Hugging Face's logo

ptoro
/
EvolCodeLlama-7b

PEFT
Safetensors
llama
axolotl
Generated from Trainer
Model card Files
xet
Community
ptoro commited on
Commit
3ae0e54
·
verified ·
1 Parent(s): 8502190

End of training

Browse files
Files changed (2) hide show
  1. README.md +229 -0
  2. adapter_model.bin +3 -0
README.md ADDED
@@ -0,0 +1,229 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: llama2
3
+ library_name: peft
4
+ tags:
5
+ - axolotl
6
+ - generated_from_trainer
7
+ base_model: codellama/CodeLlama-7b-hf
8
+ model-index:
9
+ - name: EvolCodeLlama-7b
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/OpenAccess-AI-Collective/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/OpenAccess-AI-Collective/axolotl)
17
+ <details><summary>See axolotl config</summary>
18
+
19
+ axolotl version: `0.4.0`
20
+ ```yaml
21
+ base_model: codellama/CodeLlama-7b-hf
22
+ base_model_config: codellama/CodeLlama-7b-hf
23
+ model_type: LlamaForCausalLM
24
+ tokenizer_type: LlamaTokenizer
25
+ is_llama_derived_model: true
26
+ hub_model_id: EvolCodeLlama-7b
27
+
28
+ load_in_8bit: false
29
+ load_in_4bit: true
30
+ strict: false
31
+
32
+ datasets:
33
+ - path: ptoro/Evol-Instruct-Python-1k-testing
34
+ type: alpaca
35
+ dataset_prepared_path: last_run_prepared
36
+ val_set_size: 0.02
37
+ output_dir: ./qlora-out
38
+
39
+ adapter: qlora
40
+ lora_model_dir:
41
+
42
+ sequence_len: 2048
43
+ sample_packing: true
44
+
45
+ lora_r: 32
46
+ lora_alpha: 16
47
+ lora_dropout: 0.05
48
+ lora_target_modules:
49
+ lora_target_linear: true
50
+ lora_fan_in_fan_out:
51
+
52
+ wandb_project: axolotl
53
+ wandb_entity:
54
+ wandb_watch:
55
+ wandb_run_id:
56
+ wandb_log_model:
57
+
58
+ gradient_accumulation_steps: 4
59
+ micro_batch_size: 2
60
+ num_epochs: 3
61
+ optimizer: paged_adamw_32bit
62
+ lr_scheduler: cosine
63
+ learning_rate: 0.0002
64
+
65
+ train_on_inputs: false
66
+ group_by_length: false
67
+ bf16: true
68
+ fp16: false
69
+ tf32: false
70
+
71
+ gradient_checkpointing: true
72
+ early_stopping_patience:
73
+ resume_from_checkpoint:
74
+ local_rank:
75
+ logging_steps: 1
76
+ xformers_attention:
77
+ flash_attention: true
78
+
79
+ warmup_steps: 100
80
+ eval_steps: 0.01
81
+ save_strategy: epoch
82
+ save_steps:
83
+ debug:
84
+ deepspeed:
85
+ weight_decay: 0.0
86
+ fsdp:
87
+ fsdp_config:
88
+ special_tokens:
89
+ bos_token: "<s>"
90
+ eos_token: "</s>"
91
+ unk_token: "<unk>"
92
+ ```
93
+
94
+ </details><br>
95
+
96
+ # EvolCodeLlama-7b
97
+
98
+ This model is a fine-tuned version of [codellama/CodeLlama-7b-hf](https://huggingface.co/codellama/CodeLlama-7b-hf) on the None dataset.
99
+ It achieves the following results on the evaluation set:
100
+ - Loss: 0.3828
101
+
102
+ ## Model description
103
+
104
+ More information needed
105
+
106
+ ## Intended uses & limitations
107
+
108
+ More information needed
109
+
110
+ ## Training and evaluation data
111
+
112
+ More information needed
113
+
114
+ ## Training procedure
115
+
116
+ ### Training hyperparameters
117
+
118
+ The following hyperparameters were used during training:
119
+ - learning_rate: 0.0002
120
+ - train_batch_size: 2
121
+ - eval_batch_size: 2
122
+ - seed: 42
123
+ - gradient_accumulation_steps: 4
124
+ - total_train_batch_size: 8
125
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
126
+ - lr_scheduler_type: cosine
127
+ - lr_scheduler_warmup_steps: 100
128
+ - num_epochs: 3
129
+
130
+ ### Training results
131
+
132
+ | Training Loss | Epoch | Step | Validation Loss |
133
+ |:-------------:|:-----:|:----:|:---------------:|
134
+ | 0.3627 | 0.01 | 1 | 0.5027 |
135
+ | 0.3412 | 0.03 | 4 | 0.5026 |
136
+ | 0.3806 | 0.07 | 8 | 0.5023 |
137
+ | 0.392 | 0.1 | 12 | 0.5018 |
138
+ | 0.4141 | 0.14 | 16 | 0.4999 |
139
+ | 0.3433 | 0.17 | 20 | 0.4954 |
140
+ | 0.3702 | 0.21 | 24 | 0.4851 |
141
+ | 0.2948 | 0.24 | 28 | 0.4682 |
142
+ | 0.3387 | 0.28 | 32 | 0.4499 |
143
+ | 0.2437 | 0.31 | 36 | 0.4331 |
144
+ | 0.2526 | 0.35 | 40 | 0.4221 |
145
+ | 0.2721 | 0.38 | 44 | 0.4146 |
146
+ | 0.2292 | 0.42 | 48 | 0.4089 |
147
+ | 0.1986 | 0.45 | 52 | 0.4028 |
148
+ | 0.3258 | 0.48 | 56 | 0.3983 |
149
+ | 0.3509 | 0.52 | 60 | 0.3950 |
150
+ | 0.2697 | 0.55 | 64 | 0.3926 |
151
+ | 0.2646 | 0.59 | 68 | 0.3907 |
152
+ | 0.3979 | 0.62 | 72 | 0.3900 |
153
+ | 0.2737 | 0.66 | 76 | 0.3880 |
154
+ | 0.2271 | 0.69 | 80 | 0.3865 |
155
+ | 0.247 | 0.73 | 84 | 0.3847 |
156
+ | 0.3112 | 0.76 | 88 | 0.3824 |
157
+ | 0.2724 | 0.8 | 92 | 0.3820 |
158
+ | 0.207 | 0.83 | 96 | 0.3814 |
159
+ | 0.3492 | 0.87 | 100 | 0.3810 |
160
+ | 0.2474 | 0.9 | 104 | 0.3802 |
161
+ | 0.4037 | 0.94 | 108 | 0.3785 |
162
+ | 0.2295 | 0.97 | 112 | 0.3773 |
163
+ | 0.2689 | 1.0 | 116 | 0.3760 |
164
+ | 0.2546 | 1.02 | 120 | 0.3753 |
165
+ | 0.1916 | 1.05 | 124 | 0.3768 |
166
+ | 0.2458 | 1.09 | 128 | 0.3758 |
167
+ | 0.2155 | 1.12 | 132 | 0.3768 |
168
+ | 0.2341 | 1.16 | 136 | 0.3773 |
169
+ | 0.1909 | 1.19 | 140 | 0.3793 |
170
+ | 0.1911 | 1.23 | 144 | 0.3759 |
171
+ | 0.2096 | 1.26 | 148 | 0.3761 |
172
+ | 0.2353 | 1.29 | 152 | 0.3772 |
173
+ | 0.2606 | 1.33 | 156 | 0.3773 |
174
+ | 0.1485 | 1.36 | 160 | 0.3778 |
175
+ | 0.1807 | 1.4 | 164 | 0.3749 |
176
+ | 0.2294 | 1.43 | 168 | 0.3770 |
177
+ | 0.216 | 1.47 | 172 | 0.3759 |
178
+ | 0.1791 | 1.5 | 176 | 0.3727 |
179
+ | 0.2605 | 1.54 | 180 | 0.3733 |
180
+ | 0.2838 | 1.57 | 184 | 0.3738 |
181
+ | 0.2632 | 1.61 | 188 | 0.3694 |
182
+ | 0.1839 | 1.64 | 192 | 0.3686 |
183
+ | 0.1939 | 1.68 | 196 | 0.3690 |
184
+ | 0.2413 | 1.71 | 200 | 0.3699 |
185
+ | 0.1494 | 1.74 | 204 | 0.3689 |
186
+ | 0.2782 | 1.78 | 208 | 0.3695 |
187
+ | 0.2314 | 1.81 | 212 | 0.3696 |
188
+ | 0.2499 | 1.85 | 216 | 0.3691 |
189
+ | 0.1976 | 1.88 | 220 | 0.3672 |
190
+ | 0.2587 | 1.92 | 224 | 0.3660 |
191
+ | 0.2598 | 1.95 | 228 | 0.3658 |
192
+ | 0.2686 | 1.99 | 232 | 0.3666 |
193
+ | 0.216 | 2.01 | 236 | 0.3673 |
194
+ | 0.1261 | 2.04 | 240 | 0.3723 |
195
+ | 0.1938 | 2.08 | 244 | 0.3811 |
196
+ | 0.1906 | 2.11 | 248 | 0.3869 |
197
+ | 0.1375 | 2.15 | 252 | 0.3829 |
198
+ | 0.228 | 2.18 | 256 | 0.3796 |
199
+ | 0.2524 | 2.22 | 260 | 0.3789 |
200
+ | 0.118 | 2.25 | 264 | 0.3809 |
201
+ | 0.2224 | 2.29 | 268 | 0.3834 |
202
+ | 0.1477 | 2.32 | 272 | 0.3847 |
203
+ | 0.2095 | 2.35 | 276 | 0.3849 |
204
+ | 0.1919 | 2.39 | 280 | 0.3820 |
205
+ | 0.1916 | 2.42 | 284 | 0.3804 |
206
+ | 0.1625 | 2.46 | 288 | 0.3788 |
207
+ | 0.2054 | 2.49 | 292 | 0.3794 |
208
+ | 0.1605 | 2.53 | 296 | 0.3810 |
209
+ | 0.1564 | 2.56 | 300 | 0.3819 |
210
+ | 0.196 | 2.6 | 304 | 0.3822 |
211
+ | 0.1975 | 2.63 | 308 | 0.3830 |
212
+ | 0.1406 | 2.67 | 312 | 0.3833 |
213
+ | 0.2754 | 2.7 | 316 | 0.3830 |
214
+ | 0.1544 | 2.74 | 320 | 0.3829 |
215
+ | 0.1733 | 2.77 | 324 | 0.3830 |
216
+ | 0.1862 | 2.81 | 328 | 0.3832 |
217
+ | 0.1634 | 2.84 | 332 | 0.3829 |
218
+ | 0.1966 | 2.87 | 336 | 0.3830 |
219
+ | 0.1306 | 2.91 | 340 | 0.3831 |
220
+ | 0.1444 | 2.94 | 344 | 0.3828 |
221
+
222
+
223
+ ### Framework versions
224
+
225
+ - PEFT 0.7.2.dev0
226
+ - Transformers 4.37.0
227
+ - Pytorch 2.0.1+cu118
228
+ - Datasets 2.16.1
229
+ - Tokenizers 0.15.0
adapter_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:080a9dc35cbad6849176c7703510bf01ab0c132169c7cd9aa15d3382eb7dc594
3
+ size 319977229