Hugging Face's logo

bbytxt
/
9c8daef8-4a26-4744-a09b-cadb7d64e02c

PEFT
Safetensors
bloom
axolotl
Generated from Trainer
Model card Files Community
bbytxt commited on
Commit
dda7ed7
•
1 Parent(s): 3b9b147

End of training

Browse files
Files changed (3) hide show
  1. README.md +155 -0
  2. adapter_model.bin +3 -0
  3. adapter_model.safetensors +1 -1
README.md ADDED
@@ -0,0 +1,155 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: peft
3
+ license: bigscience-bloom-rail-1.0
4
+ base_model: bigscience/bloomz-560m
5
+ tags:
6
+ - axolotl
7
+ - generated_from_trainer
8
+ model-index:
9
+ - name: 9c8daef8-4a26-4744-a09b-cadb7d64e02c
10
+ results: []
11
+ ---
12
+
13
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
14
+ should probably proofread and complete it, then remove this comment. -->
15
+
16
+ [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl)
17
+ <details><summary>See axolotl config</summary>
18
+
19
+ axolotl version: `0.4.1`
20
+ ```yaml
21
+ adapter: lora
22
+ base_model: bigscience/bloomz-560m
23
+ bf16: true
24
+ chat_template: llama3
25
+ dataset_prepared_path: null
26
+ datasets:
27
+ - data_files:
28
+ - f5914e48d1786bba_train_data.json
29
+ ds_type: json
30
+ format: custom
31
+ path: /workspace/input_data/f5914e48d1786bba_train_data.json
32
+ type:
33
+ field_input: new-context
34
+ field_instruction: new-instruction
35
+ field_output: new-response
36
+ format: '{instruction} {input}'
37
+ no_input_format: '{instruction}'
38
+ system_format: '{system}'
39
+ system_prompt: ''
40
+ debug: null
41
+ deepspeed: null
42
+ device_map: auto
43
+ early_stopping_patience: 5
44
+ eval_max_new_tokens: 128
45
+ eval_steps: 50
46
+ eval_table_size: null
47
+ evals_per_epoch: null
48
+ flash_attention: false
49
+ fp16: false
50
+ fsdp: null
51
+ fsdp_config: null
52
+ gradient_accumulation_steps: 16
53
+ gradient_checkpointing: true
54
+ group_by_length: false
55
+ hub_model_id: bbytxt/9c8daef8-4a26-4744-a09b-cadb7d64e02c
56
+ hub_repo: null
57
+ hub_strategy: checkpoint
58
+ hub_token: null
59
+ learning_rate: 0.0001
60
+ load_in_4bit: false
61
+ load_in_8bit: false
62
+ local_rank: null
63
+ logging_steps: 1
64
+ lora_alpha: 128
65
+ lora_dropout: 0.05
66
+ lora_fan_in_fan_out: null
67
+ lora_model_dir: null
68
+ lora_r: 64
69
+ lora_target_linear: true
70
+ lr_scheduler: cosine
71
+ max_memory:
72
+ 0: 70GB
73
+ max_steps: 75
74
+ micro_batch_size: 4
75
+ mlflow_experiment_name: /tmp/f5914e48d1786bba_train_data.json
76
+ model_type: AutoModelForCausalLM
77
+ num_epochs: 3
78
+ optimizer: adamw_bnb_8bit
79
+ output_dir: miner_id_24
80
+ pad_to_sequence_len: true
81
+ resume_from_checkpoint: null
82
+ s2_attention: null
83
+ sample_packing: false
84
+ save_steps: 50
85
+ saves_per_epoch: null
86
+ sequence_len: 2048
87
+ strict: false
88
+ tf32: false
89
+ tokenizer_type: AutoTokenizer
90
+ train_on_inputs: false
91
+ trust_remote_code: true
92
+ val_set_size: 0.05
93
+ wandb_entity: null
94
+ wandb_mode: online
95
+ wandb_name: 9c8daef8-4a26-4744-a09b-cadb7d64e02c
96
+ wandb_project: Gradients-On-Demand
97
+ wandb_run: your_name
98
+ wandb_runid: 9c8daef8-4a26-4744-a09b-cadb7d64e02c
99
+ warmup_steps: 20
100
+ weight_decay: 0.0
101
+ xformers_attention: null
102
+
103
+ ```
104
+
105
+ </details><br>
106
+
107
+ # 9c8daef8-4a26-4744-a09b-cadb7d64e02c
108
+
109
+ This model is a fine-tuned version of [bigscience/bloomz-560m](https://huggingface.co/bigscience/bloomz-560m) on the None dataset.
110
+ It achieves the following results on the evaluation set:
111
+ - Loss: 1.8455
112
+
113
+ ## Model description
114
+
115
+ More information needed
116
+
117
+ ## Intended uses & limitations
118
+
119
+ More information needed
120
+
121
+ ## Training and evaluation data
122
+
123
+ More information needed
124
+
125
+ ## Training procedure
126
+
127
+ ### Training hyperparameters
128
+
129
+ The following hyperparameters were used during training:
130
+ - learning_rate: 0.0001
131
+ - train_batch_size: 4
132
+ - eval_batch_size: 4
133
+ - seed: 42
134
+ - gradient_accumulation_steps: 16
135
+ - total_train_batch_size: 64
136
+ - optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
137
+ - lr_scheduler_type: cosine
138
+ - lr_scheduler_warmup_steps: 20
139
+ - training_steps: 75
140
+
141
+ ### Training results
142
+
143
+ | Training Loss | Epoch | Step | Validation Loss |
144
+ |:-------------:|:------:|:----:|:---------------:|
145
+ | 37.307 | 0.0047 | 1 | 2.2704 |
146
+ | 26.2165 | 0.2336 | 50 | 1.8455 |
147
+
148
+
149
+ ### Framework versions
150
+
151
+ - PEFT 0.13.2
152
+ - Transformers 4.46.0
153
+ - Pytorch 2.5.0+cu124
154
+ - Datasets 3.0.1
155
+ - Tokenizers 0.20.1
adapter_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:46b357726cac335ec3b6b1b0565eca402daf3e042dc7b5026a25ed9205d6eeae
3
+ size 100734154
adapter_model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:d5ecc50627cc77392872a84b376379dc560a5df887496051d111c54b00b31dd6
3
  size 100690288
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:59c4c05e5e3cb61a90bdbe376893631ff9b8cbfd8b9b4bedc9ac7530dfbe0424
3
  size 100690288