Upload recastmlp_llama/config.json with huggingface_hub
Browse files- recastmlp_llama/config.json +97 -0
recastmlp_llama/config.json
ADDED
@@ -0,0 +1,97 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"vocab_size": 128256,
|
3 |
+
"max_position_embeddings": 131072,
|
4 |
+
"hidden_size": 4096,
|
5 |
+
"intermediate_size": 14336,
|
6 |
+
"num_hidden_layers": 32,
|
7 |
+
"num_attention_heads": 32,
|
8 |
+
"num_key_value_heads": 8,
|
9 |
+
"hidden_act": "silu",
|
10 |
+
"initializer_range": 0.02,
|
11 |
+
"rms_norm_eps": 1e-05,
|
12 |
+
"pretraining_tp": 1,
|
13 |
+
"use_cache": true,
|
14 |
+
"mlp_bias": false,
|
15 |
+
"attention_bias": false,
|
16 |
+
"attention_dropout": 0.0,
|
17 |
+
"rope_theta": 500000.0,
|
18 |
+
"rope_scaling": {
|
19 |
+
"factor": 8.0,
|
20 |
+
"low_freq_factor": 1.0,
|
21 |
+
"high_freq_factor": 4.0,
|
22 |
+
"original_max_position_embeddings": 8192,
|
23 |
+
"rope_type": "llama3"
|
24 |
+
},
|
25 |
+
"torch_dtype": null,
|
26 |
+
"num_templates": 4,
|
27 |
+
"num_groups": 8,
|
28 |
+
"num_cf": 1,
|
29 |
+
"return_dict": true,
|
30 |
+
"output_hidden_states": false,
|
31 |
+
"output_attentions": false,
|
32 |
+
"torchscript": false,
|
33 |
+
"use_bfloat16": false,
|
34 |
+
"tf_legacy_loss": false,
|
35 |
+
"pruned_heads": {},
|
36 |
+
"tie_word_embeddings": false,
|
37 |
+
"chunk_size_feed_forward": 0,
|
38 |
+
"is_encoder_decoder": false,
|
39 |
+
"is_decoder": false,
|
40 |
+
"cross_attention_hidden_size": null,
|
41 |
+
"add_cross_attention": false,
|
42 |
+
"tie_encoder_decoder": false,
|
43 |
+
"max_length": 20,
|
44 |
+
"min_length": 0,
|
45 |
+
"do_sample": false,
|
46 |
+
"early_stopping": false,
|
47 |
+
"num_beams": 1,
|
48 |
+
"num_beam_groups": 1,
|
49 |
+
"diversity_penalty": 0.0,
|
50 |
+
"temperature": 1.0,
|
51 |
+
"top_k": 50,
|
52 |
+
"top_p": 1.0,
|
53 |
+
"typical_p": 1.0,
|
54 |
+
"repetition_penalty": 1.0,
|
55 |
+
"length_penalty": 1.0,
|
56 |
+
"no_repeat_ngram_size": 0,
|
57 |
+
"encoder_no_repeat_ngram_size": 0,
|
58 |
+
"bad_words_ids": null,
|
59 |
+
"num_return_sequences": 1,
|
60 |
+
"output_scores": false,
|
61 |
+
"return_dict_in_generate": false,
|
62 |
+
"forced_bos_token_id": null,
|
63 |
+
"forced_eos_token_id": null,
|
64 |
+
"remove_invalid_values": false,
|
65 |
+
"exponential_decay_length_penalty": null,
|
66 |
+
"suppress_tokens": null,
|
67 |
+
"begin_suppress_tokens": null,
|
68 |
+
"architectures": [
|
69 |
+
"RECASTMLP_LlamaForCausalLM"
|
70 |
+
],
|
71 |
+
"finetuning_task": null,
|
72 |
+
"id2label": {
|
73 |
+
"0": "LABEL_0",
|
74 |
+
"1": "LABEL_1"
|
75 |
+
},
|
76 |
+
"label2id": {
|
77 |
+
"LABEL_0": 0,
|
78 |
+
"LABEL_1": 1
|
79 |
+
},
|
80 |
+
"tokenizer_class": null,
|
81 |
+
"prefix": null,
|
82 |
+
"bos_token_id": 128000,
|
83 |
+
"pad_token_id": null,
|
84 |
+
"eos_token_id": 128001,
|
85 |
+
"sep_token_id": null,
|
86 |
+
"decoder_start_token_id": null,
|
87 |
+
"task_specific_params": null,
|
88 |
+
"problem_type": null,
|
89 |
+
"_name_or_path": "",
|
90 |
+
"transformers_version": "4.36.0",
|
91 |
+
"model_type": "recastmlp_llama",
|
92 |
+
"auto_map": {
|
93 |
+
"AutoConfig": "configuration_recastmlp_llama.RECASTMLP_llama",
|
94 |
+
"AutoModel": "modeling_recastmlp_llama.RECASTMLP_llamaModel",
|
95 |
+
"AutoModelForCausalLM": "modeling_recastmlp_llama.RECASTMLP_LlamaForCausalLM"
|
96 |
+
}
|
97 |
+
}
|