Hugging Face's logo

DUAL-GPO
/
phi-2-gpo-iter-0

PEFT
Safetensors
phi
alignment-handbook
Generated from Trainer
trl
dpo
custom_code
Model card Files Community
lole25 commited on
Commit
e34affd
•
1 Parent(s): 8a71329

Model save

Browse files
Files changed (7) hide show
  1. README.md +1 -5
  2. adapter_config.json +1 -1
  3. adapter_model.safetensors +1 -1
  4. all_results.json +5 -5
  5. train_results.json +5 -5
  6. trainer_state.json +9 -37
  7. training_args.bin +1 -1
README.md CHANGED
@@ -2,13 +2,9 @@
2
  license: mit
3
  library_name: peft
4
  tags:
5
- - alignment-handbook
6
- - generated_from_trainer
7
  - trl
8
  - dpo
9
  - generated_from_trainer
10
- datasets:
11
- - HuggingFaceH4/ultrafeedback_binarized
12
  base_model: microsoft/phi-2
13
  model-index:
14
  - name: phi-2-gpo-iter-0
@@ -20,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
20
 
21
  # phi-2-gpo-iter-0
22
 
23
- This model is a fine-tuned version of [lole25/phi-2-sft-ultrachat-lora](https://huggingface.co/lole25/phi-2-sft-ultrachat-lora) on the HuggingFaceH4/ultrafeedback_binarized dataset.
24
 
25
  ## Model description
26
 
 
2
  license: mit
3
  library_name: peft
4
  tags:
 
 
5
  - trl
6
  - dpo
7
  - generated_from_trainer
 
 
8
  base_model: microsoft/phi-2
9
  model-index:
10
  - name: phi-2-gpo-iter-0
 
16
 
17
  # phi-2-gpo-iter-0
18
 
19
+ This model is a fine-tuned version of [microsoft/phi-2](https://huggingface.co/microsoft/phi-2) on the None dataset.
20
 
21
  ## Model description
22
 
adapter_config.json CHANGED
@@ -19,9 +19,9 @@
19
  "rank_pattern": {},
20
  "revision": null,
21
  "target_modules": [
 
22
  "dense",
23
  "q_proj",
24
- "v_proj",
25
  "k_proj"
26
  ],
27
  "task_type": "CAUSAL_LM"
 
19
  "rank_pattern": {},
20
  "revision": null,
21
  "target_modules": [
22
+ "v_proj",
23
  "dense",
24
  "q_proj",
 
25
  "k_proj"
26
  ],
27
  "task_type": "CAUSAL_LM"
adapter_model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:f144d13fd4cedf5df77ead36d7d376e79ec3dbe6c14042eae2d43dc8d35030b5
3
  size 41977616
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:33a91970ad32fd59ca792aa2083a2c74b90dac687a074c2c49e8e55236c0a2c0
3
  size 41977616
all_results.json CHANGED
@@ -1,8 +1,8 @@
1
  {
2
- "epoch": 1.99,
3
- "train_loss": 0.001963847136545566,
4
- "train_runtime": 279.1577,
5
  "train_samples": 30567,
6
- "train_samples_per_second": 7.164,
7
- "train_steps_per_second": 0.444
8
  }
 
1
  {
2
+ "epoch": 1.6,
3
+ "train_loss": 0.0,
4
+ "train_runtime": 4.6383,
5
  "train_samples": 30567,
6
+ "train_samples_per_second": 4.312,
7
+ "train_steps_per_second": 0.431
8
  }
train_results.json CHANGED
@@ -1,8 +1,8 @@
1
  {
2
- "epoch": 1.99,
3
- "train_loss": 0.001963847136545566,
4
- "train_runtime": 279.1577,
5
  "train_samples": 30567,
6
- "train_samples_per_second": 7.164,
7
- "train_steps_per_second": 0.444
8
  }
 
1
  {
2
+ "epoch": 1.6,
3
+ "train_loss": 0.0,
4
+ "train_runtime": 4.6383,
5
  "train_samples": 30567,
6
+ "train_samples_per_second": 4.312,
7
+ "train_steps_per_second": 0.431
8
  }
trainer_state.json CHANGED
@@ -1,9 +1,9 @@
1
  {
2
  "best_metric": null,
3
  "best_model_checkpoint": null,
4
- "epoch": 1.992,
5
  "eval_steps": 100,
6
- "global_step": 124,
7
  "is_hyper_param_search": false,
8
  "is_local_process_zero": true,
9
  "is_world_process_zero": true,
@@ -163,45 +163,17 @@
163
  "step": 100
164
  },
165
  {
166
- "epoch": 1.77,
167
- "learning_rate": 1.937002879188285e-07,
168
- "logits/chosen": 0.9077476263046265,
169
- "logits/rejected": 0.8532902598381042,
170
- "logps/chosen": -142.3281707763672,
171
- "logps/rejected": -133.73294067382812,
172
- "loss": 0.0102,
173
- "rewards/accuracies": 0.4375,
174
- "rewards/chosen": -0.0010257468093186617,
175
- "rewards/margins": 0.00028010181267745793,
176
- "rewards/rejected": -0.0013058485928922892,
177
- "step": 110
178
- },
179
- {
180
- "epoch": 1.93,
181
- "learning_rate": 1.6003680950742728e-08,
182
- "logits/chosen": 0.9056937098503113,
183
- "logits/rejected": 0.8917877078056335,
184
- "logps/chosen": -147.62950134277344,
185
- "logps/rejected": -135.589599609375,
186
- "loss": 0.0102,
187
- "rewards/accuracies": 0.45625001192092896,
188
- "rewards/chosen": -0.00017889760783873498,
189
- "rewards/margins": 0.0004578630905598402,
190
- "rewards/rejected": -0.0006367607275024056,
191
- "step": 120
192
- },
193
- {
194
- "epoch": 1.99,
195
- "step": 124,
196
  "total_flos": 0.0,
197
- "train_loss": 0.001963847136545566,
198
- "train_runtime": 279.1577,
199
- "train_samples_per_second": 7.164,
200
- "train_steps_per_second": 0.444
201
  }
202
  ],
203
  "logging_steps": 10,
204
- "max_steps": 124,
205
  "num_input_tokens_seen": 0,
206
  "num_train_epochs": 2,
207
  "save_steps": 100,
 
1
  {
2
  "best_metric": null,
3
  "best_model_checkpoint": null,
4
+ "epoch": 1.6,
5
  "eval_steps": 100,
6
+ "global_step": 100,
7
  "is_hyper_param_search": false,
8
  "is_local_process_zero": true,
9
  "is_world_process_zero": true,
 
163
  "step": 100
164
  },
165
  {
166
+ "epoch": 1.6,
167
+ "step": 100,
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
168
  "total_flos": 0.0,
169
+ "train_loss": 0.0,
170
+ "train_runtime": 4.6383,
171
+ "train_samples_per_second": 4.312,
172
+ "train_steps_per_second": 0.431
173
  }
174
  ],
175
  "logging_steps": 10,
176
+ "max_steps": 2,
177
  "num_input_tokens_seen": 0,
178
  "num_train_epochs": 2,
179
  "save_steps": 100,
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:65d0b7c47a66f6abb907b621b64d73257ac8eedd85947a8ad6ec086e6a803cf7
3
  size 5816
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3b123edc447e2fafb1d5a4273fbf4322f2b97753c2aa1eb6b83df8a20af63c9a
3
  size 5816