layer_norm.bias": "model-00001-of-00004.safetensors",
12
+ "audio_tower.layer_norm.weight": "model-00001-of-00004.safetensors",
13
+ "audio_tower.layers.0.fc1.bias": "model-00001-of-00004.safetensors",
14
+ "audio_tower.layers.0.fc1.weight": "model-00001-of-00004.safetensors",
15
+ "audio_tower.layers.0.fc2.bias": "model-00001-of-00004.safetensors",
16
+ "audio_tower.layers.0.fc2.weight": "model-00001-of-00004.safetensors",
17
+ "audio_tower.layers.0.final_layer_norm.bias": "model-00001-of-00004.safetensors",
18
+ "audio_tower.layers.0.final_layer_norm.weight": "model-00001-of-00004.safetensors",
19
+ "audio_tower.layers.0.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
20
+ "audio_tower.layers.0.self_attn.out_proj.bias": "model-00001-of-00004.safetensors",
21
+ "audio_tower.layers.0.self_attn.out_proj.weight": "model-00001-of-00004.safetensors",
22
+ "audio_tower.layers.0.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
23
+ "audio_tower.layers.0.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
24
+ "audio_tower.layers.0.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
25
+ "audio_tower.layers.0.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
26
+ "audio_tower.layers.0.self_attn_layer_norm.bias": "model-00001-of-00004.safetensors",
27
+ "audio_tower.layers.0.self_attn_layer_norm.weight": "model-00001-of-00004.safetensors",
28
+ "audio_tower.layers.1.fc1.bias": "model-00001-of-00004.safetensors",
29
+ "audio_tower.layers.1.fc1.weight": "model-00001-of-00004.safetensors",
30
+ "audio_tower.layers.1.fc2.bias": "model-00001-of-00004.safetensors",
31
+ "audio_tower.layers.1.fc2.weight": "model-00001-of-00004.safetensors",
32
+ "audio_tower.layers.1.final_layer_norm.bias": "model-00001-of-00004.safetensors",
33
+ "audio_tower.layers.1.final_layer_norm.weight": "model-00001-of-00004.safetensors",
34
+ "audio_tower.layers.1.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
35
+ "audio_tower.layers.1.self_attn.out_proj.bias": "model-00001-of-00004.safetensors",
36
+ "audio_tower.layers.1.self_attn.out_proj.weight": "model-00001-of-00004.safetensors",
37
+ "audio_tower.layers.1.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
38
+ "audio_tower.layers.1.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
39
+ "audio_tower.layers.1.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
40
+ "audio_tower.layers.1.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
41
+ "audio_tower.layers.1.self_attn_layer_norm.bias": "model-00001-of-00004.safetensors",
42
+ "audio_tower.layers.1.self_attn_layer_norm.weight": "model-00001-of-00004.safetensors",
43
+ "audio_tower.layers.10.fc1.bias": "model-00001-of-00004.safetensors",
44
+ "audio_tower.layers.10.fc1.weight": "model-00001-of-00004.safetensors",
45
+ "audio_tower.layers.10.fc2.bias": "model-00001-of-00004.safetensors",
46
+ "audio_tower.layers.10.fc2.weight": "model-00001-of-00004.safetensors",
47
+ "audio_tower.layers.10.final_layer_norm.bias": "model-00001-of-00004.safetensors",
48
+ "audio_tower.layers.10.final_layer_norm.weight": "model-00001-of-00004.safetensors",
49
+ "audio_tower.layers.10.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
50
+ "audio_tower.layers.10.self_attn.out_proj.bias": "model-00001-of-00004.safetensors",
51
+ "audio_tower.layers.10.self_attn.out_proj.weight": "model-00001-of-00004.safetensors",
52
+ "audio_tower.layers.10.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
53
+ "audio_tower.layers.10.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
54
+ "audio_tower.layers.10.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
55
+ "audio_tower.layers.10.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
56
+ "audio_tower.layers.10.self_attn_layer_norm.bias": "model-00001-of-00004.safetensors",
57
+ "audio_tower.layers.10.self_attn_layer_norm.weight": "model-00001-of-00004.safetensors",
58
+ "audio_tower.layers.11.fc1.bias": "model-00001-of-00004.safetensors",
59
+ "audio_tower.layers.11.fc1.weight": "model-00001-of-00004.safetensors",
60
+ "audio_tower.layers.11.fc2.bias": "model-00001-of-00004.safetensors",
61
+ "audio_tower.layers.11.fc2.weight": "model-00001-of-00004.safetensors",
62
+ "audio_tower.layers.11.final_layer_norm.bias": "model-00001-of-00004.safetensors",
63
+ "audio_tower.layers.11.final_layer_norm.weight": "model-00001-of-00004.safetensors",
64
+ "audio_tower.layers.11.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
65
+ "audio_tower.layers.11.self_attn.out_proj.bias": "model-00001-of-00004.safetensors",
66
+ "audio_tower.layers.11.self_attn.out_proj.weight": "model-00001-of-00004.safetensors",
67
+ "audio_tower.layers.11.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
68
+ "audio_tower.layers.11.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
69
+ "audio_tower.layers.11.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
70
+ "audio_tower.layers.11.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
71
+ "audio_tower.layers.11.self_attn_layer_norm.bias": "model-00001-of-00004.safetensors",
72
+ "audio_tower.layers.11.self_attn_layer_norm.weight": "model-00001-of-00004.safetensors",
73
+ "audio_tower.layers.12.fc1.bias": "model-00001-of-00004.safetensors",
74
+ "audio_tower.layers.12.fc1.weight": "model-00001-of-00004.safetensors",
75
+ "audio_tower.layers.12.fc2.bias": "model-00001-of-00004.safetensors",
76
+ "audio_tower.layers.12.fc2.weight": "model-00001-of-00004.safetensors",
77
+ "audio_tower.layers.12.final_layer_norm.bias": "model-00001-of-00004.safetensors",
78
+ "audio_tower.layers.12.final_layer_norm.weight": "model-00001-of-00004.safetensors",
79
+ "audio_tower.layers.12.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
80
+ "audio_tower.layers.12.self_attn.out_proj.bias": "model-00001-of-00004.safetensors",
81
+ "audio_tower.layers.12.self_attn.out_proj.weight": "model-00001-of-00004.safetensors",
82
+ "audio_tower.layers.12.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
83
+ "audio_tower.layers.12.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
84
+ "audio_tower.layers.12.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
85
+ "audio_tower.layers.12.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
86
+ "audio_tower.layers.12.self_attn_layer_norm.bias": "model-00001-of-00004.safetensors",
87
+ "audio_tower.layers.12.self_attn_layer_norm.weight": "model-00001-of-00004.safetensors",
88
+ "audio_tower.layers.13.fc1.bias": "model-00001-of-00004.safetensors",
89
+ "audio_tower.layers.13.fc1.weight": "model-00001-of-00004.safetensors",
90
+ "audio_tower.layers.13.fc2.bias": "model-00001-of-00004.safetensors",
91
+ "audio_tower.layers.13.fc2.weight": "model-00001-of-00004.safetensors",
92
+ "audio_tower.layers.13.final_layer_norm.bias": "model-00001-of-00004.safetensors",
93
+ "audio_tower.layers.13.final_layer_norm.weight": "model-00001-of-00004.safetensors",
94
+ "audio_tower.layers.13.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
95
+ "audio_tower.layers.13.self_attn.out_proj.bias": "model-00001-of-00004.safetensors",
96
+ "audio_tower.layers.13.self_attn.out_proj.weight": "model-00001-of-00004.safetensors",
97
+ "audio_tower.layers.13.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
98
+ "audio_tower.layers.13.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
99
+ "audio_tower.layers.13.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
100
+ "audio_tower.layers.13.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
101
+ "audio_tower.layers.13.self_attn_layer_norm.bias": "model-00001-of-00004.safetensors",
102
+ "audio_tower.layers.13.self_attn_layer_norm.weight": "model-00001-of-00004.safetensors",
103
+ "audio_tower.layers.14.fc1.bias": "model-00001-of-00004.safetensors",
104
+ "audio_tower.layers.14.fc1.weight": "model-00001-of-00004.safetensors",
105
+ "audio_tower.layers.14.fc2.bias": "model-00001-of-00004.safetensors",
106
+ "audio_tower.layers.14.fc2.weight": "model-00001-of-00004.safetensors",
107
+ "audio_tower.layers.14.final_layer_norm.bias": "model-00001-of-00004.safetensors",
108
+ "audio_tower.layers.14.final_layer_norm.weight": "model-00001-of-00004.safetensors",
109
+ "audio_tower.layers.14.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
110
+ "audio_tower.layers.14.self_attn.out_proj.bias": "model-00001-of-00004.safetensors",
111
+ "audio_tower.layers.14.self_attn.out_proj.weight": "model-00001-of-00004.safetensors",
112
+ "audio_tower.layers.14.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
113
+ "audio_tower.layers.14.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
114
+ "audio_tower.layers.14.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
115
+ "audio_tower.layers.14.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
116
+ "audio_tower.layers.14.self_attn_layer_norm.bias": "model-00001-of-00004.safetensors",
117
+ "audio_tower.layers.14.self_attn_layer_norm.weight": "model-00001-of-00004.safetensors",
118
+ "audio_tower.layers.15.fc1.bias": "model-00001-of-00004.safetensors",
119
+ "audio_tower.layers.15.fc1.weight": "model-00001-of-00004.safetensors",
120
+ "audio_tower.layers.15.fc2.bias": "model-00001-of-00004.safetensors",
121
+ "audio_tower.layers.15.fc2.weight": "model-00001-of-00004.safetensors",
122
+ "audio_tower.layers.15.final_layer_norm.bias": "model-00001-of-00004.safetensors",
123
+ "audio_tower.layers.15.final_layer_norm.weight": "model-00001-of-00004.safetensors",
124
+ "audio_tower.layers.15.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
125
+ "audio_tower.layers.15.self_attn.out_proj.bias": "model-00001-of-00004.safetensors",
126
+ "audio_tower.layers.15.self_attn.out_proj.weight": "model-00001-of-00004.safetensors",
127
+ "audio_tower.layers.15.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
128
+ "audio_tower.layers.15.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
129
+ "audio_tower.layers.15.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
130
+ "audio_tower.layers.15.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
131
+ "audio_tower.layers.15.self_attn_layer_norm.bias": "model-00001-of-00004.safetensors",
132
+ "audio_tower.layers.15.self_attn_layer_norm.weight": "model-00001-of-00004.safetensors",
133
+ "audio_tower.layers.16.fc1.bias": "model-00001-of-00004.safetensors",
134
+ "audio_tower.layers.16.fc1.weight": "model-00001-of-00004.safetensors",
135
+ "audio_tower.layers.16.fc2.bias": "model-00001-of-00004.safetensors",
136
+ "audio_tower.layers.16.fc2.weight": "model-00001-of-00004.safetensors",
137
+ "audio_tower.layers.16.final_layer_norm.bias": "model-00001-of-00004.safetensors",
138
+ "audio_tower.layers.16.final_layer_norm.weight": "model-00001-of-00004.safetensors",
139
+ "audio_tower.layers.16.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
140
+ "audio_tower.layers.16.self_attn.out_proj.bias": "model-00001-of-00004.safetensors",
141
+ "audio_tower.layers.16.self_attn.out_proj.weight": "model-00001-of-00004.safetensors",
142
+ "audio_tower.layers.16.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
143
+ "audio_tower.layers.16.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
144
+ "audio_tower.layers.16.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
145
+ "audio_tower.layers.16.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
146
+ "audio_tower.layers.16.self_attn_layer_norm.bias": "model-00001-of-00004.safetensors",
147
+ "audio_tower.layers.16.self_attn_layer_norm.weight": "model-00001-of-00004.safetensors",
148
+ "audio_tower.layers.17.fc1.bias": "model-00001-of-00004.safetensors",
149
+ "audio_tower.layers.17.fc1.weight": "model-00001-of-00004.safetensors",
150
+ "audio_tower.layers.17.fc2.bias": "model-00001-of-00004.safetensors",
151
+ "audio_tower.layers.17.fc2.weight": "model-00001-of-00004.safetensors",
152
+ "audio_tower.layers.17.final_layer_norm.bias": "model-00001-of-00004.safetensors",
153
+ "audio_tower.layers.17.final_layer_norm.weight": "model-00001-of-00004.safetensors",
154
+ "audio_tower.layers.17.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
155
+ "audio_tower.layers.17.self_attn.out_proj.bias": "model-00001-of-00004.safetensors",
156
+ "audio_tower.layers.17.self_attn.out_proj.weight": "model-00001-of-00004.safetensors",
157
+ "audio_tower.layers.17.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
158
+ "audio_tower.layers.17.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
159
+ "audio_tower.layers.17.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
160
+ "audio_tower.layers.17.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
161
+ "audio_tower.layers.17.self_attn_layer_norm.bias": "model-00001-of-00004.safetensors",
162
+ "audio_tower.layers.17.self_attn_layer_norm.weight": "model-00001-of-00004.safetensors",
163
+ "audio_tower.layers.18.fc1.bias": "model-00001-of-00004.safetensors",
164
+ "audio_tower.layers.18.fc1.weight": "model-00001-of-00004.safetensors",
165
+ "audio_tower.layers.18.fc2.bias": "model-00001-of-00004.safetensors",
166
+ "audio_tower.layers.18.fc2.weight": "model-00001-of-00004.safetensors",
167
+ "audio_tower.layers.18.final_layer_norm.bias": "model-00001-of-00004.safetensors",
168
+ "audio_tower.layers.18.final_layer_norm.weight": "model-00001-of-00004.safetensors",
169
+ "audio_tower.layers.18.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
170
+ "audio_tower.layers.18.self_attn.out_proj.bias": "model-00001-of-00004.safetensors",
171
+ "audio_tower.layers.18.self_attn.out_proj.weight": "model-00001-of-00004.safetensors",
172
+ "audio_tower.layers.18.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
173
+ "audio_tower.layers.18.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
174
+ "audio_tower.layers.18.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
175
+ "audio_tower.layers.18.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
176
+ "audio_tower.layers.18.self_attn_layer_norm.bias": "model-00001-of-00004.safetensors",
177
+ "audio_tower.layers.18.self_attn_layer_norm.weight": "model-00001-of-00004.safetensors",
178
+ "audio_tower.layers.19.fc1.bias": "model-00001-of-00004.safetensors",
179
+ "audio_tower.layers.19.fc1.weight": "model-00001-of-00004.safetensors",
180
+ "audio_tower.layers.19.fc2.bias": "model-00001-of-00004.safetensors",
181
+ "audio_tower.layers.19.fc2.weight": "model-00001-of-00004.safetensors",
182
+ "audio_tower.layers.19.final_layer_norm.bias": "model-00001-of-00004.safetensors",
183
+ "audio_tower.layers.19.final_layer_norm.weight": "model-00001-of-00004.safetensors",
184
+ "audio_tower.layers.19.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
185
+ "audio_tower.layers.19.self_attn.out_proj.bias": "model-00001-of-00004.safetensors",
186
+ "audio_tower.layers.19.self_attn.out_proj.weight": "model-00001-of-00004.safetensors",
187
+ "audio_tower.layers.19.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
188
+ "audio_tower.layers.19.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
189
+ "audio_tower.layers.19.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
190
+ "audio_tower.layers.19.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
191
+ "audio_tower.layers.19.self_attn_layer_norm.bias": "model-00001-of-00004.safetensors",
192
+ "audio_tower.layers.19.self_attn_layer_norm.weight": "model-00001-of-00004.safetensors",
193
+ "audio_tower.layers.2.fc1.bias": "model-00001-of-00004.safetensors",
194
+ "audio_tower.layers.2.fc1.weight": "model-00001-of-00004.safetensors",
195
+ "audio_tower.layers.2.fc2.bias": "model-00001-of-00004.safetensors",
196
+ "audio_tower.layers.2.fc2.weight": "model-00001-of-00004.safetensors",
197
+ "audio_tower.layers.2.final_layer_norm.bias": "model-00001-of-00004.safetensors",
198
+ "audio_tower.layers.2.final_layer_norm.weight": "model-00001-of-00004.safetensors",
199
+ "audio_tower.layers.2.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
200
+ "audio_tower.layers.2.self_attn.out_proj.bias": "model-00001-of-00004.safetensors",
201
+ "audio_tower.layers.2.self_attn.out_proj.weight": "model-00001-of-00004.safetensors",
202
+ "audio_tower.layers.2.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
203
+ "audio_tower.layers.2.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
204
+ "audio_tower.layers.2.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
205
+ "audio_tower.layers.2.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
206
+ "audio_tower.layers.2.self_attn_layer_norm.bias": "model-00001-of-00004.safetensors",
207
+ "audio_tower.layers.2.self_attn_layer_norm.weight": "model-00001-of-00004.safetensors",
208
+ "audio_tower.layers.20.fc1.bias": "model-00001-of-00004.safetensors",
209
+ "audio_tower.layers.20.fc1.weight": "model-00001-of-00004.safetensors",
210
+ "audio_tower.layers.20.fc2.bias": "model-00001-of-00004.safetensors",
211
+ "audio_tower.layers.20.fc2.weight": "model-00001-of-00004.safetensors",
212
+ "audio_tower.layers.20.final_layer_norm.bias": "model-00001-of-00004.safetensors",
213
+ "audio_tower.layers.20.final_layer_norm.weight": "model-00001-of-00004.safetensors",
214
+ "audio_tower.layers.20.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
215
+ "audio_tower.layers.20.self_attn.out_proj.bias": "model-00001-of-00004.safetensors",
216
+ "audio_tower.layers.20.self_attn.out_proj.weight": "model-00001-of-00004.safetensors",
217
+ "audio_tower.layers.20.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
218
+ "audio_tower.layers.20.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
219
+ "audio_tower.layers.20.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
220
+ "audio_tower.layers.20.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
221
+ "audio_tower.layers.20.self_attn_layer_norm.bias": "model-00001-of-00004.safetensors",
222
+ "audio_tower.layers.20.self_attn_layer_norm.weight": "model-00001-of-00004.safetensors",
223
+ "audio_tower.layers.21.fc1.bias": "model-00001-of-00004.safetensors",
224
+ "audio_tower.layers.21.fc1.weight": "model-00001-of-00004.safetensors",
225
+ "audio_tower.layers.21.fc2.bias": "model-00001-of-00004.safetensors",
226
+ "audio_tower.layers.21.fc2.weight": "model-00001-of-00004.safetensors",
227
+ "audio_tower.layers.21.final_layer_norm.bias": "model-00001-of-00004.safetensors",
228
+ "audio_tower.layers.21.final_layer_norm.weight": "model-00001-of-00004.safetensors",
229
+ "audio_tower.layers.21.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
230
+ "audio_tower.layers.21.self_attn.out_proj.bias": "model-00001-of-00004.safetensors",
231
+ "audio_tower.layers.21.self_attn.out_proj.weight": "model-00001-of-00004.safetensors",
232
+ "audio_tower.layers.21.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
233
+ "audio_tower.layers.21.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
234
+ "audio_tower.layers.21.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
235
+ "audio_tower.layers.21.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
236
+ "audio_tower.layers.21.self_attn_layer_norm.bias": "model-00001-of-00004.safetensors",
237
+ "audio_tower.layers.21.self_attn_layer_norm.weight": "model-00001-of-00004.safetensors",
238
+ "audio_tower.layers.22.fc1.bias": "model-00001-of-00004.safetensors",
239
+ "audio_tower.layers.22.fc1.weight": "model-00001-of-00004.safetensors",
240
+ "audio_tower.layers.22.fc2.bias": "model-00001-of-00004.safetensors",
241
+ "audio_tower.layers.22.fc2.weight": "model-00001-of-00004.safetensors",
242
+ "audio_tower.layers.22.final_layer_norm.bias": "model-00001-of-00004.safetensors",
243
+ "audio_tower.layers.22.final_layer_norm.weight": "model-00001-of-00004.safetensors",
244
+ "audio_tower.layers.22.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
245
+ "audio_tower.layers.22.self_attn.out_proj.bias": "model-00001-of-00004.safetensors",
246
+ "audio_tower.layers.22.self_attn.out_proj.weight": "model-00001-of-00004.safetensors",
247
+ "audio_tower.layers.22.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
248
+ "audio_tower.layers.22.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
249
+ "audio_tower.layers.22.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
250
+ "audio_tower.layers.22.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
251
+ "audio_tower.layers.22.self_attn_layer_norm.bias": "model-00001-of-00004.safetensors",
252
+ "audio_tower.layers.22.self_attn_layer_norm.weight": "model-00001-of-00004.safetensors",
253
+ "audio_tower.layers.23.fc1.bias": "model-00001-of-00004.safetensors",
254
+ "audio_tower.layers.23.fc1.weight": "model-00001-of-00004.safetensors",
255
+ "audio_tower.layers.23.fc2.bias": "model-00001-of-00004.safetensors",
256
+ "audio_tower.layers.23.fc2.weight": "model-00001-of-00004.safetensors",
257
+ "audio_tower.layers.23.final_layer_norm.bias": "model-00001-of-00004.safetensors",
258
+ "audio_tower.layers.23.final_layer_norm.weight": "model-00001-of-00004.safetensors",
259
+ "audio_tower.layers.23.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
260
+ "audio_tower.layers.23.self_attn.out_proj.bias": "model-00001-of-00004.safetensors",
261
+ "audio_tower.layers.23.self_attn.out_proj.weight": "model-00001-of-00004.safetensors",
262
+ "audio_tower.layers.23.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
263
+ "audio_tower.layers.23.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
264
+ "audio_tower.layers.23.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
265
+ "audio_tower.layers.23.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
266
+ "audio_tower.layers.23.self_attn_layer_norm.bias": "model-00001-of-00004.safetensors",
267
+ "audio_tower.layers.23.self_attn_layer_norm.weight": "model-00001-of-00004.safetensors",
268
+ "audio_tower.layers.24.fc1.bias": "model-00001-of-00004.safetensors",
269
+ "audio_tower.layers.24.fc1.weight": "model-00001-of-00004.safetensors",
270
+ "audio_tower.layers.24.fc2.bias": "model-00001-of-00004.safetensors",
271
+ "audio_tower.layers.24.fc2.weight": "model-00001-of-00004.safetensors",
272
+ "audio_tower.layers.24.final_layer_norm.bias": "model-00001-of-00004.safetensors",
273
+ "audio_tower.layers.24.final_layer_norm.weight": "model-00001-of-00004.safetensors",
274
+ "audio_tower.layers.24.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
275
+ "audio_tower.layers.24.self_attn.out_proj.bias": "model-00001-of-00004.safetensors",
276
+ "audio_tower.layers.24.self_attn.out_proj.weight": "model-00001-of-00004.safetensors",
277
+ "audio_tower.layers.24.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
278
+ "audio_tower.layers.24.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
279
+ "audio_tower.layers.24.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
280
+ "audio_tower.layers.24.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
281
+ "audio_tower.layers.24.self_attn_layer_norm.bias": "model-00001-of-00004.safetensors",
282
+ "audio_tower.layers.24.self_attn_layer_norm.weight": "model-00001-of-00004.safetensors",
283
+ "audio_tower.layers.25.fc1.bias": "model-00001-of-00004.safetensors",
284
+ "audio_tower.layers.25.fc1.weight": "model-00001-of-00004.safetensors",
285
+ "audio_tower.layers.25.fc2.bias": "model-00001-of-00004.safetensors",
286
+ "audio_tower.layers.25.fc2.weight": "model-00001-of-00004.safetensors",
287
+ "audio_tower.layers.25.final_layer_norm.bias": "model-00001-of-00004.safetensors",
288
+ "audio_tower.layers.25.final_layer_norm.weight": "model-00001-of-00004.safetensors",
289
+ "audio_tower.layers.25.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
290
+ "audio_tower.layers.25.self_attn.out_proj.bias": "model-00001-of-00004.safetensors",
291
+ "audio_tower.layers.25.self_attn.out_proj.weight": "model-00001-of-00004.safetensors",
292
+ "audio_tower.layers.25.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
293
+ "audio_tower.layers.25.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
294
+ "audio_tower.layers.25.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
295
+ "audio_tower.layers.25.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
296
+ "audio_tower.layers.25.self_attn_layer_norm.bias": "model-00001-of-00004.safetensors",
297
+ "audio_tower.layers.25.self_attn_layer_norm.weight": "model-00001-of-00004.safetensors",
298
+ "audio_tower.layers.26.fc1.bias": "model-00001-of-00004.safetensors",
299
+ "audio_tower.layers.26.fc1.weight": "model-00001-of-00004.safetensors",
300
+ "audio_tower.layers.26.fc2.bias": "model-00001-of-00004.safetensors",
301
+ "audio_tower.layers.26.fc2.weight": "model-00001-of-00004.safetensors",
302
+ "audio_tower.layers.26.final_layer_norm.bias": "model-00001-of-00004.safetensors",
303
+ "audio_tower.layers.26.final_layer_norm.weight": "model-00001-of-00004.safetensors",
304
+ "audio_tower.layers.26.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
305
+ "audio_tower.layers.26.self_attn.out_proj.bias": "model-00001-of-00004.safetensors",
306
+ "audio_tower.layers.26.self_attn.out_proj.weight": "model-00001-of-00004.safetensors",
307
+ "audio_tower.layers.26.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
308
+ "audio_tower.layers.26.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
309
+ "audio_tower.layers.26.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
310
+ "audio_tower.layers.26.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
311
+ "audio_tower.layers.26.self_attn_layer_norm.bias": "model-00001-of-00004.safetensors",
312
+ "audio_tower.layers.26.self_attn_layer_norm.weight": "model-00001-of-00004.safetensors",
313
+ "audio_tower.layers.27.fc1.bias": "model-00001-of-00004.safetensors",
314
+ "audio_tower.layers.27.fc1.weight": "model-00001-of-00004.safetensors",
315
+ "audio_tower.layers.27.fc2.bias": "model-00001-of-00004.safetensors",
316
+ "audio_tower.layers.27.fc2.weight": "model-00001-of-00004.safetensors",
317
+ "audio_tower.layers.27.final_layer_norm.bias": "model-00001-of-00004.safetensors",
318
+ "audio_tower.layers.27.final_layer_norm.weight": "model-00001-of-00004.safetensors",
319
+ "audio_tower.layers.27.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
320
+ "audio_tower.layers.27.self_attn.out_proj.bias": "model-00001-of-00004.safetensors",
321
+ "audio_tower.layers.27.self_attn.out_proj.weight": "model-00001-of-00004.safetensors",
322
+ "audio_tower.layers.27.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
323
+ "audio_tower.layers.27.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
324
+ "audio_tower.layers.27.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
325
+ "audio_tower.layers.27.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
326
+ "audio_tower.layers.27.self_attn_layer_norm.bias": "model-00001-of-00004.safetensors",
327
+ "audio_tower.layers.27.self_attn_layer_norm.weight": "model-00001-of-00004.safetensors",
328
+ "audio_tower.layers.28.fc1.bias": "model-00001-of-00004.safetensors",
329
+ "audio_tower.layers.28.fc1.weight": "model-00001-of-00004.safetensors",
330
+ "audio_tower.layers.28.fc2.bias": "model-00001-of-00004.safetensors",
331
+ "audio_tower.layers.28.fc2.weight": "model-00001-of-00004.safetensors",
332
+ "audio_tower.layers.28.final_layer_norm.bias": "model-00001-of-00004.safetensors",
333
+ "audio_tower.layers.28.final_layer_norm.weight": "model-00001-of-00004.safetensors",
334
+ "audio_tower.layers.28.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
335
+ "audio_tower.layers.28.self_attn.out_proj.bias": "model-00001-of-00004.safetensors",
336
+ "audio_tower.layers.28.self_attn.out_proj.weight": "model-00001-of-00004.safetensors",
337
+ "audio_tower.layers.28.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
338
+ "audio_tower.layers.28.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
339
+ "audio_tower.layers.28.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
340
+ "audio_tower.layers.28.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
341
+ "audio_tower.layers.28.self_attn_layer_norm.bias": "model-00001-of-00004.safetensors",
342
+ "audio_tower.layers.28.self_attn_layer_norm.weight": "model-00001-of-00004.safetensors",
343
+ "audio_tower.layers.29.fc1.bias": "model-00001-of-00004.safetensors",
344
+ "audio_tower.layers.29.fc1.weight": "model-00001-of-00004.safetensors",
345
+ "audio_tower.layers.29.fc2.bias": "model-00001-of-00004.safetensors",
346
+ "audio_tower.layers.29.fc2.weight": "model-00001-of-00004.safetensors",
347
+ "audio_tower.layers.29.final_layer_norm.bias": "model-00001-of-00004.safetensors",
348
+ "audio_tower.layers.29.final_layer_norm.weight": "model-00001-of-00004.safetensors",
349
+ "audio_tower.layers.29.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
350
+ "audio_tower.layers.29.self_attn.out_proj.bias": "model-00001-of-00004.safetensors",
351
+ "audio_tower.layers.29.self_attn.out_proj.weight": "model-00001-of-00004.safetensors",
352
+ "audio_tower.layers.29.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
353
+ "audio_tower.layers.29.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
354
+ "audio_tower.layers.29.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
355
+ "audio_tower.layers.29.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
356
+ "audio_tower.layers.29.self_attn_layer_norm.bias": "model-00001-of-00004.safetensors",
357
+ "audio_tower.layers.29.self_attn_layer_norm.weight": "model-00001-of-00004.safetensors",
358
+ "audio_tower.layers.3.fc1.bias": "model-00001-of-00004.safetensors",
359
+ "audio_tower.layers.3.fc1.weight": "model-00001-of-00004.safetensors",
360
+ "audio_tower.layers.3.fc2.bias": "model-00001-of-00004.safetensors",
361
+ "audio_tower.layers.3.fc2.weight": "model-00001-of-00004.safetensors",
362
+ "audio_tower.layers.3.final_layer_norm.bias": "model-00001-of-00004.safetensors",
363
+ "audio_tower.layers.3.final_layer_norm.weight": "model-00001-of-00004.safetensors",
364
+ "audio_tower.layers.3.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
365
+ "audio_tower.layers.3.self_attn.out_proj.bias": "model-00001-of-00004.safetensors",
366
+ "audio_tower.layers.3.self_attn.out_proj.weight": "model-00001-of-00004.safetensors",
367
+ "audio_tower.layers.3.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
368
+ "audio_tower.layers.3.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
369
+ "audio_tower.layers.3.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
370
+ "audio_tower.layers.3.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
371
+ "audio_tower.layers.3.self_attn_layer_norm.bias": "model-00001-of-00004.safetensors",
372
+ "audio_tower.layers.3.self_attn_layer_norm.weight": "model-00001-of-00004.safetensors",
373
+ "audio_tower.layers.30.fc1.bias": "model-00001-of-00004.safetensors",
374
+ "audio_tower.layers.30.fc1.weight": "model-00001-of-00004.safetensors",
375
+ "audio_tower.layers.30.fc2.bias": "model-00001-of-00004.safetensors",
376
+ "audio_tower.layers.30.fc2.weight": "model-00001-of-00004.safetensors",
377
+ "audio_tower.layers.30.final_layer_norm.bias": "model-00001-of-00004.safetensors",
378
+ "audio_tower.layers.30.final_layer_norm.weight": "model-00001-of-00004.safetensors",
379
+ "audio_tower.layers.30.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
380
+ "audio_tower.layers.30.self_attn.out_proj.bias": "model-00001-of-00004.safetensors",
381
+ "audio_tower.layers.30.self_attn.out_proj.weight": "model-00001-of-00004.safetensors",
382
+ "audio_tower.layers.30.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
383
+ "audio_tower.layers.30.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
384
+ "audio_tower.layers.30.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
385
+ "audio_tower.layers.30.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
386
+ "audio_tower.layers.30.self_attn_layer_norm.bias": "model-00001-of-00004.safetensors",
387
+ "audio_tower.layers.30.self_attn_layer_norm.weight": "model-00001-of-00004.safetensors",
388
+ "audio_tower.layers.31.fc1.bias": "model-00001-of-00004.safetensors",
389
+ "audio_tower.layers.31.fc1.weight": "model-00001-of-00004.safetensors",
390
+ "audio_tower.layers.31.fc2.bias": "model-00001-of-00004.safetensors",
391
+ "audio_tower.layers.31.fc2.weight": "model-00001-of-00004.safetensors",
392
+ "audio_tower.layers.31.final_layer_norm.bias": "model-00001-of-00004.safetensors",
393
+ "audio_tower.layers.31.final_layer_norm.weight": "model-00001-of-00004.safetensors",
394
+ "audio_tower.layers.31.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
395
+ "audio_tower.layers.31.self_attn.out_proj.bias": "model-00001-of-00004.safetensors",
396
+ "audio_tower.layers.31.self_attn.out_proj.weight": "model-00001-of-00004.safetensors",
397
+ "audio_tower.layers.31.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
398
+ "audio_tower.layers.31.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
399
+ "audio_tower.layers.31.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
400
+ "audio_tower.layers.31.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
401
+ "audio_tower.layers.31.self_attn_layer_norm.bias": "model-00001-of-00004.safetensors",
402
+ "audio_tower.layers.31.self_attn_layer_norm.weight": "model-00001-of-00004.safetensors",
403
+ "audio_tower.layers.4.fc1.bias": "model-00001-of-00004.safetensors",
404
+ "audio_tower.layers.4.fc1.weight": "model-00001-of-00004.safetensors",
405
+ "audio_tower.layers.4.fc2.bias": "model-00001-of-00004.safetensors",
406
+ "audio_tower.layers.4.fc2.weight": "model-00001-of-00004.safetensors",
407
+ "audio_tower.layers.4.final_layer_norm.bias": "model-00001-of-00004.safetensors",
408
+ "audio_tower.layers.4.final_layer_norm.weight": "model-00001-of-00004.safetensors",
409
+ "audio_tower.layers.4.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
410
+ "audio_tower.layers.4.self_attn.out_proj.bias": "model-00001-of-00004.safetensors",
411
+ "audio_tower.layers.4.self_attn.out_proj.weight": "model-00001-of-00004.safetensors",
412
+ "audio_tower.layers.4.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
413
+ "audio_tower.layers.4.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
414
+ "audio_tower.layers.4.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
415
+ "audio_tower.layers.4.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
416
+ "audio_tower.layers.4.self_attn_layer_norm.bias": "model-00001-of-00004.safetensors",
417
+ "audio_tower.layers.4.self_attn_layer_norm.weight": "model-00001-of-00004.safetensors",
418
+ "audio_tower.layers.5.fc1.bias": "model-00001-of-00004.safetensors",
419
+ "audio_tower.layers.5.fc1.weight": "model-00001-of-00004.safetensors",
420
+ "audio_tower.layers.5.fc2.bias": "model-00001-of-00004.safetensors",
421
+ "audio_tower.layers.5.fc2.weight": "model-00001-of-00004.safetensors",
422
+ "audio_tower.layers.5.final_layer_norm.bias": "model-00001-of-00004.safetensors",
423
+ "audio_tower.layers.5.final_layer_norm.weight": "model-00001-of-00004.safetensors",
424
+ "audio_tower.layers.5.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
425
+ "audio_tower.layers.5.self_attn.out_proj.bias": "model-00001-of-00004.safetensors",
426
+ "audio_tower.layers.5.self_attn.out_proj.weight": "model-00001-of-00004.safetensors",
427
+ "audio_tower.layers.5.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
428
+ "audio_tower.layers.5.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
429
+ "audio_tower.layers.5.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
430
+ "audio_tower.layers.5.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
431
+ "audio_tower.layers.5.self_attn_layer_norm.bias": "model-00001-of-00004.safetensors",
432
+ "audio_tower.layers.5.self_attn_layer_norm.weight": "model-00001-of-00004.safetensors",
433
+ "audio_tower.layers.6.fc1.bias": "model-00001-of-00004.safetensors",
434
+ "audio_tower.layers.6.fc1.weight": "model-00001-of-00004.safetensors",
435
+ "audio_tower.layers.6.fc2.bias": "model-00001-of-00004.safetensors",
436
+ "audio_tower.layers.6.fc2.weight": "model-00001-of-00004.safetensors",
437
+ "audio_tower.layers.6.final_layer_norm.bias": "model-00001-of-00004.safetensors",
438
+ "audio_tower.layers.6.final_layer_norm.weight": "model-00001-of-00004.safetensors",
439
+ "audio_tower.layers.6.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
440
+ "audio_tower.layers.6.self_attn.out_proj.bias": "model-00001-of-00004.safetensors",
441
+ "audio_tower.layers.6.self_attn.out_proj.weight": "model-00001-of-00004.safetensors",
442
+ "audio_tower.layers.6.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
443
+ "audio_tower.layers.6.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
444
+ "audio_tower.layers.6.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
445
+ "audio_tower.layers.6.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
446
+ "audio_tower.layers.6.self_attn_layer_norm.bias": "model-00001-of-00004.safetensors",
447
+ "audio_tower.layers.6.self_attn_layer_norm.weight": "model-00001-of-00004.safetensors",
448
+ "audio_tower.layers.7.fc1.bias": "model-00001-of-00004.safetensors",
449
+ "audio_tower.layers.7.fc1.weight": "model-00001-of-00004.safetensors",
450
+ "audio_tower.layers.7.fc2.bias": "model-00001-of-00004.safetensors",
451
+ "audio_tower.layers.7.fc2.weight": "model-00001-of-00004.safetensors",
452
+ "audio_tower.layers.7.final_layer_norm.bias": "model-00001-of-00004.safetensors",
453
+ "audio_tower.layers.7.final_layer_norm.weight": "model-00001-of-00004.safetensors",
454
+ "audio_tower.layers.7.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
455
+ "audio_tower.layers.7.self_attn.out_proj.bias": "model-00001-of-00004.safetensors",
456
+ "audio_tower.layers.7.self_attn.out_proj.weight": "model-00001-of-00004.safetensors",
457
+ "audio_tower.layers.7.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
458
+ "audio_tower.layers.7.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
459
+ "audio_tower.layers.7.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
460
+ "audio_tower.layers.7.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
461
+ "audio_tower.layers.7.self_attn_layer_norm.bias": "model-00001-of-00004.safetensors",
462
+ "audio_tower.layers.7.self_attn_layer_norm.weight": "model-00001-of-00004.safetensors",
463
+ "audio_tower.layers.8.fc1.bias": "model-00001-of-00004.safetensors",
464
+ "audio_tower.layers.8.fc1.weight": "model-00001-of-00004.safetensors",
465
+ "audio_tower.layers.8.fc2.bias": "model-00001-of-00004.safetensors",
466
+ "audio_tower.layers.8.fc2.weight": "model-00001-of-00004.safetensors",
467
+ "audio_tower.layers.8.final_layer_norm.bias": "model-00001-of-00004.safetensors",
468
+ "audio_tower.layers.8.final_layer_norm.weight": "model-00001-of-00004.safetensors",
469
+ "audio_tower.layers.8.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
470
+ "audio_tower.layers.8.self_attn.out_proj.bias": "model-00001-of-00004.safetensors",
471
+ "audio_tower.layers.8.self_attn.out_proj.weight": "model-00001-of-00004.safetensors",
472
+ "audio_tower.layers.8.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
473
+ "audio_tower.layers.8.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
474
+ "audio_tower.layers.8.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
475
+ "audio_tower.layers.8.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
476
+ "audio_tower.layers.8.self_attn_layer_norm.bias": "model-00001-of-00004.safetensors",
477
+ "audio_tower.layers.8.self_attn_layer_norm.weight": "model-00001-of-00004.safetensors",
478
+ "audio_tower.layers.9.fc1.bias": "model-00001-of-00004.safetensors",
479
+ "audio_tower.layers.9.fc1.weight": "model-00001-of-00004.safetensors",
480
+ "audio_tower.layers.9.fc2.bias": "model-00001-of-00004.safetensors",
481
+ "audio_tower.layers.9.fc2.weight": "model-00001-of-00004.safetensors",
482
+ "audio_tower.layers.9.final_layer_norm.bias": "model-00001-of-00004.safetensors",
483
+ "audio_tower.layers.9.final_layer_norm.weight": "model-00001-of-00004.safetensors",
484
+ "audio_tower.layers.9.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
485
+ "audio_tower.layers.9.self_attn.out_proj.bias": "model-00001-of-00004.safetensors",
486
+ "audio_tower.layers.9.self_attn.out_proj.weight": "model-00001-of-00004.safetensors",
487
+ "audio_tower.layers.9.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
488
+ "audio_tower.layers.9.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
489
+ "audio_tower.layers.9.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
490
+ "audio_tower.layers.9.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
491
+ "audio_tower.layers.9.self_attn_layer_norm.bias": "model-00001-of-00004.safetensors",
492
+ "audio_tower.layers.9.self_attn_layer_norm.weight": "model-00001-of-00004.safetensors",
493
+ "language_model.lm_head.weight": "model-00004-of-00004.safetensors",
494
+ "language_model.model.embed_tokens.weight": "model-00001-of-00004.safetensors",
495
+ "language_model.model.layers.0.input_layernorm.weight": "model-00001-of-00004.safetensors",
496
+ "language_model.model.layers.0.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
497
+ "language_model.model.layers.0.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
498
+ "language_model.model.layers.0.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
499
+ "language_model.model.layers.0.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
500
+ "language_model.model.layers.0.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
501
+ "language_model.model.layers.0.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
502
+ "language_model.model.layers.0.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
503
+ "language_model.model.layers.0.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
504
+ "language_model.model.layers.1.input_layernorm.weight": "model-00001-of-00004.safetensors",
505
+ "language_model.model.layers.1.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
506
+ "language_model.model.layers.1.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
507
+ "language_model.model.layers.1.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
508
+ "language_model.model.layers.1.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
509
+ "language_model.model.layers.1.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
510
+ "language_model.model.layers.1.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
511
+ "language_model.model.layers.1.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
512
+ "language_model.model.layers.1.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
513
+ "language_model.model.layers.10.input_layernorm.weight": "model-00002-of-00004.safetensors",
514
+ "language_model.model.layers.10.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
515
+ "language_model.model.layers.10.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
516
+ "language_model.model.layers.10.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
517
+ "language_model.model.layers.10.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
518
+ "language_model.model.layers.10.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
519
+ "language_model.model.layers.10.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
520
+ "language_model.model.layers.10.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
521
+ "language_model.model.layers.10.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
522
+ "language_model.model.layers.11.input_layernorm.weight": "model-00002-of-00004.safetensors",
523
+ "language_model.model.layers.11.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
524
+ "language_model.model.layers.11.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
525
+ "language_model.model.layers.11.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
526
+ "language_model.model.layers.11.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
527
+ "language_model.model.layers.11.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
528
+ "language_model.model.layers.11.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
529
+ "language_model.model.layers.11.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
530
+ "language_model.model.layers.11.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
531
+ "language_model.model.layers.12.input_layernorm.weight": "model-00002-of-00004.safetensors",
532
+ "language_model.model.layers.12.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
533
+ "language_model.model.layers.12.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
534
+ "language_model.model.layers.12.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
535
+ "language_model.model.layers.12.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
536
+ "language_model.model.layers.12.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
537
+ "language_model.model.layers.12.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
538
+ "language_model.model.layers.12.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
539
+ "language_model.model.layers.12.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
540
+ "language_model.model.layers.13.input_layernorm.weight": "model-00002-of-00004.safetensors",
541
+ "language_model.model.layers.13.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
542
+ "language_model.model.layers.13.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
543
+ "language_model.model.layers.13.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
544
+ "language_model.model.layers.13.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
545
+ "language_model.model.layers.13.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
546
+ "language_model.model.layers.13.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
547
+ "language_model.model.layers.13.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
548
+ "language_model.model.layers.13.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
549
+ "language_model.model.layers.14.input_layernorm.weight": "model-00002-of-00004.safetensors",
550
+ "language_model.model.layers.14.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
551
+ "language_model.model.layers.14.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
552
+ "language_model.model.layers.14.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
553
+ "language_model.model.layers.14.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
554
+ "language_model.model.layers.14.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
555
+ "language_model.model.layers.14.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
556
+ "language_model.model.layers.14.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
557
+ "language_model.model.layers.14.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
558
+ "language_model.model.layers.15.input_layernorm.weight": "model-00002-of-00004.safetensors",
559
+ "language_model.model.layers.15.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
560
+ "language_model.model.layers.15.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
561
+ "language_model.model.layers.15.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
562
+ "language_model.model.layers.15.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
563
+ "language_model.model.layers.15.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
564
+ "language_model.model.layers.15.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
565
+ "language_model.model.layers.15.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
566
+ "language_model.model.layers.15.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
567
+ "language_model.model.layers.16.input_layernorm.weight": "model-00002-of-00004.safetensors",
568
+ "language_model.model.layers.16.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
569
+ "language_model.model.layers.16.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
570
+ "language_model.model.layers.16.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
571
+ "language_model.model.layers.16.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
572
+ "language_model.model.layers.16.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
573
+ "language_model.model.layers.16.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
574
+ "language_model.model.layers.16.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
575
+ "language_model.model.layers.16.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
576
+ "language_model.model.layers.17.input_layernorm.weight": "model-00003-of-00004.safetensors",
577
+ "language_model.model.layers.17.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
578
+ "language_model.model.layers.17.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
579
+ "language_model.model.layers.17.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
580
+ "language_model.model.layers.17.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
581
+ "language_model.model.layers.17.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
582
+ "language_model.model.layers.17.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
583
+ "language_model.model.layers.17.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
584
+ "language_model.model.layers.17.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
585
+ "language_model.model.layers.18.input_layernorm.weight": "model-00003-of-00004.safetensors",
586
+ "language_model.model.layers.18.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
587
+ "language_model.model.layers.18.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
588
+ "language_model.model.layers.18.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
589
+ "language_model.model.layers.18.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
590
+ "language_model.model.layers.18.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
591
+ "language_model.model.layers.18.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
592
+ "language_model.model.layers.18.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
593
+ "language_model.model.layers.18.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
594
+ "language_model.model.layers.19.input_layernorm.weight": "model-00003-of-00004.safetensors",
595
+ "language_model.model.layers.19.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
596
+ "language_model.model.layers.19.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
597
+ "language_model.model.layers.19.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
598
+ "language_model.model.layers.19.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
599
+ "language_model.model.layers.19.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
600
+ "language_model.model.layers.19.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
601
+ "language_model.model.layers.19.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
602
+ "language_model.model.layers.19.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
603
+ "language_model.model.layers.2.input_layernorm.weight": "model-00001-of-00004.safetensors",
604
+ "language_model.model.layers.2.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
605
+ "language_model.model.layers.2.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
606
+ "language_model.model.layers.2.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
607
+ "language_model.model.layers.2.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
608
+ "language_model.model.layers.2.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
609
+ "language_model.model.layers.2.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
610
+ "language_model.model.layers.2.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
611
+ "language_model.model.layers.2.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
612
+ "language_model.model.layers.20.input_layernorm.weight": "model-00003-of-00004.safetensors",
613
+ "language_model.model.layers.20.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
614
+ "language_model.model.layers.20.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
615
+ "language_model.model.layers.20.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
616
+ "language_model.model.layers.20.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
617
+ "language_model.model.layers.20.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
618
+ "language_model.model.layers.20.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
619
+ "language_model.model.layers.20.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
620
+ "language_model.model.layers.20.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
621
+ "language_model.model.layers.21.input_layernorm.weight": "model-00003-of-00004.safetensors",
622
+ "language_model.model.layers.21.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
623
+ "language_model.model.layers.21.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
624
+ "language_model.model.layers.21.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
625
+ "language_model.model.layers.21.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
626
+ "language_model.model.layers.21.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
627
+ "language_model.model.layers.21.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
628
+ "language_model.model.layers.21.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
629
+ "language_model.model.layers.21.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
630
+ "language_model.model.layers.22.input_layernorm.weight": "model-00003-of-00004.safetensors",
631
+ "language_model.model.layers.22.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
632
+ "language_model.model.layers.22.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
633
+ "language_model.model.layers.22.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
634
+ "language_model.model.layers.22.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
635
+ "language_model.model.layers.22.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
636
+ "language_model.model.layers.22.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
637
+ "language_model.model.layers.22.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
638
+ "language_model.model.layers.22.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
639
+ "language_model.model.layers.23.input_layernorm.weight": "model-00003-of-00004.safetensors",
640
+ "language_model.model.layers.23.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
641
+ "language_model.model.layers.23.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
642
+ "language_model.model.layers.23.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
643
+ "language_model.model.layers.23.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
644
+ "language_model.model.layers.23.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
645
+ "language_model.model.layers.23.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
646
+ "language_model.model.layers.23.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
647
+ "language_model.model.layers.23.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
648
+ "language_model.model.layers.24.input_layernorm.weight": "model-00003-of-00004.safetensors",
649
+ "language_model.model.layers.24.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
650
+ "language_model.model.layers.24.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
651
+ "language_model.model.layers.24.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
652
+ "language_model.model.layers.24.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
653
+ "language_model.model.layers.24.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
654
+ "language_model.model.layers.24.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
655
+ "language_model.model.layers.24.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
656
+ "language_model.model.layers.24.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
657
+ "language_model.model.layers.25.input_layernorm.weight": "model-00003-of-00004.safetensors",
658
+ "language_model.model.layers.25.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
659
+ "language_model.model.layers.25.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
660
+ "language_model.model.layers.25.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
661
+ "language_model.model.layers.25.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
662
+ "language_model.model.layers.25.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
663
+ "language_model.model.layers.25.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
664
+ "language_model.model.layers.25.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
665
+ "language_model.model.layers.25.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
666
+ "language_model.model.layers.26.input_layernorm.weight": "model-00003-of-00004.safetensors",
667
+ "language_model.model.layers.26.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
668
+ "language_model.model.layers.26.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
669
+ "language_model.model.layers.26.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
670
+ "language_model.model.layers.26.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
671
+ "language_model.model.layers.26.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
672
+ "language_model.model.layers.26.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
673
+ "language_model.model.layers.26.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
674
+ "language_model.model.layers.26.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
675
+ "language_model.model.layers.27.input_layernorm.weight": "model-00003-of-00004.safetensors",
676
+ "language_model.model.layers.27.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
677
+ "language_model.model.layers.27.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
678
+ "language_model.model.layers.27.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
679
+ "language_model.model.layers.27.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
680
+ "language_model.model.layers.27.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
681
+ "language_model.model.layers.27.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
682
+ "language_model.model.layers.27.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
683
+ "language_model.model.layers.27.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
684
+ "language_model.model.layers.28.input_layernorm.weight": "model-00004-of-00004.safetensors",
685
+ "language_model.model.layers.28.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
686
+ "language_model.model.layers.28.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
687
+ "language_model.model.layers.28.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
688
+ "language_model.model.layers.28.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
689
+ "language_model.model.layers.28.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
690
+ "language_model.model.layers.28.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
691
+ "language_model.model.layers.28.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
692
+ "language_model.model.layers.28.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
693
+ "language_model.model.layers.29.input_layernorm.weight": "model-00004-of-00004.safetensors",
694
+ "language_model.model.layers.29.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
695
+ "language_model.model.layers.29.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
696
+ "language_model.model.layers.29.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
697
+ "language_model.model.layers.29.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
698
+ "language_model.model.layers.29.self_attn.k_proj.weight": "model-00004-of-00004.safetensors",
699
+ "language_model.model.layers.29.self_attn.o_proj.weight": "model-00004-of-00004.safetensors",
700
+ "language_model.model.layers.29.self_attn.q_proj.weight": "model-00004-of-00004.safetensors",
701
+ "language_model.model.layers.29.self_attn.v_proj.weight": "model-00004-of-00004.safetensors",
702
+ "language_model.model.layers.3.input_layernorm.weight": "model-00001-of-00004.safetensors",
703
+ "language_model.model.layers.3.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
704
+ "language_model.model.layers.3.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
705
+ "language_model.model.layers.3.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
706
+ "language_model.model.layers.3.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
707
+ "language_model.model.layers.3.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
708
+ "language_model.model.layers.3.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
709
+ "language_model.model.layers.3.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
710
+ "language_model.model.layers.3.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
711
+ "language_model.model.layers.30.input_layernorm.weight": "model-00004-of-00004.safetensors",
712
+ "language_model.model.layers.30.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
713
+ "language_model.model.layers.30.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
714
+ "language_model.model.layers.30.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
715
+ "language_model.model.layers.30.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
716
+ "language_model.model.layers.30.self_attn.k_proj.weight": "model-00004-of-00004.safetensors",
717
+ "language_model.model.layers.30.self_attn.o_proj.weight": "model-00004-of-00004.safetensors",
718
+ "language_model.model.layers.30.self_attn.q_proj.weight": "model-00004-of-00004.safetensors",
719
+ "language_model.model.layers.30.self_attn.v_proj.weight": "model-00004-of-00004.safetensors",
720
+ "language_model.model.layers.31.input_layernorm.weight": "model-00004-of-00004.safetensors",
721
+ "language_model.model.layers.31.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
722
+ "language_model.model.layers.31.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
723
+ "language_model.model.layers.31.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
724
+ "language_model.model.layers.31.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
725
+ "language_model.model.layers.31.self_attn.k_proj.weight": "model-00004-of-00004.safetensors",
726
+ "language_model.model.layers.31.self_attn.o_proj.weight": "model-00004-of-00004.safetensors",
727
+ "language_model.model.layers.31.self_attn.q_proj.weight": "model-00004-of-00004.safetensors",
728
+ "language_model.model.layers.31.self_attn.v_proj.weight": "model-00004-of-00004.safetensors",
729
+ "language_model.model.layers.4.input_layernorm.weight": "model-00001-of-00004.safetensors",
730
+ "language_model.model.layers.4.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
731
+ "language_model.model.layers.4.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
732
+ "language_model.model.layers.4.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
733
+ "language_model.model.layers.4.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
734
+ "language_model.model.layers.4.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
735
+ "language_model.model.layers.4.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
736
+ "language_model.model.layers.4.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
737
+ "language_model.model.layers.4.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
738
+ "language_model.model.layers.5.input_layernorm.weight": "model-00002-of-00004.safetensors",
739
+ "language_model.model.layers.5.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
740
+ "language_model.model.layers.5.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
741
+ "language_model.model.layers.5.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
742
+ "language_model.model.layers.5.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
743
+ "language_model.model.layers.5.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
744
+ "language_model.model.layers.5.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
745
+ "language_model.model.layers.5.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
746
+ "language_model.model.layers.5.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
747
+ "language_model.model.layers.6.input_layernorm.weight": "model-00002-of-00004.safetensors",
748
+ "language_model.model.layers.6.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
749
+ "language_model.model.layers.6.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
750
+ "language_model.model.layers.6.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
751
+ "language_model.model.layers.6.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
752
+ "language_model.model.layers.6.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
753
+ "language_model.model.layers.6.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
754
+ "language_model.model.layers.6.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
755
+ "language_model.model.layers.6.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
756
+ "language_model.model.layers.7.input_layernorm.weight": "model-00002-of-00004.safetensors",
757
+ "language_model.model.layers.7.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
758
+ "language_model.model.layers.7.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
759
+ "language_model.model.layers.7.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
760
+ "language_model.model.layers.7.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
761
+ "language_model.model.layers.7.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
762
+ "language_model.model.layers.7.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
763
+ "language_model.model.layers.7.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
764
+ "language_model.model.layers.7.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
765
+ "language_model.model.layers.8.input_layernorm.weight": "model-00002-of-00004.safetensors",
766
+ "language_model.model.layers.8.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
767
+ "language_model.model.layers.8.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
768
+ "language_model.model.layers.8.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
769
+ "language_model.model.layers.8.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
770
+ "language_model.model.layers.8.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
771
+ "language_model.model.layers.8.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
772
+ "language_model.model.layers.8.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
773
+ "language_model.model.layers.8.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
774
+ "language_model.model.layers.9.input_layernorm.weight": "model-00002-of-00004.safetensors",
775
+ "language_model.model.layers.9.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
776
+ "language_model.model.layers.9.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
777
+ "language_model.model.layers.9.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
778
+ "language_model.model.layers.9.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
779
+ "language_model.model.layers.9.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
780
+ "language_model.model.layers.9.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
781
+ "language_model.model.layers.9.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
782
+ "language_model.model.layers.9.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
783
+ "language_model.model.norm.weight": "model-00004-of-00004.safetensors",
784
+ "multi_modal_projector.linear_1.weight": "model-00001-of-00004.safetensors",
785
+ "multi_modal_projector.linear_2.weight": "model-00001-of-00004.safetensors",
786
+ "multi_modal_projector.ln_post.weight": "model-00001-of-00004.safetensors",
787
+ "multi_modal_projector.ln_pre.weight": "model-00001-of-00004.safetensors"
788
+ }
789
+ }
shuka_config.py ADDED
@@ -0,0 +1,149 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ####################################################################
2
+ ## Built on top of Ultravox: https://github.com/fixie-ai/ultravox ##
3
+ ####################################################################
4
+
5
+ import dataclasses
6
+ from typing import Any, Dict, List, Optional
7
+
8
+ import transformers
9
+
10
+
11
+ @dataclasses.dataclass
12
+ class LoraConfigSimplified:
13
+ """
14
+ Low Rank Approximation (LoRA) configuration.
15
+
16
+ Used for language and audio models separately.
17
+ """
18
+
19
+ # The rank of the approximation
20
+ r: int = 0
21
+ lora_alpha: float = 8
22
+ target_modules: Optional[List[str]] = dataclasses.field(
23
+ default_factory=lambda: ["k_proj", "q_proj", "linear_k", "linear_q"]
24
+ )
25
+
26
+
27
+ class ShukaConfig(transformers.PretrainedConfig):
28
+ r"""
29
+ This is the configuration class to store the configuration of a [`ShukaForConditionalGeneration`]. It is used to instantiate an
30
+ Shuka model according to the specified arguments, defining the model architecture.
31
+
32
+ Configuration objects inherit from [`PretrainedConfig`] and can be used to control the model outputs. Read the
33
+ documentation from [`PretrainedConfig`] for more information.
34
+
35
+ Args:
36
+ audio_config (`Wav2Vec2Config`, *optional*):
37
+ Custom audio config or dict
38
+ text_config (`Union[AutoConfig, dict]`, *optional*):
39
+ The config object of the text backbone. Can be any of `LlamaConfig` or `MistralConfig`.
40
+ ignore_index (`int`, *optional*, defaults to -100):
41
+ The ignore index for the loss function.
42
+ audio_token_index (`int`, *optional*, defaults to 32000):
43
+ The audio token index to encode the audio prompt.
44
+ encoder_ds_factor (`int`, *optional*, defaults to 320):
45
+ The audio encoder downsample factor.
46
+ stack_factor (`int`, *optional*, defaults to 8):
47
+ Audio downsampling factor for the multimodal projector.
48
+ norm_init (`float`, *optional*, defaults to 0.4):
49
+ The initialization value for the layer normalization.
50
+ projector_act (`str`, *optional*, defaults to `"silu"`):
51
+ The activation function used by the multimodal projector.
52
+ text_model_lora_config (`LoraConfigSimplified`, *optional*):
53
+ The LoRA configuration for finetuning the text model.
54
+ audio_model_lora_config (`LoraConfigSimplified`, *optional*):
55
+ The LoRA configuration for finetuning the audio model.
56
+
57
+
58
+ Example:
59
+
60
+ ```python
61
+ >>> from transformers import ShukaForConditionalGeneration, Wav2Vec2Config, ShukaConfig, LlamaConfig
62
+
63
+ >>> # Initializing an audio encoder config
64
+ >>> audio_config = Wav2Vec2Config()
65
+
66
+ >>> # Initializing a Llama config
67
+ >>> text_config = LlamaConfig()
68
+
69
+ >>> # Initializing a default configuration
70
+ >>> configuration = ShukaConfig(audio_config, text_config)
71
+
72
+ >>> # Initializing a completely untrained model from the configuration
73
+ >>> model = ShukaForConditionalGeneration(configuration)
74
+
75
+ >>> # Accessing the model configuration
76
+ >>> configuration = model.config
77
+
78
+ >>> # Initialize a model from pretrained checkpoints and random projector weights
79
+ >>> config = ShukaConfig(audio_model_id="facebook/wav2vec2-base-960h", text_model_id="meta-llama/Llama-2-7b-chat-hf")
80
+ ```"""
81
+
82
+ model_type = "shuka"
83
+ is_composition = False
84
+
85
+ def __init__(
86
+ self,
87
+ audio_config: Optional[Dict[str, Any]] = None,
88
+ text_config: Optional[Dict[str, Any]] = None,
89
+ audio_model_id: Optional[str] = None,
90
+ text_model_id: Optional[str] = None,
91
+ ignore_index: int = -100,
92
+ audio_token_index: int = 32000,
93
+ hidden_size: int = 4096,
94
+ encoder_ds_factor: int = 320,
95
+ stack_factor: int = 8,
96
+ norm_init: float = 0.4,
97
+ projector_act: str = "silu",
98
+ text_model_lora_config: Optional[LoraConfigSimplified] = None,
99
+ audio_model_lora_config: Optional[LoraConfigSimplified] = None,
100
+ **kwargs,
101
+ ):
102
+ self.ignore_index = ignore_index
103
+
104
+ self.audio_model_id = audio_model_id
105
+ self.text_model_id = text_model_id
106
+ self.audio_token_index = audio_token_index
107
+
108
+ self.hidden_size = hidden_size
109
+ self.encoder_ds_factor = encoder_ds_factor
110
+ self.stack_factor = stack_factor
111
+ self.norm_init = norm_init
112
+ self.projector_act = projector_act
113
+
114
+ if text_model_id is not None:
115
+ self.text_config: transformers.LlamaConfig = (
116
+ transformers.AutoConfig.from_pretrained(text_model_id)
117
+ )
118
+ else:
119
+ text_config = text_config or {}
120
+ self.text_config = transformers.CONFIG_MAPPING[
121
+ text_config.get("model_type", "llama")
122
+ ](**text_config)
123
+
124
+ if audio_model_id is not None:
125
+ self.audio_config: transformers.PretrainedConfig = (
126
+ transformers.AutoConfig.from_pretrained(audio_model_id)
127
+ )
128
+ else:
129
+ audio_config = audio_config or {}
130
+ self.audio_config = transformers.CONFIG_MAPPING[
131
+ audio_config.get("model_type", "wav2vec2")
132
+ ](**audio_config)
133
+
134
+ self.text_model_lora_config = (
135
+ text_model_lora_config
136
+ if isinstance(text_model_lora_config, dict)
137
+ else dataclasses.asdict(text_model_lora_config or LoraConfigSimplified())
138
+ )
139
+ self.audio_model_lora_config = (
140
+ audio_model_lora_config
141
+ if isinstance(audio_model_lora_config, dict)
142
+ else dataclasses.asdict(audio_model_lora_config or LoraConfigSimplified())
143
+ )
144
+
145
+ self.vocab_size = self.text_config.vocab_size
146
+
147
+ self.initializer_range = self.text_config.initializer_range
148
+
149
+ super().__init__(**kwargs)
shuka_model.py ADDED
@@ -0,0 +1,413 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ####################################################################
2
+ ## Built on top of Ultravox: https://github.com/fixie-ai/ultravox ##
3
+ ####################################################################
4
+
5
+ import logging
6
+ from typing import Any, Dict, Optional, Set, Tuple, Union
7
+
8
+ import peft
9
+ import torch
10
+ import torch.nn as nn
11
+ import torch.nn.functional as F
12
+ import transformers
13
+ import transformers.activations
14
+ import transformers.modeling_outputs
15
+ import transformers.models
16
+
17
+ from .shuka_config import ShukaConfig
18
+ from .whisper_model_modified import WhisperEncoder as ModifiedWhisperEncoder
19
+
20
+
21
+ class ShukaModel(
22
+ transformers.LlamaPreTrainedModel,
23
+ transformers.GenerationMixin,
24
+ ):
25
+ """
26
+ The Shuka model which consists of an audio encoder and a language model.
27
+
28
+ Audio input is processed by the audio encoder, then every `stack_factor` frames are stacked together and
29
+ projected to the language model's embedding space using a few linear layers.
30
+ The text is embedded by the language model as usual and then the audio and text embeddings are merged together.
31
+
32
+ A special token `<|audio|>` is used to indicate the start of the audio embeddings in the merged embeddings.
33
+
34
+ Parameters:
35
+ config: Model configuration class with all the parameters of the model.
36
+ """
37
+
38
+ config_class = ShukaConfig
39
+ config: ShukaConfig # for type hinting
40
+ _no_split_modules = ["Wav2Vec2Model", "WhisperEncoder", "LlamaDecoderLayer"]
41
+
42
+ def __init__(self, config: ShukaConfig):
43
+ super().__init__(config)
44
+
45
+ self.keep_params: Set[str] = set()
46
+ self.vocab_size = config.vocab_size
47
+
48
+ self.audio_tower = self._create_audio_tower(config)
49
+ self.multi_modal_projector = ShukaProjector(config)
50
+ self.language_model = self._create_language_model(config)
51
+
52
+ self.post_init()
53
+
54
+ def get_input_embeddings(self):
55
+ return self.language_model.get_input_embeddings()
56
+
57
+ def set_input_embeddings(self, value):
58
+ self.language_model.set_input_embeddings(value)
59
+
60
+ def get_output_embeddings(self):
61
+ return self.language_model.get_output_embeddings()
62
+
63
+ def set_output_embeddings(self, new_embeddings):
64
+ self.language_model.set_output_embeddings(new_embeddings)
65
+
66
+ def set_decoder(self, decoder):
67
+ self.language_model.set_decoder(decoder)
68
+
69
+ def get_decoder(self):
70
+ return self.language_model.get_decoder()
71
+
72
+ def tie_weights(self):
73
+ return self.language_model.tie_weights()
74
+
75
+ def _setup_cache(
76
+ self, cache_cls, max_batch_size: int, max_cache_len: Optional[int] = None
77
+ ):
78
+ self.language_model._setup_cache(cache_cls, max_batch_size, max_cache_len)
79
+
80
+ def _reorder_cache(self, past_key_values, beam_idx):
81
+ return self.language_model._reorder_cache(past_key_values, beam_idx)
82
+
83
+ def resize_token_embeddings(
84
+ self,
85
+ new_num_tokens: Optional[int] = None,
86
+ pad_to_multiple_of: Optional[int] = None,
87
+ ) -> nn.Embedding:
88
+ model_embeds = self.language_model.resize_token_embeddings(
89
+ new_num_tokens, pad_to_multiple_of
90
+ )
91
+ # update vocab size
92
+ self.config.text_config.vocab_size = model_embeds.num_embeddings
93
+ self.config.vocab_size = model_embeds.num_embeddings
94
+ self.vocab_size = model_embeds.num_embeddings
95
+ return model_embeds
96
+
97
+ def forward(
98
+ self,
99
+ input_ids: torch.Tensor,
100
+ audio_values: Optional[torch.Tensor] = None,
101
+ inputs_embeds: Optional[torch.Tensor] = None,
102
+ labels: Optional[torch.Tensor] = None,
103
+ attention_mask: Optional[torch.Tensor] = None,
104
+ audio_token_start_idx: Optional[torch.Tensor] = None,
105
+ audio_token_len: Optional[torch.Tensor] = None,
106
+ past_key_values: Optional[Tuple] = None,
107
+ **kwargs,
108
+ ) -> Union[Tuple, transformers.modeling_outputs.CausalLMOutputWithPast]:
109
+ """
110
+ Forward pass for the Shuka model.
111
+
112
+ `input_ids` are the tokenized text input. They are embedded by the language model as usual.
113
+ `audio_values` are processed by the audio encoder and then every `stack_factor` frames are stacked together and
114
+ projected to the language model's embedding space using a few linear layers.
115
+ The audio and text embeddings are merged together. A special token `<|audio|>` is used to indicate the start
116
+ of the audio embeddings in the merged embeddings.
117
+
118
+ Args:
119
+ input_ids: The tokenized text input.
120
+ audio_values: The processed audio values.
121
+ inputs_embeds: The embeddings for the input tokens.
122
+ labels: The tokenized text labels.
123
+ attention_mask: The attention mask for the input.
124
+ position_ids: The position ids for the input.
125
+ past_key_values: The past key value cache for the language model attention layers.
126
+ **kwargs: Additional keyword arguments. Passed directly to the language model.
127
+ """
128
+ if inputs_embeds is None:
129
+ # B x T -> B x T x D
130
+ inputs_embeds = self.get_input_embeddings().forward(input_ids)
131
+
132
+ if audio_values is not None:
133
+ assert (
134
+ audio_token_start_idx is not None and audio_token_len is not None
135
+ ), "audio_token_start_idx and audio_token_len must be provided if audio_values are provided."
136
+ assert (
137
+ len(audio_token_start_idx) == len(audio_token_len) == len(audio_values)
138
+ ), "audio_token_start_idx, audio_token_len, and audio_values must have the same batch size."
139
+
140
+ audio_values = audio_values.to(inputs_embeds.dtype)
141
+ # B x A/3200 x D
142
+ audio_tower_output = self.audio_tower.forward(
143
+ audio_values
144
+ ).last_hidden_state
145
+
146
+ audio_embeds = self.multi_modal_projector.forward(audio_tower_output)
147
+
148
+ # combine audio and text embeddings
149
+ for i, (audio, start, length) in enumerate(
150
+ zip(audio_embeds, audio_token_start_idx, audio_token_len)
151
+ ):
152
+ length = min(length, audio.shape[0])
153
+ inputs_embeds[i, start : start + length] = audio[:length]
154
+
155
+ lm_output = self.language_model.forward(
156
+ inputs_embeds=inputs_embeds,
157
+ labels=labels,
158
+ attention_mask=attention_mask,
159
+ past_key_values=past_key_values,
160
+ **kwargs,
161
+ )
162
+
163
+ return lm_output
164
+
165
+ def prepare_inputs_for_generation(
166
+ self,
167
+ input_ids: torch.Tensor,
168
+ audio_values: Optional[torch.FloatTensor] = None,
169
+ audio_token_start_idx: Optional[torch.Tensor] = None,
170
+ audio_token_len: Optional[torch.Tensor] = None,
171
+ past_key_values: Optional[Tuple] = None,
172
+ attention_mask: Optional[torch.Tensor] = None,
173
+ inputs_embeds: Optional[torch.Tensor] = None,
174
+ **kwargs,
175
+ ) -> Dict[str, Any]:
176
+ model_input = self.language_model.prepare_inputs_for_generation(
177
+ input_ids=input_ids,
178
+ past_key_values=past_key_values,
179
+ attention_mask=attention_mask,
180
+ inputs_embeds=inputs_embeds,
181
+ **kwargs,
182
+ )
183
+
184
+ if past_key_values is None and audio_values is not None:
185
+ # We only want to use audio features in the 1st generation step
186
+ model_input["audio_values"] = audio_values
187
+ model_input["audio_token_start_idx"] = audio_token_start_idx
188
+ model_input["audio_token_len"] = audio_token_len
189
+
190
+ return model_input
191
+
192
+ @classmethod
193
+ def _create_audio_tower(cls, config: ShukaConfig) -> Union[
194
+ transformers.Wav2Vec2Model,
195
+ transformers.models.whisper.modeling_whisper.WhisperEncoder,
196
+ ]:
197
+ if config.audio_model_id is not None:
198
+ if (
199
+ "whisper" in config.audio_model_id is not None
200
+ or "saaras" in config.audio_model_id is not None
201
+ ):
202
+ audio_tower = ModifiedWhisperEncoder.from_pretrained(
203
+ config.audio_model_id
204
+ )
205
+ else:
206
+ audio_tower = transformers.AutoModel.from_pretrained(
207
+ config.audio_model_id
208
+ )
209
+ else:
210
+ if (
211
+ "whisper" in config.audio_config._name_or_path
212
+ or "saaras" in config.audio_config._name_or_path
213
+ ):
214
+ audio_tower = ModifiedWhisperEncoder(config.audio_config)
215
+ else:
216
+ audio_tower = transformers.AutoModel.from_config(config.audio_config)
217
+
218
+ if isinstance(
219
+ audio_tower,
220
+ (transformers.Wav2Vec2BertModel, transformers.WhisperModel),
221
+ ):
222
+ # For these models we only need the encoder part
223
+ # Wav2Vec2BertModel -> Wav2Vec2BertEncoder
224
+ # WhisperModel -> WhisperEncoder
225
+ audio_tower = audio_tower.encoder
226
+
227
+ audio_tower = apply_lora(audio_tower, config.audio_model_lora_config)
228
+ return audio_tower
229
+
230
+ @classmethod
231
+ def _create_language_model(
232
+ cls, config: ShukaConfig
233
+ ) -> transformers.LlamaForCausalLM:
234
+ if config.text_model_id is not None:
235
+ language_model = transformers.AutoModelForCausalLM.from_pretrained(
236
+ config.text_model_id, attn_implementation=config._attn_implementation
237
+ )
238
+ else:
239
+ language_model = transformers.AutoModelForCausalLM.from_config(
240
+ config.text_config, attn_implementation=config._attn_implementation
241
+ )
242
+
243
+ language_model = apply_lora(language_model, config.text_model_lora_config)
244
+ return language_model
245
+
246
+ def merge_and_unload(self):
247
+ if isinstance(self.language_model, peft.PeftModel):
248
+ self.language_model = self.language_model.merge_and_unload()
249
+ if isinstance(self.audio_tower, peft.PeftModel):
250
+ self.audio_tower = self.audio_tower.merge_and_unload()
251
+
252
+ self.config.text_model_id = None
253
+ self.keep_params.update(
254
+ set(
255
+ [
256
+ f"language_model.{name}"
257
+ for name, _ in self.language_model.named_parameters()
258
+ ]
259
+ )
260
+ )
261
+ self.config.audio_model_id = None
262
+ self.keep_params.update(
263
+ set(
264
+ [
265
+ f"audio_tower.{name}"
266
+ for name, _ in self.audio_tower.named_parameters()
267
+ ]
268
+ )
269
+ )
270
+ for param in ["text_model_lora_config", "audio_model_lora_config"]:
271
+ if hasattr(self.config, param):
272
+ delattr(self.config, param)
273
+
274
+ def push_to_hub(self, *args, **kwargs):
275
+ self.merge_and_unload()
276
+ self.to(self.language_model.dtype)
277
+ return super().push_to_hub(*args, **kwargs)
278
+
279
+ def state_dict(self, *args, **kwargs):
280
+ named_params = dict(self.named_parameters())
281
+ state_dict = super().state_dict(*args, **kwargs)
282
+
283
+ state_dict = {
284
+ k: v
285
+ for k, v in state_dict.items()
286
+ if k in self.keep_params
287
+ or (k in named_params and named_params[k].requires_grad)
288
+ }
289
+ return state_dict
290
+
291
+ def load_state_dict(
292
+ self,
293
+ state_dict: Dict[str, Any],
294
+ *args,
295
+ **kwargs,
296
+ ):
297
+ self.keep_params.update(set(state_dict.keys()))
298
+ return super().load_state_dict(state_dict, *args, **kwargs)
299
+
300
+ def print_trainable_parameters(self):
301
+ """
302
+ Prints the number of trainable parameters in the model (reuses Peft model's method)
303
+ """
304
+ count_params = peft.peft_model.PeftModel.get_nb_trainable_parameters
305
+
306
+ trainable_params, all_param = count_params(self)
307
+
308
+ logging.info(
309
+ f"trainable params: {trainable_params:,d} || all params: {all_param:,d}"
310
+ f" || trainable%: {100 * trainable_params / all_param:.1f}%"
311
+ )
312
+
313
+ lm_trainable_params, lm_all_params = count_params(self.language_model)
314
+ audio_trainable_params, audio_all_params = count_params(self.audio_tower)
315
+
316
+ projector_trainable_params = (
317
+ trainable_params - lm_trainable_params - audio_trainable_params
318
+ )
319
+ projector_all_params = all_param - lm_all_params - audio_all_params
320
+
321
+ logging.info(
322
+ f"Trainable%: "
323
+ f" LLM: {100 * lm_trainable_params / lm_all_params:.1f}%"
324
+ f" || Audio Encoder: {100 * audio_trainable_params / audio_all_params:.1f}%"
325
+ f" || Projector: {100 * projector_trainable_params / projector_all_params:.1f}%"
326
+ )
327
+
328
+
329
+ def apply_lora(model: torch.nn.Module, lora_config: dict) -> torch.nn.Module:
330
+ """
331
+ Applies LoRA finetuning to the model. If the `r` parameter is set to 0, the model is frozen instead.
332
+ """
333
+ lora_config = peft.LoraConfig(**lora_config or {})
334
+
335
+ if lora_config.r == 0:
336
+ # freeze the model entirely
337
+ for param in model.parameters():
338
+ param.requires_grad = False
339
+ else:
340
+ model = peft.get_peft_model(model, lora_config)
341
+
342
+ return model
343
+
344
+
345
+ class StackAudioFrames(nn.Module):
346
+ """
347
+ Stack the audio embedding frames to reduce the sequence length by a factor of `stack_factor`.
348
+
349
+ The number of output frames will be `ceil(T / stack_factor) + 1` where `T` is the number of input frames.
350
+ NOTE: the extra +1 is intentional: in case the number of audio tokens are over-estimated by the processor,
351
+ we want to make sure `processor.audio_token_replacement` (i.e. EOS) doesn't get leaked into the middle of embeddings.
352
+ In most cases this extra padding will get removed in the model's forward function so it has no effect.
353
+ """
354
+
355
+ def __init__(self, stack_factor: int = 8):
356
+ super().__init__()
357
+ self.stack_factor = stack_factor
358
+
359
+ def forward(self, audio_embeds: torch.Tensor) -> torch.Tensor:
360
+ B, T, C = audio_embeds.shape
361
+ T_pad = (T + self.stack_factor - 1) // self.stack_factor * self.stack_factor
362
+ audio_embeds = F.pad(audio_embeds, (0, 0, 0, T_pad - T + self.stack_factor))
363
+ B, T, C = audio_embeds.shape
364
+ audio_embeds = audio_embeds.view(
365
+ B, T // self.stack_factor, C * self.stack_factor
366
+ )
367
+ return audio_embeds
368
+
369
+
370
+ class RMSNorm(transformers.models.llama.modeling_llama.LlamaRMSNorm):
371
+ def __init__(self, hidden_size: int, init: float = 1, eps: float = 1e-6):
372
+ super().__init__(hidden_size=hidden_size, eps=eps)
373
+ self.weight.data.fill_(init)
374
+
375
+
376
+ class SwiGLU(nn.Module):
377
+ def forward(self, x):
378
+ x, gate = x.chunk(2, dim=-1)
379
+ return F.silu(gate) * x
380
+
381
+
382
+ class ShukaProjector(nn.Sequential):
383
+ def __init__(self, config: ShukaConfig):
384
+ super().__init__()
385
+ self.hidden_dim = config.hidden_size
386
+ self._pad_and_stack = StackAudioFrames(config.stack_factor)
387
+ dim = config.audio_config.hidden_size * config.stack_factor
388
+ self.ln_pre = RMSNorm(dim, init=config.norm_init)
389
+ self.linear_1 = nn.Linear(dim, self.hidden_dim, bias=False)
390
+ dim = self.hidden_dim
391
+ self.act = transformers.activations.get_activation(config.projector_act)
392
+ dim = dim // 2 if config.projector_act == "swiglu" else dim
393
+ self.linear_2 = nn.Linear(dim, config.text_config.hidden_size, bias=False)
394
+ self.ln_post = RMSNorm(config.text_config.hidden_size, init=config.norm_init)
395
+
396
+ def forward(self, audio_features: torch.Tensor) -> torch.Tensor:
397
+ audio_features = self._pad_and_stack(audio_features)
398
+ audio_features = self.ln_pre(audio_features)
399
+ hidden_states = self.linear_1(audio_features)
400
+ hidden_states = self.act(hidden_states)
401
+ hidden_states = self.linear_2(hidden_states)
402
+ hidden_states = self.ln_post(hidden_states)
403
+ return hidden_states
404
+
405
+
406
+ ShukaConfig.register_for_auto_class()
407
+ ShukaModel.register_for_auto_class()
408
+
409
+ transformers.AutoConfig.register("shuka", ShukaConfig)
410
+ transformers.AutoModel.register(ShukaConfig, ShukaModel)
411
+
412
+ transformers.activations.ACT2FN["swiglu"] = SwiGLU
413
+ transformers.activations.ACT2FN["swiglu"] = SwiGLU
shuka_pipeline.py ADDED
@@ -0,0 +1,131 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ####################################################################
2
+ ## Built on top of Ultravox: https://github.com/fixie-ai/ultravox ##
3
+ ####################################################################
4
+
5
+ import logging
6
+ from typing import Any, Dict, List, Optional
7
+
8
+ import numpy as np
9
+ import transformers
10
+
11
+ # We must use relative import in this directory to allow uploading to HF Hub
12
+ # Even "from . import X" pattern doesn't work (undocumented and unclear why)
13
+ from .shuka_model import ShukaModel
14
+ from .shuka_processing import ShukaProcessor
15
+
16
+
17
+ class ShukaPipeline(transformers.Pipeline):
18
+ def __init__(
19
+ self,
20
+ model: ShukaModel,
21
+ tokenizer: Optional[transformers.PreTrainedTokenizerBase] = None,
22
+ audio_processor: Optional[transformers.ProcessorMixin] = None,
23
+ **kwargs
24
+ ):
25
+ if tokenizer is None:
26
+ try:
27
+ tokenizer = transformers.AutoTokenizer.from_pretrained(
28
+ model.config._name_or_path
29
+ )
30
+ except:
31
+ tokenizer = transformers.AutoTokenizer.from_pretrained(
32
+ model.config.text_model_id or model.config.text_config._name_or_path
33
+ )
34
+
35
+ if audio_processor is None:
36
+ audio_processor = transformers.AutoProcessor.from_pretrained(
37
+ model.config.audio_model_id or model.config.audio_config._name_or_path
38
+ )
39
+
40
+ self.processor = ShukaProcessor(
41
+ audio_processor=audio_processor,
42
+ tokenizer=tokenizer,
43
+ stack_factor=model.config.stack_factor,
44
+ )
45
+
46
+ super().__init__(model=model, tokenizer=tokenizer, **kwargs)
47
+
48
+ def _sanitize_parameters(self, **kwargs):
49
+ generation_keys = ["temperature", "max_new_tokens", "repetition_penalty"]
50
+ generation_kwargs = {k: kwargs[k] for k in kwargs if k in generation_keys}
51
+ return {}, generation_kwargs, {}
52
+
53
+ def preprocess(self, inputs: Dict[str, Any]):
54
+ turns: list = inputs.get("turns", [])
55
+
56
+ audio = inputs.get("audio", None)
57
+ # Convert to float32 if needed.
58
+ if isinstance(audio, np.ndarray):
59
+ if audio.dtype == np.float64:
60
+ audio = audio.astype(np.float32)
61
+ elif audio.dtype == np.int16:
62
+ audio = audio.astype(np.float32) / np.float32(32768.0)
63
+ elif audio.dtype == np.int32:
64
+ audio = audio.astype(np.float32) / np.float32(2147483648.0)
65
+
66
+ if audio is not None and (len(turns) == 0 or turns[-1]["role"] != "user"):
67
+ prompt = inputs.get("prompt", "<|audio|>")
68
+ if "<|audio|>" not in prompt:
69
+ logging.warning(
70
+ "Prompt does not contain '<|audio|>', appending '<|audio|>' to the end of the prompt."
71
+ )
72
+
73
+ prompt += " <|audio|>"
74
+ turns.append({"role": "user", "content": prompt})
75
+
76
+ text = self.processor.tokenizer.apply_chat_template(
77
+ turns, add_generation_prompt=True, tokenize=False
78
+ )
79
+
80
+ if "sampling_rate" not in inputs and audio is not None:
81
+ logging.warning(
82
+ "No sampling rate provided, using default of 16kHz. We highly recommend providing the correct sampling rate."
83
+ )
84
+
85
+ output = self.processor(
86
+ text=text,
87
+ audio=audio,
88
+ sampling_rate=inputs.get("sampling_rate", 16000),
89
+ )
90
+ if "audio_values" in output:
91
+ output["audio_values"] = output["audio_values"].to(self.model.dtype)
92
+
93
+ return output
94
+
95
+ def _forward(
96
+ self,
97
+ model_inputs: Dict[str, Any],
98
+ temperature: Optional[float] = None,
99
+ max_new_tokens: Optional[int] = None,
100
+ repetition_penalty: float = 1.1,
101
+ ) -> List[int]:
102
+ temperature = temperature or None
103
+ do_sample = temperature is not None
104
+
105
+ terminators = [self.tokenizer.eos_token_id]
106
+ if "<|eot_id|>" in self.tokenizer.added_tokens_encoder:
107
+ terminators.append(self.tokenizer.convert_tokens_to_ids("<|eot_id|>"))
108
+
109
+ input_len = model_inputs["input_ids"].shape[1]
110
+
111
+ outputs = self.model.generate(
112
+ **model_inputs,
113
+ do_sample=do_sample,
114
+ temperature=temperature,
115
+ max_new_tokens=max_new_tokens,
116
+ repetition_penalty=repetition_penalty,
117
+ eos_token_id=terminators
118
+ )
119
+ return outputs[0][input_len:]
120
+
121
+ def postprocess(self, model_outputs) -> str:
122
+ output_text = self.tokenizer.decode(model_outputs, skip_special_tokens=True)
123
+ return output_text
124
+
125
+
126
+ transformers.pipelines.PIPELINE_REGISTRY.register_pipeline(
127
+ "shuka-pipeline",
128
+ pipeline_class=ShukaPipeline,
129
+ pt_model=transformers.AutoModel,
130
+ type="multimodal",
131
+ )
shuka_processing.py ADDED
@@ -0,0 +1,176 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ####################################################################
2
+ ## Built on top of Ultravox: https://github.com/fixie-ai/ultravox ##
3
+ ####################################################################
4
+
5
+ from typing import Optional, Union
6
+
7
+ import numpy as np
8
+ import torch
9
+ import transformers
10
+
11
+
12
+ class ShukaProcessor(transformers.ProcessorMixin):
13
+ """
14
+ Constructs an Shuka processor which wraps an audio processor and a tokenizer into a single processor.
15
+
16
+ Args:
17
+ audio_processor: The audio processor for the audio encoder.
18
+ tokenizer: The tokenizer for the language model.
19
+ """
20
+
21
+ attributes = ["audio_processor", "tokenizer"]
22
+ audio_processor_class = (
23
+ "Wav2Vec2Processor",
24
+ "SeamlessM4TFeatureExtractor",
25
+ "WhisperProcessor",
26
+ )
27
+ tokenizer_class = (
28
+ "PreTrainedTokenizer",
29
+ "PreTrainedTokenizerFast",
30
+ )
31
+
32
+ tokenizer: transformers.PreTrainedTokenizerBase
33
+ audio_processor: transformers.ProcessorMixin
34
+
35
+ def __init__(
36
+ self,
37
+ audio_processor=None,
38
+ tokenizer=None,
39
+ audio_padding: str = "longest",
40
+ encoder_ds_factor: int = 320,
41
+ stack_factor: int = 8,
42
+ audio_placeholder: str = "<|audio|>",
43
+ ):
44
+ """
45
+ Args:
46
+ audio_processor: The audio processor for the audio encoder.
47
+ tokenizer: The tokenizer for the language model.
48
+ audio_padding: The padding strategy for the audio encoder.
49
+ encoder_ds_factor: The downsample factor of the audio encoder.
50
+ stack_factor: The factor by which the audio encoder output is stacked in the multimodal projector.
51
+ audio_placeholder: The placeholder for the audio in the text.
52
+ """
53
+ self.audio_padding = audio_padding
54
+ self.encoder_ds_factor = encoder_ds_factor
55
+ self.stack_factor = stack_factor
56
+ self.audio_placeholder = audio_placeholder
57
+ self.audio_token_replacement = tokenizer.eos_token
58
+ assert (
59
+ self.audio_token_replacement is not None
60
+ ), "The tokenizer has no EOS token. Cannot recover."
61
+ super().__init__(audio_processor=audio_processor, tokenizer=tokenizer)
62
+
63
+ def __call__(
64
+ self,
65
+ text: Optional[str] = None,
66
+ audio: Optional[Union[np.ndarray, torch.Tensor]] = None,
67
+ sampling_rate: Optional[int] = None,
68
+ return_tensors: Optional[
69
+ Union[str, transformers.TensorType]
70
+ ] = transformers.TensorType.PYTORCH,
71
+ **kwargs,
72
+ ) -> transformers.BatchFeature:
73
+ """
74
+ Main method to prepare for the model one text sequence and audio. This method forwards the `text`
75
+ and `kwargs` arguments to PreTrainedTokenizerFast's [`~PreTrainedTokenizerFast.__call__`] if `text` is not `None` to encode
76
+ the text. To prepare the audio(s), this method forwards the `audio`, `sampling_rate` and `kwargs` arguments to
77
+ audio processor's [`~Wav2Vec2Processor.__call__`] if `audio` is not `None`. Please refer to the docstring
78
+ of the above two methods for more information.
79
+
80
+ Args:
81
+ text (`str`, `List[str]`):
82
+ The sequence to be encoded. Sequence can be a string or (pretokenized string).
83
+ audio (`np.ndarray`, `torch.Tensor`, `List[np.ndarray]`, `List[torch.Tensor]`):
84
+ The audio to be prepared. Audio can be NumPy array or PyTorch tensor. In case of a
85
+ NumPy array/PyTorch tensor, each audio should be of shape (C, T), where C is a number of channels, and T the
86
+ sample length of the audio.
87
+ sampling_rate (`int`, *optional*, defaults to 16000):
88
+ Sampling rate of the input audio. We expect 16kHz audio. Don't change this value unless you know what
89
+ you are doing.
90
+ return_tensors (`str` or [`~utils.TensorType`], *optional*):
91
+ If set, will return tensors of a particular framework. Acceptable values are:
92
+
93
+ - `'tf'`: Return TensorFlow `tf.constant` objects.
94
+ - `'pt'`: Return PyTorch `torch.Tensor` objects.
95
+ - `'np'`: Return NumPy `np.ndarray` objects.
96
+ - `'jax'`: Return JAX `jnp.ndarray` objects.
97
+
98
+ Returns:
99
+ [`BatchFeature`]: A [`BatchFeature`] with the following fields:
100
+
101
+ - **input_ids** -- List of token ids to be fed to a model. Returned when `text` is not `None`.
102
+ - **attention_mask** -- List of indices specifying which tokens should be attended to by the model (when
103
+ `return_attention_mask=True` or if *"attention_mask"* is in `self.model_input_names` and if `text` is not
104
+ `None`).
105
+ - **audio_values** -- Processed audio values to be fed to a model. Returned when `audio` is not `None`.
106
+ - **audio_token_len** -- Predicted number of audio frames: this value is guaranteed to be a close upper bound.
107
+ Returned when `audio` is not `None`.
108
+ - **audio_token_start_idx** -- The index in the tokenized text where the audio starts. Returned when `audio` is not `None`.
109
+ """
110
+ # TODO: Add support for multiple audio and text inputs.
111
+ data = {}
112
+ audio_embed_frames = 0
113
+ if audio is not None and len(audio) > 0:
114
+ if self.audio_padding == "max_length":
115
+ # 30 seconds is the expected length for Whisper
116
+ assert sampling_rate is not None, "Sampling rate must be provided."
117
+ audio_len = 30 * sampling_rate
118
+ else:
119
+ audio_len = audio.shape[-1]
120
+ # It's guaranteed that the number of frames is less than or equal to this amount.
121
+ # For Whisper this is exact AFAICT, but for Wav2Vec2 it's an upper bound.
122
+ # Currently, StackAudioFrames makes sure an over-estimation won't cause issues by padding the audio embeddings.
123
+ nb_encoder_frames = int(round(audio_len / self.encoder_ds_factor + 1e-4))
124
+ audio_embed_frames = int(np.ceil(nb_encoder_frames / self.stack_factor))
125
+ data["audio_token_len"] = [audio_embed_frames]
126
+
127
+ x = self.audio_processor(
128
+ audio,
129
+ sampling_rate=sampling_rate,
130
+ padding="longest",
131
+ max_length=audio_len,
132
+ **kwargs,
133
+ )
134
+ if "input_features" in x:
135
+ data["audio_values"] = x.input_features
136
+ else:
137
+ data["audio_values"] = x.input_values
138
+
139
+ if text is not None:
140
+ assert isinstance(
141
+ text, str
142
+ ), "Text must be a string. Batch mode not supported yet."
143
+ if self.audio_placeholder in text:
144
+ if "audio_token_len" not in data:
145
+ raise ValueError(
146
+ f"audio must be provided when using audio placeholder ({self.audio_placeholder}) in text."
147
+ )
148
+
149
+ start_idx = len(
150
+ self.tokenizer.encode(
151
+ text[: text.index(self.audio_placeholder)],
152
+ add_special_tokens=False,
153
+ )
154
+ )
155
+ data["audio_token_start_idx"] = [start_idx]
156
+ text = text.replace(
157
+ self.audio_placeholder,
158
+ self.audio_token_replacement * audio_embed_frames,
159
+ )
160
+
161
+ # Special tokens like BOS should already have been added by the caller.
162
+ data.update(self.tokenizer([text], add_special_tokens=False, **kwargs))
163
+
164
+ return transformers.BatchFeature(data=data, tensor_type=return_tensors)
165
+
166
+ def batch_decode(self, *args, **kwargs):
167
+ return self.tokenizer.batch_decode(*args, **kwargs)
168
+
169
+ def decode(self, *args, **kwargs):
170
+ return self.tokenizer.decode(*args, **kwargs)
171
+
172
+ @property
173
+ def model_input_names(self):
174
+ tokenizer_input_names = self.tokenizer.model_input_names
175
+ audio_processor_input_names = self.audio_processor.model_input_names
176
+ return list(set(tokenizer_input_names + audio_processor_input_names))
special_tokens_map.json ADDED
@@ -0,0 +1,17 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "content": "<|begin_of_text|>",
4
+ "lstrip": false,
5
+ "normalized": false,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "eos_token": {
10
+ "content": "<|eot_id|>",
11
+ "lstrip": false,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "pad_token": "<|eot_id|>"
17
+ }
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1,2063 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "128000": {
4
+ "content": "<|begin_of_text|>",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "128001": {
12
+ "content": "<|end_of_text|>",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "128002": {
20
+ "content": "<|reserved_special_token_0|>",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "128003": {
28
+ "content": "<|reserved_special_token_1|>",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ },
35
+ "128004": {
36
+ "content": "<|reserved_special_token_2|>",
37
+ "lstrip": false,
38
+ "normalized": false,
39
+ "rstrip": false,
40
+ "single_word": false,
41
+ "special": true
42
+ },
43
+ "128005": {
44
+ "content": "<|reserved_special_token_3|>",
45
+ "lstrip": false,
46
+ "normalized": false,
47
+ "rstrip": false,
48
+ "single_word": false,
49
+ "special": true
50
+ },
51
+ "128006": {
52
+ "content": "<|start_header_id|>",
53
+ "lstrip": false,
54
+ "normalized": false,
55
+ "rstrip": false,
56
+ "single_word": false,
57
+ "special": true
58
+ },
59
+ "128007": {
60
+ "content": "<|end_header_id|>",
61
+ "lstrip": false,
62
+ "normalized": false,
63
+ "rstrip": false,
64
+ "single_word": false,
65
+ "special": true
66
+ },
67
+ "128008": {
68
+ "content": "<|reserved_special_token_4|>",
69
+ "lstrip": false,
70
+ "normalized": false,
71
+ "rstrip": false,
72
+ "single_word": false,
73
+ "special": true
74
+ },
75
+ "128009": {
76
+ "content": "<|eot_id|>",
77
+ "lstrip": false,
78
+ "normalized": false,
79
+ "rstrip": false,
80
+ "single_word": false,
81
+ "special": true
82
+ },
83
+ "128010": {
84
+ "content": "<|reserved_special_token_5|>",
85
+ "lstrip": false,
86
+ "normalized": false,
87
+ "rstrip": false,
88
+ "single_word": false,
89
+ "special": true
90
+ },
91
+ "128011": {
92
+ "content": "<|reserved_special_token_6|>",
93
+ "lstrip": false,
94
+ "normalized": false,
95
+ "rstrip": false,
96
+ "single_word": false,
97
+ "special": true
98
+ },
99
+ "128012": {
100
+ "content": "<|reserved_special_token_7|>",
101
+ "lstrip": false,
102
+ "normalized": false,
103
+ "rstrip": false,
104
+ "single_word": false,
105
+ "special": true
106
+ },
107
+ "128013": {
108
+ "content": "<|reserved_special_token_8|>",
109
+ "lstrip": false,
110
+ "normalized": false,
111
+ "rstrip": false,
112
+ "single_word": false,
113
+ "special": true
114
+ },
115
+ "128014": {
116
+ "content": "<|reserved_special_token_9|>",
117
+ "lstrip": false,
118
+ "normalized": false,
119
+ "rstrip": false,
120
+ "single_word": false,
121
+ "special": true
122
+ },
123
+ "128015": {
124
+ "content": "<|reserved_special_token_10|>",
125
+ "lstrip": false,
126
+ "normalized": false,
127
+ "rstrip": false,
128
+ "single_word": false,
129
+ "special": true
130
+ },
131
+ "128016": {
132
+ "content": "<|reserved_special_token_11|>",
133
+ "lstrip": false,
134
+ "normalized": false,
135
+ "rstrip": false,
136
+ "single_word": false,
137
+ "special": true
138
+ },
139
+ "128017": {
140
+ "content": "<|reserved_special_token_12|>",
141
+ "lstrip": false,
142
+ "normalized": false,
143
+ "rstrip": false,
144
+ "single_word": false,
145
+ "special": true
146
+ },
147
+ "128018": {
148
+ "content": "<|reserved_special_token_13|>",
149
+ "lstrip": false,
150
+ "normalized": false,
151
+ "rstrip": false,
152
+ "single_word": false,
153
+ "special": true
154
+ },
155
+ "128019": {
156
+ "content": "<|reserved_special_token_14|>",
157
+ "lstrip": false,
158
+ "normalized": false,
159
+ "rstrip": false,
160
+ "single_word": false,
161
+ "special": true
162
+ },
163
+ "128020": {
164
+ "content": "<|reserved_special_token_15|>",
165
+ "lstrip": false,
166
+ "normalized": false,
167
+ "rstrip": false,
168
+ "single_word": false,
169
+ "special": true
170
+ },
171
+ "128021": {
172
+ "content": "<|reserved_special_token_16|>",
173
+ "lstrip": false,
174
+ "normalized": false,
175
+ "rstrip": false,
176
+ "single_word": false,
177
+ "special": true
178
+ },
179
+ "128022": {
180
+ "content": "<|reserved_special_token_17|>",
181
+ "lstrip": false,
182
+ "normalized": false,
183
+ "rstrip": false,
184
+ "single_word": false,
185
+ "special": true
186
+ },
187
+ "128023": {
188
+ "content": "<|reserved_special_token_18|>",
189
+ "lstrip": false,
190
+ "normalized": false,
191
+ "rstrip": false,
192
+ "single_word": false,
193
+ "special": true
194
+ },
195
+ "128024": {
196
+ "content": "<|reserved_special_token_19|>",
197
+ "lstrip": false,
198
+ "normalized": false,
199
+ "rstrip": false,
200
+ "single_word": false,
201
+ "special": true
202
+ },
203
+ "128025": {
204
+ "content": "<|reserved_special_token_20|>",
205
+ "lstrip": false,
206
+ "normalized": false,
207
+ "rstrip": false,
208
+ "single_word": false,
209
+ "special": true
210
+ },
211
+ "128026": {
212
+ "content": "<|reserved_special_token_21|>",
213
+ "lstrip": false,
214
+ "normalized": false,
215
+ "rstrip": false,
216
+ "single_word": false,
217
+ "special": true
218
+ },
219
+ "128027": {
220
+ "content": "<|reserved_special_token_22|>",
221
+ "lstrip": false,
222
+ "normalized": false,
223
+ "rstrip": false,
224
+ "single_word": false,
225
+ "special": true
226
+ },
227
+ "128028": {
228
+ "content": "<|reserved_special_token_23|>",
229
+ "lstrip": false,
230
+ "normalized": false,
231
+ "rstrip": false,
232
+ "single_word": false,
233
+ "special": true
234
+ },
235
+ "128029": {
236
+ "content": "<|reserved_special_token_24|>",
237
+ "lstrip": false,
238
+ "normalized": false,
239
+ "rstrip": false,
240
+ "single_word": false,
241
+ "special": true
242
+ },
243
+ "128030": {
244
+ "content": "<|reserved_special_token_25|>",
245
+ "lstrip": false,
246
+ "normalized": false,
247
+ "rstrip": false,
248
+ "single_word": false,
249
+ "special": true
250
+ },
251
+ "128031": {
252
+ "content": "<|reserved_special_token_26|>",
253
+ "lstrip": false,
254
+ "normalized": false,
255
+ "rstrip": false,
256
+ "single_word": false,
257
+ "special": true
258
+ },
259
+ "128032": {
260
+ "content": "<|reserved_special_token_27|>",
261
+ "lstrip": false,
262
+ "normalized": false,
263
+ "rstrip": false,
264
+ "single_word": false,
265
+ "special": true
266
+ },
267
+ "128033": {
268
+ "content": "<|reserved_special_token_28|>",
269
+ "lstrip": false,
270
+ "normalized": false,
271
+ "rstrip": false,
272
+ "single_word": false,
273
+ "special": true
274
+ },
275
+ "128034": {
276
+ "content": "<|reserved_special_token_29|>",
277
+ "lstrip": false,
278
+ "normalized": false,
279
+ "rstrip": false,
280
+ "single_word": false,
281
+ "special": true
282
+ },
283
+ "128035": {
284
+ "content": "<|reserved_special_token_30|>",
285
+ "lstrip": false,
286
+ "normalized": false,
287
+ "rstrip": false,
288
+ "single_word": false,
289
+ "special": true
290
+ },
291
+ "128036": {
292
+ "content": "<|reserved_special_token_31|>",
293
+ "lstrip": false,
294
+ "normalized": false,
295
+ "rstrip": false,
296
+ "single_word": false,
297
+ "special": true
298
+ },
299
+ "128037": {
300
+ "content": "<|reserved_special_token_32|>",
301
+ "lstrip": false,
302
+ "normalized": false,
303
+ "rstrip": false,
304
+ "single_word": false,
305
+ "special": true
306
+ },
307
+ "128038": {
308
+ "content": "<|reserved_special_token_33|>",
309
+ "lstrip": false,
310
+ "normalized": false,
311
+ "rstrip": false,
312
+ "single_word": false,
313
+ "special": true
314
+ },
315
+ "128039": {
316
+ "content": "<|reserved_special_token_34|>",
317
+ "lstrip": false,
318
+ "normalized": false,
319
+ "rstrip": false,
320
+ "single_word": false,
321
+ "special": true
322
+ },
323
+ "128040": {
324
+ "content": "<|reserved_special_token_35|>",
325
+ "lstrip": false,
326
+ "normalized": false,
327
+ "rstrip": false,
328
+ "single_word": false,
329
+ "special": true
330
+ },
331
+ "128041": {
332
+ "content": "<|reserved_special_token_36|>",
333
+ "lstrip": false,
334
+ "normalized": false,
335
+ "rstrip": false,
336
+ "single_word": false,
337
+ "special": true
338
+ },
339
+ "128042": {
340
+ "content": "<|reserved_special_token_37|>",
341
+ "lstrip": false,
342
+ "normalized": false,
343
+ "rstrip": false,
344
+ "single_word": false,
345
+ "special": true
346
+ },
347
+ "128043": {
348
+ "content": "<|reserved_special_token_38|>",
349
+ "lstrip": false,
350
+ "normalized": false,
351
+ "rstrip": false,
352
+ "single_word": false,
353
+ "special": true
354
+ },
355
+ "128044": {
356
+ "content": "<|reserved_special_token_39|>",
357
+ "lstrip": false,
358
+ "normalized": false,
359
+ "rstrip": false,
360
+ "single_word": false,
361
+ "special": true
362
+ },
363
+ "128045": {
364
+ "content": "<|reserved_special_token_40|>",
365
+ "lstrip": false,
366
+ "normalized": false,
367
+ "rstrip": false,
368
+ "single_word": false,
369
+ "special": true
370
+ },
371
+ "128046": {
372
+ "content": "<|reserved_special_token_41|>",
373
+ "lstrip": false,
374
+ "normalized": false,
375
+ "rstrip": false,
376
+ "single_word": false,
377
+ "special": true
378
+ },
379
+ "128047": {
380
+ "content": "<|reserved_special_token_42|>",
381
+ "lstrip": false,
382
+ "normalized": false,
383
+ "rstrip": false,
384
+ "single_word": false,
385
+ "special": true
386
+ },
387
+ "128048": {
388
+ "content": "<|reserved_special_token_43|>",
389
+ "lstrip": false,
390
+ "normalized": false,
391
+ "rstrip": false,
392
+ "single_word": false,
393
+ "special": true
394
+ },
395
+ "128049": {
396
+ "content": "<|reserved_special_token_44|>",
397
+ "lstrip": false,
398
+ "normalized": false,
399
+ "rstrip": false,
400
+ "single_word": false,
401
+ "special": true
402
+ },
403
+ "128050": {
404
+ "content": "<|reserved_special_token_45|>",
405
+ "lstrip": false,
406
+ "normalized": false,
407
+ "rstrip": false,
408
+ "single_word": false,
409
+ "special": true
410
+ },
411
+ "128051": {
412
+ "content": "<|reserved_special_token_46|>",
413
+ "lstrip": false,
414
+ "normalized": false,
415
+ "rstrip": false,
416
+ "single_word": false,
417
+ "special": true
418
+ },
419
+ "128052": {
420
+ "content": "<|reserved_special_token_47|>",
421
+ "lstrip": false,
422
+ "normalized": false,
423
+ "rstrip": false,
424
+ "single_word": false,
425
+ "special": true
426
+ },
427
+ "128053": {
428
+ "content": "<|reserved_special_token_48|>",
429
+ "lstrip": false,
430
+ "normalized": false,
431
+ "rstrip": false,
432
+ "single_word": false,
433
+ "special": true
434
+ },
435
+ "128054": {
436
+ "content": "<|reserved_special_token_49|>",
437
+ "lstrip": false,
438
+ "normalized": false,
439
+ "rstrip": false,
440
+ "single_word": false,
441
+ "special": true
442
+ },
443
+ "128055": {
444
+ "content": "<|reserved_special_token_50|>",
445
+ "lstrip": false,
446
+ "normalized": false,
447
+ "rstrip": false,
448
+ "single_word": false,
449
+ "special": true
450
+ },
451
+ "128056": {
452
+ "content": "<|reserved_special_token_51|>",
453
+ "lstrip": false,
454
+ "normalized": false,
455
+ "rstrip": false,
456
+ "single_word": false,
457
+ "special": true
458
+ },
459
+ "128057": {
460
+ "content": "<|reserved_special_token_52|>",
461
+ "lstrip": false,
462
+ "normalized": false,
463
+ "rstrip": false,
464
+ "single_word": false,
465
+ "special": true
466
+ },
467
+ "128058": {
468
+ "content": "<|reserved_special_token_53|>",
469
+ "lstrip": false,
470
+ "normalized": false,
471
+ "rstrip": false,
472
+ "single_word": false,
473
+ "special": true
474
+ },
475
+ "128059": {
476
+ "content": "<|reserved_special_token_54|>",
477
+ "lstrip": false,
478
+ "normalized": false,
479
+ "rstrip": false,
480
+ "single_word": false,
481
+ "special": true
482
+ },
483
+ "128060": {
484
+ "content": "<|reserved_special_token_55|>",
485
+ "lstrip": false,
486
+ "normalized": false,
487
+ "rstrip": false,
488
+ "single_word": false,
489
+ "special": true
490
+ },
491
+ "128061": {
492
+ "content": "<|reserved_special_token_56|>",
493
+ "lstrip": false,
494
+ "normalized": false,
495
+ "rstrip": false,
496
+ "single_word": false,
497
+ "special": true
498
+ },
499
+ "128062": {
500
+ "content": "<|reserved_special_token_57|>",
501
+ "lstrip": false,
502
+ "normalized": false,
503
+ "rstrip": false,
504
+ "single_word": false,
505
+ "special": true
506
+ },
507
+ "128063": {
508
+ "content": "<|reserved_special_token_58|>",
509
+ "lstrip": false,
510
+ "normalized": false,
511
+ "rstrip": false,
512
+ "single_word": false,
513
+ "special": true
514
+ },
515
+ "128064": {
516
+ "content": "<|reserved_special_token_59|>",
517
+ "lstrip": false,
518
+ "normalized": false,
519
+ "rstrip": false,
520
+ "single_word": false,
521
+ "special": true
522
+ },
523
+ "128065": {
524
+ "content": "<|reserved_special_token_60|>",
525
+ "lstrip": false,
526
+ "normalized": false,
527
+ "rstrip": false,
528
+ "single_word": false,
529
+ "special": true
530
+ },
531
+ "128066": {
532
+ "content": "<|reserved_special_token_61|>",
533
+ "lstrip": false,
534
+ "normalized": false,
535
+ "rstrip": false,
536
+ "single_word": false,
537
+ "special": true
538
+ },
539
+ "128067": {
540
+ "content": "<|reserved_special_token_62|>",
541
+ "lstrip": false,
542
+ "normalized": false,
543
+ "rstrip": false,
544
+ "single_word": false,
545
+ "special": true
546
+ },
547
+ "128068": {
548
+ "content": "<|reserved_special_token_63|>",
549
+ "lstrip": false,
550
+ "normalized": false,
551
+ "rstrip": false,
552
+ "single_word": false,
553
+ "special": true
554
+ },
555
+ "128069": {
556
+ "content": "<|reserved_special_token_64|>",
557
+ "lstrip": false,
558
+ "normalized": false,
559
+ "rstrip": false,
560
+ "single_word": false,
561
+ "special": true
562
+ },
563
+ "128070": {
564
+ "content": "<|reserved_special_token_65|>",
565
+ "lstrip": false,
566
+ "normalized": false,
567
+ "rstrip": false,
568
+ "single_word": false,
569
+ "special": true
570
+ },
571
+ "128071": {
572
+ "content": "<|reserved_special_token_66|>",
573
+ "lstrip": false,
574
+ "normalized": false,
575
+ "rstrip": false,
576
+ "single_word": false,
577
+ "special": true
578
+ },
579
+ "128072": {
580
+ "content": "<|reserved_special_token_67|>",
581
+ "lstrip": false,
582
+ "normalized": false,
583
+ "rstrip": false,
584
+ "single_word": false,
585
+ "special": true
586
+ },
587
+ "128073": {
588
+ "content": "<|reserved_special_token_68|>",
589
+ "lstrip": false,
590
+ "normalized": false,
591
+ "rstrip": false,
592
+ "single_word": false,
593
+ "special": true
594
+ },
595
+ "128074": {
596
+ "content": "<|reserved_special_token_69|>",
597
+ "lstrip": false,
598
+ "normalized": false,
599
+ "rstrip": false,
600
+ "single_word": false,
601
+ "special": true
602
+ },
603
+ "128075": {
604
+ "content": "<|reserved_special_token_70|>",
605
+ "lstrip": false,
606
+ "normalized": false,
607
+ "rstrip": false,
608
+ "single_word": false,
609
+ "special": true
610
+ },
611
+ "128076": {
612
+ "content": "<|reserved_special_token_71|>",
613
+ "lstrip": false,
614
+ "normalized": false,
615
+ "rstrip": false,
616
+ "single_word": false,
617
+ "special": true
618
+ },
619
+ "128077": {
620
+ "content": "<|reserved_special_token_72|>",
621
+ "lstrip": false,
622
+ "normalized": false,
623
+ "rstrip": false,
624
+ "single_word": false,
625
+ "special": true
626
+ },
627
+ "128078": {
628
+ "content": "<|reserved_special_token_73|>",
629
+ "lstrip": false,
630
+ "normalized": false,
631
+ "rstrip": false,
632
+ "single_word": false,
633
+ "special": true
634
+ },
635
+ "128079": {
636
+ "content": "<|reserved_special_token_74|>",
637
+ "lstrip": false,
638
+ "normalized": false,
639
+ "rstrip": false,
640
+ "single_word": false,
641
+ "special": true
642
+ },
643
+ "128080": {
644
+ "content": "<|reserved_special_token_75|>",
645
+ "lstrip": false,
646
+ "normalized": false,
647
+ "rstrip": false,
648
+ "single_word": false,
649
+ "special": true
650
+ },
651
+ "128081": {
652
+ "content": "<|reserved_special_token_76|>",
653
+ "lstrip": false,
654
+ "normalized": false,
655
+ "rstrip": false,
656
+ "single_word": false,
657
+ "special": true
658
+ },
659
+ "128082": {
660
+ "content": "<|reserved_special_token_77|>",
661
+ "lstrip": false,
662
+ "normalized": false,
663
+ "rstrip": false,
664
+ "single_word": false,
665
+ "special": true
666
+ },
667
+ "128083": {
668
+ "content": "<|reserved_special_token_78|>",
669
+ "lstrip": false,
670
+ "normalized": false,
671
+ "rstrip": false,
672
+ "single_word": false,
673
+ "special": true
674
+ },
675
+ "128084": {
676
+ "content": "<|reserved_special_token_79|>",
677
+ "lstrip": false,
678
+ "normalized": false,
679
+ "rstrip": false,
680
+ "single_word": false,
681
+ "special": true
682
+ },
683
+ "128085": {
684
+ "content": "<|reserved_special_token_80|>",
685
+ "lstrip": false,
686
+ "normalized": false,
687
+ "rstrip": false,
688
+ "single_word": false,
689
+ "special": true
690
+ },
691
+ "128086": {
692
+ "content": "<|reserved_special_token_81|>",
693
+ "lstrip": false,
694
+ "normalized": false,
695
+ "rstrip": false,
696
+ "single_word": false,
697
+ "special": true
698
+ },
699
+ "128087": {
700
+ "content": "<|reserved_special_token_82|>",
701
+ "lstrip": false,
702
+ "normalized": false,
703
+ "rstrip": false,
704
+ "single_word": false,
705
+ "special": true
706
+ },
707
+ "128088": {
708
+ "content": "<|reserved_special_token_83|>",
709
+ "lstrip": false,
710
+ "normalized": false,
711
+ "rstrip": false,
712
+ "single_word": false,
713
+ "special": true
714
+ },
715
+ "128089": {
716
+ "content": "<|reserved_special_token_84|>",
717
+ "lstrip": false,
718
+ "normalized": false,
719
+ "rstrip": false,
720
+ "single_word": false,
721
+ "special": true
722
+ },
723
+ "128090": {
724
+ "content": "<|reserved_special_token_85|>",
725
+ "lstrip": false,
726
+ "normalized": false,
727
+ "rstrip": false,
728
+ "single_word": false,
729
+ "special": true
730
+ },
731
+ "128091": {
732
+ "content": "<|reserved_special_token_86|>",
733
+ "lstrip": false,
734
+ "normalized": false,
735
+ "rstrip": false,
736
+ "single_word": false,
737
+ "special": true
738
+ },
739
+ "128092": {
740
+ "content": "<|reserved_special_token_87|>",
741
+ "lstrip": false,
742
+ "normalized": false,
743
+ "rstrip": false,
744
+ "single_word": false,
745
+ "special": true
746
+ },
747
+ "128093": {
748
+ "content": "<|reserved_special_token_88|>",
749
+ "lstrip": false,
750
+ "normalized": false,
751
+ "rstrip": false,
752
+ "single_word": false,
753
+ "special": true
754
+ },
755
+ "128094": {
756
+ "content": "<|reserved_special_token_89|>",
757
+ "lstrip": false,
758
+ "normalized": false,
759
+ "rstrip": false,
760
+ "single_word": false,
761
+ "special": true
762
+ },
763
+ "128095": {
764
+ "content": "<|reserved_special_token_90|>",
765
+ "lstrip": false,
766
+ "normalized": false,
767
+ "rstrip": false,
768
+ "single_word": false,
769
+ "special": true
770
+ },
771
+ "128096": {
772
+ "content": "<|reserved_special_token_91|>",
773
+ "lstrip": false,
774
+ "normalized": false,
775
+ "rstrip": false,
776
+ "single_word": false,
777
+ "special": true
778
+ },
779
+ "128097": {
780
+ "content": "<|reserved_special_token_92|>",
781
+ "lstrip": false,
782
+ "normalized": false,
783
+ "rstrip": false,
784
+ "single_word": false,
785
+ "special": true
786
+ },
787
+ "128098": {
788
+ "content": "<|reserved_special_token_93|>",
789
+ "lstrip": false,
790
+ "normalized": false,
791
+ "rstrip": false,
792
+ "single_word": false,
793
+ "special": true
794
+ },
795
+ "128099": {
796
+ "content": "<|reserved_special_token_94|>",
797
+ "lstrip": false,
798
+ "normalized": false,
799
+ "rstrip": false,
800
+ "single_word": false,
801
+ "special": true
802
+ },
803
+ "128100": {
804
+ "content": "<|reserved_special_token_95|>",
805
+ "lstrip": false,
806
+ "normalized": false,
807
+ "rstrip": false,
808
+ "single_word": false,
809
+ "special": true
810
+ },
811
+ "128101": {
812
+ "content": "<|reserved_special_token_96|>",
813
+ "lstrip": false,
814
+ "normalized": false,
815
+ "rstrip": false,
816
+ "single_word": false,
817
+ "special": true
818
+ },
819
+ "128102": {
820
+ "content": "<|reserved_special_token_97|>",
821
+ "lstrip": false,
822
+ "normalized": false,
823
+ "rstrip": false,
824
+ "single_word": false,
825
+ "special": true
826
+ },
827
+ "128103": {
828
+ "content": "<|reserved_special_token_98|>",
829
+ "lstrip": false,
830
+ "normalized": false,
831
+ "rstrip": false,
832
+ "single_word": false,
833
+ "special": true
834
+ },
835
+ "128104": {
836
+ "content": "<|reserved_special_token_99|>",
837
+ "lstrip": false,
838
+ "normalized": false,
839
+ "rstrip": false,
840
+ "single_word": false,
841
+ "special": true
842
+ },
843
+ "128105": {
844
+ "content": "<|reserved_special_token_100|>",
845
+ "lstrip": false,
846
+ "normalized": false,
847
+ "rstrip": false,
848
+ "single_word": false,
849
+ "special": true
850
+ },
851
+ "128106": {
852
+ "content": "<|reserved_special_token_101|>",
853
+ "lstrip": false,
854
+ "normalized": false,
855
+ "rstrip": false,
856
+ "single_word": false,
857
+ "special": true
858
+ },
859
+ "128107": {
860
+ "content": "<|reserved_special_token_102|>",
861
+ "lstrip": false,
862
+ "normalized": false,
863
+ "rstrip": false,
864
+ "single_word": false,
865
+ "special": true
866
+ },
867
+ "128108": {
868
+ "content": "<|reserved_special_token_103|>",
869
+ "lstrip": false,
870
+ "normalized": false,
871
+ "rstrip": false,
872
+ "single_word": false,
873
+ "special": true
874
+ },
875
+ "128109": {
876
+ "content": "<|reserved_special_token_104|>",
877
+ "lstrip": false,
878
+ "normalized": false,
879
+ "rstrip": false,
880
+ "single_word": false,
881
+ "special": true
882
+ },
883
+ "128110": {
884
+ "content": "<|reserved_special_token_105|>",
885
+ "lstrip": false,
886
+ "normalized": false,
887
+ "rstrip": false,
888
+ "single_word": false,
889
+ "special": true
890
+ },
891
+ "128111": {
892
+ "content": "<|reserved_special_token_106|>",
893
+ "lstrip": false,
894
+ "normalized": false,
895
+ "rstrip": false,
896
+ "single_word": false,
897
+ "special": true
898
+ },
899
+ "128112": {
900
+ "content": "<|reserved_special_token_107|>",
901
+ "lstrip": false,
902
+ "normalized": false,
903
+ "rstrip": false,
904
+ "single_word": false,
905
+ "special": true
906
+ },
907
+ "128113": {
908
+ "content": "<|reserved_special_token_108|>",
909
+ "lstrip": false,
910
+ "normalized": false,
911
+ "rstrip": false,
912
+ "single_word": false,
913
+ "special": true
914
+ },
915
+ "128114": {
916
+ "content": "<|reserved_special_token_109|>",
917
+ "lstrip": false,
918
+ "normalized": false,
919
+ "rstrip": false,
920
+ "single_word": false,
921
+ "special": true
922
+ },
923
+ "128115": {
924
+ "content": "<|reserved_special_token_110|>",
925
+ "lstrip": false,
926
+ "normalized": false,
927
+ "rstrip": false,
928
+ "single_word": false,
929
+ "special": true
930
+ },
931
+ "128116": {
932
+ "content": "<|reserved_special_token_111|>",
933
+ "lstrip": false,
934
+ "normalized": false,
935
+ "rstrip": false,
936
+ "single_word": false,
937
+ "special": true
938
+ },
939
+ "128117": {
940
+ "content": "<|reserved_special_token_112|>",
941
+ "lstrip": false,
942
+ "normalized": false,
943
+ "rstrip": false,
944
+ "single_word": false,
945
+ "special": true
946
+ },
947
+ "128118": {
948
+ "content": "<|reserved_special_token_113|>",
949
+ "lstrip": false,
950
+ "normalized": false,
951
+ "rstrip": false,
952
+ "single_word": false,
953
+ "special": true
954
+ },
955
+ "128119": {
956
+ "content": "<|reserved_special_token_114|>",
957
+ "lstrip": false,
958
+ "normalized": false,
959
+ "rstrip": false,
960
+ "single_word": false,
961
+ "special": true
962
+ },
963
+ "128120": {
964
+ "content": "<|reserved_special_token_115|>",
965
+ "lstrip": false,
966
+ "normalized": false,
967
+ "rstrip": false,
968
+ "single_word": false,
969
+ "special": true
970
+ },
971
+ "128121": {
972
+ "content": "<|reserved_special_token_116|>",
973
+ "lstrip": false,
974
+ "normalized": false,
975
+ "rstrip": false,
976
+ "single_word": false,
977
+ "special": true
978
+ },
979
+ "128122": {
980
+ "content": "<|reserved_special_token_117|>",
981
+ "lstrip": false,
982
+ "normalized": false,
983
+ "rstrip": false,
984
+ "single_word": false,
985
+ "special": true
986
+ },
987
+ "128123": {
988
+ "content": "<|reserved_special_token_118|>",
989
+ "lstrip": false,
990
+ "normalized": false,
991
+ "rstrip": false,
992
+ "single_word": false,
993
+ "special": true
994
+ },
995
+ "128124": {
996
+ "content": "<|reserved_special_token_119|>",
997
+ "lstrip": false,
998
+ "normalized": false,
999
+ "rstrip": false,
1000
+ "single_word": false,
1001
+ "special": true
1002
+ },
1003
+ "128125": {
1004
+ "content": "<|reserved_special_token_120|>",
1005
+ "lstrip": false,
1006
+ "normalized": false,
1007
+ "rstrip": false,
1008
+ "single_word": false,
1009
+ "special": true
1010
+ },
1011
+ "128126": {
1012
+ "content": "<|reserved_special_token_121|>",
1013
+ "lstrip": false,
1014
+ "normalized": false,
1015
+ "rstrip": false,
1016
+ "single_word": false,
1017
+ "special": true
1018
+ },
1019
+ "128127": {
1020
+ "content": "<|reserved_special_token_122|>",
1021
+ "lstrip": false,
1022
+ "normalized": false,
1023
+ "rstrip": false,
1024
+ "single_word": false,
1025
+ "special": true
1026
+ },
1027
+ "128128": {
1028
+ "content": "<|reserved_special_token_123|>",
1029
+ "lstrip": false,
1030
+ "normalized": false,
1031
+ "rstrip": false,
1032
+ "single_word": false,
1033
+ "special": true
1034
+ },
1035
+ "128129": {
1036
+ "content": "<|reserved_special_token_124|>",
1037
+ "lstrip": false,
1038
+ "normalized": false,
1039
+ "rstrip": false,
1040
+ "single_word": false,
1041
+ "special": true
1042
+ },
1043
+ "128130": {
1044
+ "content": "<|reserved_special_token_125|>",
1045
+ "lstrip": false,
1046
+ "normalized": false,
1047
+ "rstrip": false,
1048
+ "single_word": false,
1049
+ "special": true
1050
+ },
1051
+ "128131": {
1052
+ "content": "<|reserved_special_token_126|>",
1053
+ "lstrip": false,
1054
+ "normalized": false,
1055
+ "rstrip": false,
1056
+ "single_word": false,
1057
+ "special": true
1058
+ },
1059
+ "128132": {
1060
+ "content": "<|reserved_special_token_127|>",
1061
+ "lstrip": false,
1062
+ "normalized": false,
1063
+ "rstrip": false,
1064
+ "single_word": false,
1065
+ "special": true
1066
+ },
1067
+ "128133": {
1068
+ "content": "<|reserved_special_token_128|>",
1069
+ "lstrip": false,
1070
+ "normalized": false,
1071
+ "rstrip": false,
1072
+ "single_word": false,
1073
+ "special": true
1074
+ },
1075
+ "128134": {
1076
+ "content": "<|reserved_special_token_129|>",
1077
+ "lstrip": false,
1078
+ "normalized": false,
1079
+ "rstrip": false,
1080
+ "single_word": false,
1081
+ "special": true
1082
+ },
1083
+ "128135": {
1084
+ "content": "<|reserved_special_token_130|>",
1085
+ "lstrip": false,
1086
+ "normalized": false,
1087
+ "rstrip": false,
1088
+ "single_word": false,
1089
+ "special": true
1090
+ },
1091
+ "128136": {
1092
+ "content": "<|reserved_special_token_131|>",
1093
+ "lstrip": false,
1094
+ "normalized": false,
1095
+ "rstrip": false,
1096
+ "single_word": false,
1097
+ "special": true
1098
+ },
1099
+ "128137": {
1100
+ "content": "<|reserved_special_token_132|>",
1101
+ "lstrip": false,
1102
+ "normalized": false,
1103
+ "rstrip": false,
1104
+ "single_word": false,
1105
+ "special": true
1106
+ },
1107
+ "128138": {
1108
+ "content": "<|reserved_special_token_133|>",
1109
+ "lstrip": false,
1110
+ "normalized": false,
1111
+ "rstrip": false,
1112
+ "single_word": false,
1113
+ "special": true
1114
+ },
1115
+ "128139": {
1116
+ "content": "<|reserved_special_token_134|>",
1117
+ "lstrip": false,
1118
+ "normalized": false,
1119
+ "rstrip": false,
1120
+ "single_word": false,
1121
+ "special": true
1122
+ },
1123
+ "128140": {
1124
+ "content": "<|reserved_special_token_135|>",
1125
+ "lstrip": false,
1126
+ "normalized": false,
1127
+ "rstrip": false,
1128
+ "single_word": false,
1129
+ "special": true
1130
+ },
1131
+ "128141": {
1132
+ "content": "<|reserved_special_token_136|>",
1133
+ "lstrip": false,
1134
+ "normalized": false,
1135
+ "rstrip": false,
1136
+ "single_word": false,
1137
+ "special": true
1138
+ },
1139
+ "128142": {
1140
+ "content": "<|reserved_special_token_137|>",
1141
+ "lstrip": false,
1142
+ "normalized": false,
1143
+ "rstrip": false,
1144
+ "single_word": false,
1145
+ "special": true
1146
+ },
1147
+ "128143": {
1148
+ "content": "<|reserved_special_token_138|>",
1149
+ "lstrip": false,
1150
+ "normalized": false,
1151
+ "rstrip": false,
1152
+ "single_word": false,
1153
+ "special": true
1154
+ },
1155
+ "128144": {
1156
+ "content": "<|reserved_special_token_139|>",
1157
+ "lstrip": false,
1158
+ "normalized": false,
1159
+ "rstrip": false,
1160
+ "single_word": false,
1161
+ "special": true
1162
+ },
1163
+ "128145": {
1164
+ "content": "<|reserved_special_token_140|>",
1165
+ "lstrip": false,
1166
+ "normalized": false,
1167
+ "rstrip": false,
1168
+ "single_word": false,
1169
+ "special": true
1170
+ },
1171
+ "128146": {
1172
+ "content": "<|reserved_special_token_141|>",
1173
+ "lstrip": false,
1174
+ "normalized": false,
1175
+ "rstrip": false,
1176
+ "single_word": false,
1177
+ "special": true
1178
+ },
1179
+ "128147": {
1180
+ "content": "<|reserved_special_token_142|>",
1181
+ "lstrip": false,
1182
+ "normalized": false,
1183
+ "rstrip": false,
1184
+ "single_word": false,
1185
+ "special": true
1186
+ },
1187
+ "128148": {
1188
+ "content": "<|reserved_special_token_143|>",
1189
+ "lstrip": false,
1190
+ "normalized": false,
1191
+ "rstrip": false,
1192
+ "single_word": false,
1193
+ "special": true
1194
+ },
1195
+ "128149": {
1196
+ "content": "<|reserved_special_token_144|>",
1197
+ "lstrip": false,
1198
+ "normalized": false,
1199
+ "rstrip": false,
1200
+ "single_word": false,
1201
+ "special": true
1202
+ },
1203
+ "128150": {
1204
+ "content": "<|reserved_special_token_145|>",
1205
+ "lstrip": false,
1206
+ "normalized": false,
1207
+ "rstrip": false,
1208
+ "single_word": false,
1209
+ "special": true
1210
+ },
1211
+ "128151": {
1212
+ "content": "<|reserved_special_token_146|>",
1213
+ "lstrip": false,
1214
+ "normalized": false,
1215
+ "rstrip": false,
1216
+ "single_word": false,
1217
+ "special": true
1218
+ },
1219
+ "128152": {
1220
+ "content": "<|reserved_special_token_147|>",
1221
+ "lstrip": false,
1222
+ "normalized": false,
1223
+ "rstrip": false,
1224
+ "single_word": false,
1225
+ "special": true
1226
+ },
1227
+ "128153": {
1228
+ "content": "<|reserved_special_token_148|>",
1229
+ "lstrip": false,
1230
+ "normalized": false,
1231
+ "rstrip": false,
1232
+ "single_word": false,
1233
+ "special": true
1234
+ },
1235
+ "128154": {
1236
+ "content": "<|reserved_special_token_149|>",
1237
+ "lstrip": false,
1238
+ "normalized": false,
1239
+ "rstrip": false,
1240
+ "single_word": false,
1241
+ "special": true
1242
+ },
1243
+ "128155": {
1244
+ "content": "<|reserved_special_token_150|>",
1245
+ "lstrip": false,
1246
+ "normalized": false,
1247
+ "rstrip": false,
1248
+ "single_word": false,
1249
+ "special": true
1250
+ },
1251
+ "128156": {
1252
+ "content": "<|reserved_special_token_151|>",
1253
+ "lstrip": false,
1254
+ "normalized": false,
1255
+ "rstrip": false,
1256
+ "single_word": false,
1257
+ "special": true
1258
+ },
1259
+ "128157": {
1260
+ "content": "<|reserved_special_token_152|>",
1261
+ "lstrip": false,
1262
+ "normalized": false,
1263
+ "rstrip": false,
1264
+ "single_word": false,
1265
+ "special": true
1266
+ },
1267
+ "128158": {
1268
+ "content": "<|reserved_special_token_153|>",
1269
+ "lstrip": false,
1270
+ "normalized": false,
1271
+ "rstrip": false,
1272
+ "single_word": false,
1273
+ "special": true
1274
+ },
1275
+ "128159": {
1276
+ "content": "<|reserved_special_token_154|>",
1277
+ "lstrip": false,
1278
+ "normalized": false,
1279
+ "rstrip": false,
1280
+ "single_word": false,
1281
+ "special": true
1282
+ },
1283
+ "128160": {
1284
+ "content": "<|reserved_special_token_155|>",
1285
+ "lstrip": false,
1286
+ "normalized": false,
1287
+ "rstrip": false,
1288
+ "single_word": false,
1289
+ "special": true
1290
+ },
1291
+ "128161": {
1292
+ "content": "<|reserved_special_token_156|>",
1293
+ "lstrip": false,
1294
+ "normalized": false,
1295
+ "rstrip": false,
1296
+ "single_word": false,
1297
+ "special": true
1298
+ },
1299
+ "128162": {
1300
+ "content": "<|reserved_special_token_157|>",
1301
+ "lstrip": false,
1302
+ "normalized": false,
1303
+ "rstrip": false,
1304
+ "single_word": false,
1305
+ "special": true
1306
+ },
1307
+ "128163": {
1308
+ "content": "<|reserved_special_token_158|>",
1309
+ "lstrip": false,
1310
+ "normalized": false,
1311
+ "rstrip": false,
1312
+ "single_word": false,
1313
+ "special": true
1314
+ },
1315
+ "128164": {
1316
+ "content": "<|reserved_special_token_159|>",
1317
+ "lstrip": false,
1318
+ "normalized": false,
1319
+ "rstrip": false,
1320
+ "single_word": false,
1321
+ "special": true
1322
+ },
1323
+ "128165": {
1324
+ "content": "<|reserved_special_token_160|>",
1325
+ "lstrip": false,
1326
+ "normalized": false,
1327
+ "rstrip": false,
1328
+ "single_word": false,
1329
+ "special": true
1330
+ },
1331
+ "128166": {
1332
+ "content": "<|reserved_special_token_161|>",
1333
+ "lstrip": false,
1334
+ "normalized": false,
1335
+ "rstrip": false,
1336
+ "single_word": false,
1337
+ "special": true
1338
+ },
1339
+ "128167": {
1340
+ "content": "<|reserved_special_token_162|>",
1341
+ "lstrip": false,
1342
+ "normalized": false,
1343
+ "rstrip": false,
1344
+ "single_word": false,
1345
+ "special": true
1346
+ },
1347
+ "128168": {
1348
+ "content": "<|reserved_special_token_163|>",
1349
+ "lstrip": false,
1350
+ "normalized": false,
1351
+ "rstrip": false,
1352
+ "single_word": false,
1353
+ "special": true
1354
+ },
1355
+ "128169": {
1356
+ "content": "<|reserved_special_token_164|>",
1357
+ "lstrip": false,
1358
+ "normalized": false,
1359
+ "rstrip": false,
1360
+ "single_word": false,
1361
+ "special": true
1362
+ },
1363
+ "128170": {
1364
+ "content": "<|reserved_special_token_165|>",
1365
+ "lstrip": false,
1366
+ "normalized": false,
1367
+ "rstrip": false,
1368
+ "single_word": false,
1369
+ "special": true
1370
+ },
1371
+ "128171": {
1372
+ "content": "<|reserved_special_token_166|>",
1373
+ "lstrip": false,
1374
+ "normalized": false,
1375
+ "rstrip": false,
1376
+ "single_word": false,
1377
+ "special": true
1378
+ },
1379
+ "128172": {
1380
+ "content": "<|reserved_special_token_167|>",
1381
+ "lstrip": false,
1382
+ "normalized": false,
1383
+ "rstrip": false,
1384
+ "single_word": false,
1385
+ "special": true
1386
+ },
1387
+ "128173": {
1388
+ "content": "<|reserved_special_token_168|>",
1389
+ "lstrip": false,
1390
+ "normalized": false,
1391
+ "rstrip": false,
1392
+ "single_word": false,
1393
+ "special": true
1394
+ },
1395
+ "128174": {
1396
+ "content": "<|reserved_special_token_169|>",
1397
+ "lstrip": false,
1398
+ "normalized": false,
1399
+ "rstrip": false,
1400
+ "single_word": false,
1401
+ "special": true
1402
+ },
1403
+ "128175": {
1404
+ "content": "<|reserved_special_token_170|>",
1405
+ "lstrip": false,
1406
+ "normalized": false,
1407
+ "rstrip": false,
1408
+ "single_word": false,
1409
+ "special": true
1410
+ },
1411
+ "128176": {
1412
+ "content": "<|reserved_special_token_171|>",
1413
+ "lstrip": false,
1414
+ "normalized": false,
1415
+ "rstrip": false,
1416
+ "single_word": false,
1417
+ "special": true
1418
+ },
1419
+ "128177": {
1420
+ "content": "<|reserved_special_token_172|>",
1421
+ "lstrip": false,
1422
+ "normalized": false,
1423
+ "rstrip": false,
1424
+ "single_word": false,
1425
+ "special": true
1426
+ },
1427
+ "128178": {
1428
+ "content": "<|reserved_special_token_173|>",
1429
+ "lstrip": false,
1430
+ "normalized": false,
1431
+ "rstrip": false,
1432
+ "single_word": false,
1433
+ "special": true
1434
+ },
1435
+ "128179": {
1436
+ "content": "<|reserved_special_token_174|>",
1437
+ "lstrip": false,
1438
+ "normalized": false,
1439
+ "rstrip": false,
1440
+ "single_word": false,
1441
+ "special": true
1442
+ },
1443
+ "128180": {
1444
+ "content": "<|reserved_special_token_175|>",
1445
+ "lstrip": false,
1446
+ "normalized": false,
1447
+ "rstrip": false,
1448
+ "single_word": false,
1449
+ "special": true
1450
+ },
1451
+ "128181": {
1452
+ "content": "<|reserved_special_token_176|>",
1453
+ "lstrip": false,
1454
+ "normalized": false,
1455
+ "rstrip": false,
1456
+ "single_word": false,
1457
+ "special": true
1458
+ },
1459
+ "128182": {
1460
+ "content": "<|reserved_special_token_177|>",
1461
+ "lstrip": false,
1462
+ "normalized": false,
1463
+ "rstrip": false,
1464
+ "single_word": false,
1465
+ "special": true
1466
+ },
1467
+ "128183": {
1468
+ "content": "<|reserved_special_token_178|>",
1469
+ "lstrip": false,
1470
+ "normalized": false,
1471
+ "rstrip": false,
1472
+ "single_word": false,
1473
+ "special": true
1474
+ },
1475
+ "128184": {
1476
+ "content": "<|reserved_special_token_179|>",
1477
+ "lstrip": false,
1478
+ "normalized": false,
1479
+ "rstrip": false,
1480
+ "single_word": false,
1481
+ "special": true
1482
+ },
1483
+ "128185": {
1484
+ "content": "<|reserved_special_token_180|>",
1485
+ "lstrip": false,
1486
+ "normalized": false,
1487
+ "rstrip": false,
1488
+ "single_word": false,
1489
+ "special": true
1490
+ },
1491
+ "128186": {
1492
+ "content": "<|reserved_special_token_181|>",
1493
+ "lstrip": false,
1494
+ "normalized": false,
1495
+ "rstrip": false,
1496
+ "single_word": false,
1497
+ "special": true
1498
+ },
1499
+ "128187": {
1500
+ "content": "<|reserved_special_token_182|>",
1501
+ "lstrip": false,
1502
+ "normalized": false,
1503
+ "rstrip": false,
1504
+ "single_word": false,
1505
+ "special": true
1506
+ },
1507
+ "128188": {
1508
+ "content": "<|reserved_special_token_183|>",
1509
+ "lstrip": false,
1510
+ "normalized": false,
1511
+ "rstrip": false,
1512
+ "single_word": false,
1513
+ "special": true
1514
+ },
1515
+ "128189": {
1516
+ "content": "<|reserved_special_token_184|>",
1517
+ "lstrip": false,
1518
+ "normalized": false,
1519
+ "rstrip": false,
1520
+ "single_word": false,
1521
+ "special": true
1522
+ },
1523
+ "128190": {
1524
+ "content": "<|reserved_special_token_185|>",
1525
+ "lstrip": false,
1526
+ "normalized": false,
1527
+ "rstrip": false,
1528
+ "single_word": false,
1529
+ "special": true
1530
+ },
1531
+ "128191": {
1532
+ "content": "<|reserved_special_token_186|>",
1533
+ "lstrip": false,
1534
+ "normalized": false,
1535
+ "rstrip": false,
1536
+ "single_word": false,
1537
+ "special": true
1538
+ },
1539
+ "128192": {
1540
+ "content": "<|reserved_special_token_187|>",
1541
+ "lstrip": false,
1542
+ "normalized": false,
1543
+ "rstrip": false,
1544
+ "single_word": false,
1545
+ "special": true
1546
+ },
1547
+ "128193": {
1548
+ "content": "<|reserved_special_token_188|>",
1549
+ "lstrip": false,
1550
+ "normalized": false,
1551
+ "rstrip": false,
1552
+ "single_word": false,
1553
+ "special": true
1554
+ },
1555
+ "128194": {
1556
+ "content": "<|reserved_special_token_189|>",
1557
+ "lstrip": false,
1558
+ "normalized": false,
1559
+ "rstrip": false,
1560
+ "single_word": false,
1561
+ "special": true
1562
+ },
1563
+ "128195": {
1564
+ "content": "<|reserved_special_token_190|>",
1565
+ "lstrip": false,
1566
+ "normalized": false,
1567
+ "rstrip": false,
1568
+ "single_word": false,
1569
+ "special": true
1570
+ },
1571
+ "128196": {
1572
+ "content": "<|reserved_special_token_191|>",
1573
+ "lstrip": false,
1574
+ "normalized": false,
1575
+ "rstrip": false,
1576
+ "single_word": false,
1577
+ "special": true
1578
+ },
1579
+ "128197": {
1580
+ "content": "<|reserved_special_token_192|>",
1581
+ "lstrip": false,
1582
+ "normalized": false,
1583
+ "rstrip": false,
1584
+ "single_word": false,
1585
+ "special": true
1586
+ },
1587
+ "128198": {
1588
+ "content": "<|reserved_special_token_193|>",
1589
+ "lstrip": false,
1590
+ "normalized": false,
1591
+ "rstrip": false,
1592
+ "single_word": false,
1593
+ "special": true
1594
+ },
1595
+ "128199": {
1596
+ "content": "<|reserved_special_token_194|>",
1597
+ "lstrip": false,
1598
+ "normalized": false,
1599
+ "rstrip": false,
1600
+ "single_word": false,
1601
+ "special": true
1602
+ },
1603
+ "128200": {
1604
+ "content": "<|reserved_special_token_195|>",
1605
+ "lstrip": false,
1606
+ "normalized": false,
1607
+ "rstrip": false,
1608
+ "single_word": false,
1609
+ "special": true
1610
+ },
1611
+ "128201": {
1612
+ "content": "<|reserved_special_token_196|>",
1613
+ "lstrip": false,
1614
+ "normalized": false,
1615
+ "rstrip": false,
1616
+ "single_word": false,
1617
+ "special": true
1618
+ },
1619
+ "128202": {
1620
+ "content": "<|reserved_special_token_197|>",
1621
+ "lstrip": false,
1622
+ "normalized": false,
1623
+ "rstrip": false,
1624
+ "single_word": false,
1625
+ "special": true
1626
+ },
1627
+ "128203": {
1628
+ "content": "<|reserved_special_token_198|>",
1629
+ "lstrip": false,
1630
+ "normalized": false,
1631
+ "rstrip": false,
1632
+ "single_word": false,
1633
+ "special": true
1634
+ },
1635
+ "128204": {
1636
+ "content": "<|reserved_special_token_199|>",
1637
+ "lstrip": false,
1638
+ "normalized": false,
1639
+ "rstrip": false,
1640
+ "single_word": false,
1641
+ "special": true
1642
+ },
1643
+ "128205": {
1644
+ "content": "<|reserved_special_token_200|>",
1645
+ "lstrip": false,
1646
+ "normalized": false,
1647
+ "rstrip": false,
1648
+ "single_word": false,
1649
+ "special": true
1650
+ },
1651
+ "128206": {
1652
+ "content": "<|reserved_special_token_201|>",
1653
+ "lstrip": false,
1654
+ "normalized": false,
1655
+ "rstrip": false,
1656
+ "single_word": false,
1657
+ "special": true
1658
+ },
1659
+ "128207": {
1660
+ "content": "<|reserved_special_token_202|>",
1661
+ "lstrip": false,
1662
+ "normalized": false,
1663
+ "rstrip": false,
1664
+ "single_word": false,
1665
+ "special": true
1666
+ },
1667
+ "128208": {
1668
+ "content": "<|reserved_special_token_203|>",
1669
+ "lstrip": false,
1670
+ "normalized": false,
1671
+ "rstrip": false,
1672
+ "single_word": false,
1673
+ "special": true
1674
+ },
1675
+ "128209": {
1676
+ "content": "<|reserved_special_token_204|>",
1677
+ "lstrip": false,
1678
+ "normalized": false,
1679
+ "rstrip": false,
1680
+ "single_word": false,
1681
+ "special": true
1682
+ },
1683
+ "128210": {
1684
+ "content": "<|reserved_special_token_205|>",
1685
+ "lstrip": false,
1686
+ "normalized": false,
1687
+ "rstrip": false,
1688
+ "single_word": false,
1689
+ "special": true
1690
+ },
1691
+ "128211": {
1692
+ "content": "<|reserved_special_token_206|>",
1693
+ "lstrip": false,
1694
+ "normalized": false,
1695
+ "rstrip": false,
1696
+ "single_word": false,
1697
+ "special": true
1698
+ },
1699
+ "128212": {
1700
+ "content": "<|reserved_special_token_207|>",
1701
+ "lstrip": false,
1702
+ "normalized": false,
1703
+ "rstrip": false,
1704
+ "single_word": false,
1705
+ "special": true
1706
+ },
1707
+ "128213": {
1708
+ "content": "<|reserved_special_token_208|>",
1709
+ "lstrip": false,
1710
+ "normalized": false,
1711
+ "rstrip": false,
1712
+ "single_word": false,
1713
+ "special": true
1714
+ },
1715
+ "128214": {
1716
+ "content": "<|reserved_special_token_209|>",
1717
+ "lstrip": false,
1718
+ "normalized": false,
1719
+ "rstrip": false,
1720
+ "single_word": false,
1721
+ "special": true
1722
+ },
1723
+ "128215": {
1724
+ "content": "<|reserved_special_token_210|>",
1725
+ "lstrip": false,
1726
+ "normalized": false,
1727
+ "rstrip": false,
1728
+ "single_word": false,
1729
+ "special": true
1730
+ },
1731
+ "128216": {
1732
+ "content": "<|reserved_special_token_211|>",
1733
+ "lstrip": false,
1734
+ "normalized": false,
1735
+ "rstrip": false,
1736
+ "single_word": false,
1737
+ "special": true
1738
+ },
1739
+ "128217": {
1740
+ "content": "<|reserved_special_token_212|>",
1741
+ "lstrip": false,
1742
+ "normalized": false,
1743
+ "rstrip": false,
1744
+ "single_word": false,
1745
+ "special": true
1746
+ },
1747
+ "128218": {
1748
+ "content": "<|reserved_special_token_213|>",
1749
+ "lstrip": false,
1750
+ "normalized": false,
1751
+ "rstrip": false,
1752
+ "single_word": false,
1753
+ "special": true
1754
+ },
1755
+ "128219": {
1756
+ "content": "<|reserved_special_token_214|>",
1757
+ "lstrip": false,
1758
+ "normalized": false,
1759
+ "rstrip": false,
1760
+ "single_word": false,
1761
+ "special": true
1762
+ },
1763
+ "128220": {
1764
+ "content": "<|reserved_special_token_215|>",
1765
+ "lstrip": false,
1766
+ "normalized": false,
1767
+ "rstrip": false,
1768
+ "single_word": false,
1769
+ "special": true
1770
+ },
1771
+ "128221": {
1772
+ "content": "<|reserved_special_token_216|>",
1773
+ "lstrip": false,
1774
+ "normalized": false,
1775
+ "rstrip": false,
1776
+ "single_word": false,
1777
+ "special": true
1778
+ },
1779
+ "128222": {
1780
+ "content": "<|reserved_special_token_217|>",
1781
+ "lstrip": false,
1782
+ "normalized": false,
1783
+ "rstrip": false,
1784
+ "single_word": false,
1785
+ "special": true
1786
+ },
1787
+ "128223": {
1788
+ "content": "<|reserved_special_token_218|>",
1789
+ "lstrip": false,
1790
+ "normalized": false,
1791
+ "rstrip": false,
1792
+ "single_word": false,
1793
+ "special": true
1794
+ },
1795
+ "128224": {
1796
+ "content": "<|reserved_special_token_219|>",
1797
+ "lstrip": false,
1798
+ "normalized": false,
1799
+ "rstrip": false,
1800
+ "single_word": false,
1801
+ "special": true
1802
+ },
1803
+ "128225": {
1804
+ "content": "<|reserved_special_token_220|>",
1805
+ "lstrip": false,
1806
+ "normalized": false,
1807
+ "rstrip": false,
1808
+ "single_word": false,
1809
+ "special": true
1810
+ },
1811
+ "128226": {
1812
+ "content": "<|reserved_special_token_221|>",
1813
+ "lstrip": false,
1814
+ "normalized": false,
1815
+ "rstrip": false,
1816
+ "single_word": false,
1817
+ "special": true
1818
+ },
1819
+ "128227": {
1820
+ "content": "<|reserved_special_token_222|>",
1821
+ "lstrip": false,
1822
+ "normalized": false,
1823
+ "rstrip": false,
1824
+ "single_word": false,
1825
+ "special": true
1826
+ },
1827
+ "128228": {
1828
+ "content": "<|reserved_special_token_223|>",
1829
+ "lstrip": false,
1830
+ "normalized": false,
1831
+ "rstrip": false,
1832
+ "single_word": false,
1833
+ "special": true
1834
+ },
1835
+ "128229": {
1836
+ "content": "<|reserved_special_token_224|>",
1837
+ "lstrip": false,
1838
+ "normalized": false,
1839
+ "rstrip": false,
1840
+ "single_word": false,
1841
+ "special": true
1842
+ },
1843
+ "128230": {
1844
+ "content": "<|reserved_special_token_225|>",
1845
+ "lstrip": false,
1846
+ "normalized": false,
1847
+ "rstrip": false,
1848
+ "single_word": false,
1849
+ "special": true
1850
+ },
1851
+ "128231": {
1852
+ "content": "<|reserved_special_token_226|>",
1853
+ "lstrip": false,
1854
+ "normalized": false,
1855
+ "rstrip": false,
1856
+ "single_word": false,
1857
+ "special": true
1858
+ },
1859
+ "128232": {
1860
+ "content": "<|reserved_special_token_227|>",
1861
+ "lstrip": false,
1862
+ "normalized": false,
1863
+ "rstrip": false,
1864
+ "single_word": false,
1865
+ "special": true
1866
+ },
1867
+ "128233": {
1868
+ "content": "<|reserved_special_token_228|>",
1869
+ "lstrip": false,
1870
+ "normalized": false,
1871
+ "rstrip": false,
1872
+ "single_word": false,
1873
+ "special": true
1874
+ },
1875
+ "128234": {
1876
+ "content": "<|reserved_special_token_229|>",
1877
+ "lstrip": false,
1878
+ "normalized": false,
1879
+ "rstrip": false,
1880
+ "single_word": false,
1881
+ "special": true
1882
+ },
1883
+ "128235": {
1884
+ "content": "<|reserved_special_token_230|>",
1885
+ "lstrip": false,
1886
+ "normalized": false,
1887
+ "rstrip": false,
1888
+ "single_word": false,
1889
+ "special": true
1890
+ },
1891
+ "128236": {
1892
+ "content": "<|reserved_special_token_231|>",
1893
+ "lstrip": false,
1894
+ "normalized": false,
1895
+ "rstrip": false,
1896
+ "single_word": false,
1897
+ "special": true
1898
+ },
1899
+ "128237": {
1900
+ "content": "<|reserved_special_token_232|>",
1901
+ "lstrip": false,
1902
+ "normalized": false,
1903
+ "rstrip": false,
1904
+ "single_word": false,
1905
+ "special": true
1906
+ },
1907
+ "128238": {
1908
+ "content": "<|reserved_special_token_233|>",
1909
+ "lstrip": false,
1910
+ "normalized": false,
1911
+ "rstrip": false,
1912
+ "single_word": false,
1913
+ "special": true
1914
+ },
1915
+ "128239": {
1916
+ "content": "<|reserved_special_token_234|>",
1917
+ "lstrip": false,
1918
+ "normalized": false,
1919
+ "rstrip": false,
1920
+ "single_word": false,
1921
+ "special": true
1922
+ },
1923
+ "128240": {
1924
+ "content": "<|reserved_special_token_235|>",
1925
+ "lstrip": false,
1926
+ "normalized": false,
1927
+ "rstrip": false,
1928
+ "single_word": false,
1929
+ "special": true
1930
+ },
1931
+ "128241": {
1932
+ "content": "<|reserved_special_token_236|>",
1933
+ "lstrip": false,
1934
+ "normalized": false,
1935
+ "rstrip": false,
1936
+ "single_word": false,
1937
+ "special": true
1938
+ },
1939
+ "128242": {
1940
+ "content": "<|reserved_special_token_237|>",
1941
+ "lstrip": false,
1942
+ "normalized": false,
1943
+ "rstrip": false,
1944
+ "single_word": false,
1945
+ "special": true
1946
+ },
1947
+ "128243": {
1948
+ "content": "<|reserved_special_token_238|>",
1949
+ "lstrip": false,
1950
+ "normalized": false,
1951
+ "rstrip": false,
1952
+ "single_word": false,
1953
+ "special": true
1954
+ },
1955
+ "128244": {
1956
+ "content": "<|reserved_special_token_239|>",
1957
+ "lstrip": false,
1958
+ "normalized": false,
1959
+ "rstrip": false,
1960
+ "single_word": false,
1961
+ "special": true
1962
+ },
1963
+ "128245": {
1964
+ "content": "<|reserved_special_token_240|>",
1965
+ "lstrip": false,
1966
+ "normalized": false,
1967
+ "rstrip": false,
1968
+ "single_word": false,
1969
+ "special": true
1970
+ },
1971
+ "128246": {
1972
+ "content": "<|reserved_special_token_241|>",
1973
+ "lstrip": false,
1974
+ "normalized": false,
1975
+ "rstrip": false,
1976
+ "single_word": false,
1977
+ "special": true
1978
+ },
1979
+ "128247": {
1980
+ "content": "<|reserved_special_token_242|>",
1981
+ "lstrip": false,
1982
+ "normalized": false,
1983
+ "rstrip": false,
1984
+ "single_word": false,
1985
+ "special": true
1986
+ },
1987
+ "128248": {
1988
+ "content": "<|reserved_special_token_243|>",
1989
+ "lstrip": false,
1990
+ "normalized": false,
1991
+ "rstrip": false,
1992
+ "single_word": false,
1993
+ "special": true
1994
+ },
1995
+ "128249": {
1996
+ "content": "<|reserved_special_token_244|>",
1997
+ "lstrip": false,
1998
+ "normalized": false,
1999
+ "rstrip": false,
2000
+ "single_word": false,
2001
+ "special": true
2002
+ },
2003
+ "128250": {
2004
+ "content": "<|reserved_special_token_245|>",
2005
+ "lstrip": false,
2006
+ "normalized": false,
2007
+ "rstrip": false,
2008
+ "single_word": false,
2009
+ "special": true
2010
+ },
2011
+ "128251": {
2012
+ "content": "<|reserved_special_token_246|>",
2013
+ "lstrip": false,
2014
+ "normalized": false,
2015
+ "rstrip": false,
2016
+ "single_word": false,
2017
+ "special": true
2018
+ },
2019
+ "128252": {
2020
+ "content": "<|reserved_special_token_247|>",
2021
+ "lstrip": false,
2022
+ "normalized": false,
2023
+ "rstrip": false,
2024
+ "single_word": false,
2025
+ "special": true
2026
+ },
2027
+ "128253": {
2028
+ "content": "<|reserved_special_token_248|>",
2029
+ "lstrip": false,
2030
+ "normalized": false,
2031
+ "rstrip": false,
2032
+ "single_word": false,
2033
+ "special": true
2034
+ },
2035
+ "128254": {
2036
+ "content": "<|reserved_special_token_249|>",
2037
+ "lstrip": false,
2038
+ "normalized": false,
2039
+ "rstrip": false,
2040
+ "single_word": false,
2041
+ "special": true
2042
+ },
2043
+ "128255": {
2044
+ "content": "<|reserved_special_token_250|>",
2045
+ "lstrip": false,
2046
+ "normalized": false,
2047
+ "rstrip": false,
2048
+ "single_word": false,
2049
+ "special": true
2050
+ }
2051
+ },
2052
+ "bos_token": "<|begin_of_text|>",
2053
+ "chat_template": "{% set loop_messages = messages %}{% for message in loop_messages %}{% set content = '<|start_header_id|>' + message['role'] + '<|end_header_id|>\n\n'+ message['content'] | trim + '<|eot_id|>' %}{% if loop.index0 == 0 %}{% set content = bos_token + content %}{% endif %}{{ content }}{% endfor %}{% if add_generation_prompt %}{{ '<|start_header_id|>assistant<|end_header_id|>\n\n' }}{% endif %}",
2054
+ "clean_up_tokenization_spaces": true,
2055
+ "eos_token": "<|eot_id|>",
2056
+ "model_input_names": [
2057
+ "input_ids",
2058
+ "attention_mask"
2059
+ ],
2060
+ "model_max_length": 1000000000000000019884624838656,
2061
+ "pad_token": "<|eot_id|>",
2062
+ "tokenizer_class": "PreTrainedTokenizerFast"
2063
+ }
whisper_model_modified.py ADDED
@@ -0,0 +1,144 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ####################################################################
2
+ ## Built on top of Ultravox: https://github.com/fixie-ai/ultravox ##
3
+ ####################################################################
4
+ # modified from https://github.com/huggingface/transformers/blob/main/src/transformers/models/whisper/modeling_whisper.py
5
+ # see this issue for the commentary: https://github.com/huggingface/transformers/issues/25744
6
+ #
7
+ # Copyright 2022 The OpenAI Authors and The HuggingFace Inc. team. All rights reserved.
8
+ #
9
+ # Licensed under the Apache License, Version 2.0 (the "License");
10
+ # you may not use this file except in compliance with the License.
11
+ # You may obtain a copy of the License at
12
+ #
13
+ # http://www.apache.org/licenses/LICENSE-2.0
14
+ #
15
+ # Unless required by applicable law or agreed to in writing, software
16
+ # distributed under the License is distributed on an "AS IS" BASIS,
17
+ # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
18
+ # See the License for the specific language governing permissions and
19
+ # limitations under the License.
20
+ import torch
21
+ import torch.nn as nn
22
+ import transformers
23
+ import transformers.modeling_outputs
24
+ from transformers.models.whisper import modeling_whisper as whisper
25
+
26
+
27
+ class WhisperEncoder(whisper.WhisperEncoder):
28
+ """
29
+ Encoder portion of OpenAI's Whisper model.
30
+
31
+ This implementation is a slightly modified version of HF Transformers' Whisper Encoder, with only a few fixes:
32
+ 1. base_model_prefix updated to allow for doing `.from_pretrained` directly on the encoder
33
+ 2. allow less than 30 second of audio padding to be passed in:
34
+ - relaxed ValueError check for `input_features` length to be less than or equal to `expected_seq_length` instead of strictly equal
35
+ - embed_pos is now sliced to match the length of `inputs_embeds`
36
+
37
+ Original: https://github.com/huggingface/transformers/blob/main/src/transformers/models/whisper/modeling_whisper.py
38
+ """
39
+
40
+ base_model_prefix = "model.encoder"
41
+
42
+ def forward(
43
+ self,
44
+ input_features,
45
+ attention_mask=None,
46
+ head_mask=None,
47
+ output_attentions=None,
48
+ output_hidden_states=None,
49
+ return_dict=None,
50
+ ):
51
+ expected_seq_length = (
52
+ self.config.max_source_positions
53
+ * self.conv1.stride[0]
54
+ * self.conv2.stride[0]
55
+ )
56
+ if input_features.shape[-1] > expected_seq_length:
57
+ raise ValueError(
58
+ f"Whisper expects the mel input features to be of length {expected_seq_length} or less, but found {input_features.shape[-1]}. Make sure to pad the input mel features to {expected_seq_length}."
59
+ )
60
+
61
+ output_attentions = (
62
+ output_attentions
63
+ if output_attentions is not None
64
+ else self.config.output_attentions
65
+ )
66
+ output_hidden_states = (
67
+ output_hidden_states
68
+ if output_hidden_states is not None
69
+ else self.config.output_hidden_states
70
+ )
71
+ return_dict = (
72
+ return_dict if return_dict is not None else self.config.use_return_dict
73
+ )
74
+ inputs_embeds = nn.functional.gelu(self.conv1(input_features))
75
+ inputs_embeds = nn.functional.gelu(self.conv2(inputs_embeds))
76
+
77
+ inputs_embeds = inputs_embeds.permute(0, 2, 1)
78
+ embed_pos = self.embed_positions.weight[: inputs_embeds.size(-2)]
79
+
80
+ hidden_states = inputs_embeds + embed_pos
81
+ hidden_states = nn.functional.dropout(
82
+ hidden_states, p=self.dropout, training=self.training
83
+ )
84
+
85
+ encoder_states = () if output_hidden_states else None
86
+ all_attentions = () if output_attentions else None
87
+
88
+ # check if head_mask has a correct number of layers specified if desired
89
+ if head_mask is not None:
90
+ assert head_mask.size()[0] == (
91
+ len(self.layers)
92
+ ), f"The head_mask should be specified for {len(self.layers)} layers, but it is for {head_mask.size()[0]}."
93
+
94
+ for idx, encoder_layer in enumerate(self.layers):
95
+ if output_hidden_states:
96
+ encoder_states = encoder_states + (hidden_states,)
97
+ # add LayerDrop (see https://arxiv.org/abs/1909.11556 for description)
98
+ to_drop = False
99
+ if self.training:
100
+ dropout_probability = torch.rand([])
101
+ if dropout_probability < self.layerdrop: # skip the layer
102
+ to_drop = True
103
+
104
+ if to_drop:
105
+ layer_outputs = (None, None)
106
+ else:
107
+ if self.gradient_checkpointing and self.training:
108
+ layer_outputs = self._gradient_checkpointing_func(
109
+ encoder_layer.__call__,
110
+ hidden_states,
111
+ None,
112
+ (head_mask[idx] if head_mask is not None else None),
113
+ output_attentions,
114
+ )
115
+ else:
116
+ layer_outputs = encoder_layer(
117
+ hidden_states,
118
+ None,
119
+ layer_head_mask=(
120
+ head_mask[idx] if head_mask is not None else None
121
+ ),
122
+ output_attentions=output_attentions,
123
+ )
124
+
125
+ hidden_states = layer_outputs[0]
126
+
127
+ if output_attentions:
128
+ all_attentions = all_attentions + (layer_outputs[1],)
129
+
130
+ hidden_states = self.layer_norm(hidden_states)
131
+ if output_hidden_states:
132
+ encoder_states = encoder_states + (hidden_states,)
133
+
134
+ if not return_dict:
135
+ return tuple(
136
+ v
137
+ for v in [hidden_states, encoder_states, all_attentions]
138
+ if v is not None
139
+ )
140
+ return transformers.modeling_outputs.BaseModelOutput(
141
+ last_hidden_state=hidden_states,
142
+ hidden_states=encoder_states,
143
+ attentions=all_attentions,
144
+ )