Hugging Face's logo

willtensora
/
ad641a5b-ef11-4278-80e4-9119f53c47f4

PEFT
Safetensors
llama
axolotl
Generated from Trainer
Model card Files Community
willtensora commited on
Commit
a63e195
·
verified ·
1 Parent(s): 7a1ac56

End of training

Browse files
Files changed (3) hide show
  1. README.md +174 -0
  2. adapter_model.bin +3 -0
  3. adapter_model.safetensors +1 -1
README.md ADDED
@@ -0,0 +1,174 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: peft
3
+ license: apache-2.0
4
+ base_model: TinyLlama/TinyLlama-1.1B-Chat-v0.6
5
+ tags:
6
+ - axolotl
7
+ - generated_from_trainer
8
+ model-index:
9
+ - name: ad641a5b-ef11-4278-80e4-9119f53c47f4
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: TinyLlama/TinyLlama-1.1B-Chat-v0.6
23
+ batch_size: 8
24
+ bf16: true
25
+ chat_template: tokenizer_default_fallback_alpaca
26
+ datasets:
27
+ - data_files:
28
+ - 11bfaf21b106be7f_train_data.json
29
+ ds_type: json
30
+ format: custom
31
+ path: /workspace/input_data/11bfaf21b106be7f_train_data.json
32
+ type:
33
+ field_input: project_and_commit_id
34
+ field_instruction: source
35
+ field_output: target
36
+ format: '{instruction} {input}'
37
+ no_input_format: '{instruction}'
38
+ system_format: '{system}'
39
+ system_prompt: ''
40
+ early_stopping_patience: 3
41
+ eval_steps: 50
42
+ flash_attention: true
43
+ gpu_memory_limit: 80GiB
44
+ gradient_checkpointing: true
45
+ group_by_length: true
46
+ hub_model_id: willtensora/ad641a5b-ef11-4278-80e4-9119f53c47f4
47
+ hub_strategy: checkpoint
48
+ learning_rate: 0.0002
49
+ logging_steps: 10
50
+ lora_alpha: 256
51
+ lora_dropout: 0.1
52
+ lora_r: 128
53
+ lora_target_linear: true
54
+ lr_scheduler: cosine
55
+ micro_batch_size: 1
56
+ model_type: AutoModelForCausalLM
57
+ num_epochs: 100
58
+ optimizer: adamw_bnb_8bit
59
+ output_dir: miner_id_24
60
+ pad_to_sequence_len: true
61
+ resize_token_embeddings_to_32x: false
62
+ sample_packing: false
63
+ save_steps: 50
64
+ sequence_len: 2048
65
+ tokenizer_type: LlamaTokenizerFast
66
+ train_on_inputs: false
67
+ trust_remote_code: true
68
+ val_set_size: 0.1
69
+ wandb_entity: ''
70
+ wandb_mode: online
71
+ wandb_project: Gradients-On-Demand
72
+ wandb_run: your_name
73
+ wandb_runid: default
74
+ warmup_ratio: 0.05
75
+ xformers_attention: true
76
+
77
+ ```
78
+
79
+ </details><br>
80
+
81
+ # ad641a5b-ef11-4278-80e4-9119f53c47f4
82
+
83
+ This model is a fine-tuned version of [TinyLlama/TinyLlama-1.1B-Chat-v0.6](https://huggingface.co/TinyLlama/TinyLlama-1.1B-Chat-v0.6) on the None dataset.
84
+ It achieves the following results on the evaluation set:
85
+ - Loss: 0.2704
86
+
87
+ ## Model description
88
+
89
+ More information needed
90
+
91
+ ## Intended uses & limitations
92
+
93
+ More information needed
94
+
95
+ ## Training and evaluation data
96
+
97
+ More information needed
98
+
99
+ ## Training procedure
100
+
101
+ ### Training hyperparameters
102
+
103
+ The following hyperparameters were used during training:
104
+ - learning_rate: 0.0002
105
+ - train_batch_size: 1
106
+ - eval_batch_size: 1
107
+ - seed: 42
108
+ - distributed_type: multi-GPU
109
+ - num_devices: 8
110
+ - total_train_batch_size: 8
111
+ - total_eval_batch_size: 8
112
+ - optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
113
+ - lr_scheduler_type: cosine
114
+ - lr_scheduler_warmup_steps: 589
115
+ - num_epochs: 100
116
+
117
+ ### Training results
118
+
119
+ | Training Loss | Epoch | Step | Validation Loss |
120
+ |:-------------:|:------:|:----:|:---------------:|
121
+ | No log | 0.0011 | 1 | 2.0060 |
122
+ | 1.2664 | 0.0530 | 50 | 1.2713 |
123
+ | 0.9659 | 0.1060 | 100 | 0.9642 |
124
+ | 0.7596 | 0.1591 | 150 | 0.8311 |
125
+ | 0.7092 | 0.2121 | 200 | 0.7583 |
126
+ | 0.6691 | 0.2651 | 250 | 0.6944 |
127
+ | 0.65 | 0.3181 | 300 | 0.6577 |
128
+ | 0.6149 | 0.3712 | 350 | 0.6268 |
129
+ | 0.5929 | 0.4242 | 400 | 0.5945 |
130
+ | 0.5319 | 0.4772 | 450 | 0.5820 |
131
+ | 0.5136 | 0.5302 | 500 | 0.5576 |
132
+ | 0.5258 | 0.5832 | 550 | 0.5367 |
133
+ | 0.4476 | 0.6363 | 600 | 0.5141 |
134
+ | 0.5018 | 0.6893 | 650 | 0.4943 |
135
+ | 0.4851 | 0.7423 | 700 | 0.4861 |
136
+ | 0.41 | 0.7953 | 750 | 0.4693 |
137
+ | 0.4625 | 0.8484 | 800 | 0.4552 |
138
+ | 0.4909 | 0.9014 | 850 | 0.4421 |
139
+ | 0.3885 | 0.9544 | 900 | 0.4196 |
140
+ | 0.3408 | 1.0074 | 950 | 0.4111 |
141
+ | 0.2804 | 1.0604 | 1000 | 0.4020 |
142
+ | 0.3503 | 1.1135 | 1050 | 0.3875 |
143
+ | 0.291 | 1.1665 | 1100 | 0.3958 |
144
+ | 0.3025 | 1.2195 | 1150 | 0.3849 |
145
+ | 0.2749 | 1.2725 | 1200 | 0.3729 |
146
+ | 0.3222 | 1.3256 | 1250 | 0.3631 |
147
+ | 0.2895 | 1.3786 | 1300 | 0.3570 |
148
+ | 0.2994 | 1.4316 | 1350 | 0.3470 |
149
+ | 0.3055 | 1.4846 | 1400 | 0.3431 |
150
+ | 0.2252 | 1.5376 | 1450 | 0.3351 |
151
+ | 0.2816 | 1.5907 | 1500 | 0.3214 |
152
+ | 0.3065 | 1.6437 | 1550 | 0.3163 |
153
+ | 0.2727 | 1.6967 | 1600 | 0.3158 |
154
+ | 0.2673 | 1.7497 | 1650 | 0.3123 |
155
+ | 0.276 | 1.8028 | 1700 | 0.3090 |
156
+ | 0.217 | 1.8558 | 1750 | 0.3021 |
157
+ | 0.2712 | 1.9088 | 1800 | 0.2950 |
158
+ | 0.2175 | 1.9618 | 1850 | 0.2927 |
159
+ | 0.1561 | 2.0148 | 1900 | 0.2911 |
160
+ | 0.1557 | 2.0679 | 1950 | 0.2773 |
161
+ | 0.1404 | 2.1209 | 2000 | 0.2725 |
162
+ | 0.1386 | 2.1739 | 2050 | 0.2696 |
163
+ | 0.1224 | 2.2269 | 2100 | 0.2780 |
164
+ | 0.1535 | 2.2800 | 2150 | 0.2713 |
165
+ | 0.1494 | 2.3330 | 2200 | 0.2704 |
166
+
167
+
168
+ ### Framework versions
169
+
170
+ - PEFT 0.13.2
171
+ - Transformers 4.46.0
172
+ - Pytorch 2.5.0+cu124
173
+ - Datasets 3.0.1
174
+ - Tokenizers 0.20.1
adapter_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b05c6860190bb84986157728f6c1b611cee5778235af9765e6aed540596ebc91
3
+ size 403813018
adapter_model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:c73215e6d914edf8142dd717a36edb3fb738801179e09242da3da6e9f1d10e79
3
  size 403743472
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8d3b3ecb3d46a57ac86b1539d1dbee095f0308d7d48e60904b852911fb26c8f5
3
  size 403743472