Hugging Face's logo

MichaelR207
/
mistral-sft-7b-dpo-qlora

PEFT
Safetensors
mistral
alignment-handbook
trl
dpo
Generated from Trainer
Model card Files Community
MichaelR207 commited on
Commit
53d9518
•
1 Parent(s): 59f4671

Model save

Browse files
Files changed (7) hide show
  1. README.md +1 -17
  2. adapter_config.json +3 -3
  3. adapter_model.safetensors +1 -1
  4. all_results.json +5 -18
  5. train_results.json +5 -5
  6. trainer_state.json +9 -9
  7. training_args.bin +1 -1
README.md CHANGED
@@ -2,16 +2,10 @@
2
  license: mit
3
  library_name: peft
4
  tags:
5
- - alignment-handbook
6
- - trl
7
- - dpo
8
- - generated_from_trainer
9
  - trl
10
  - dpo
11
  - generated_from_trainer
12
  base_model: HuggingFaceH4/mistral-7b-sft-beta
13
- datasets:
14
- - HuggingFaceH4/ultrafeedback_binarized
15
  model-index:
16
  - name: mistral-sft-7b-dpo-qlora
17
  results: []
@@ -22,17 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
22
 
23
  # mistral-sft-7b-dpo-qlora
24
 
25
- This model is a fine-tuned version of [HuggingFaceH4/mistral-7b-sft-beta](https://huggingface.co/HuggingFaceH4/mistral-7b-sft-beta) on the HuggingFaceH4/ultrafeedback_binarized dataset.
26
- It achieves the following results on the evaluation set:
27
- - Loss: 0.6932
28
- - Rewards/chosen: 0.0023
29
- - Rewards/rejected: 0.0013
30
- - Rewards/accuracies: 0.5625
31
- - Rewards/margins: 0.0009
32
- - Logps/rejected: -122.8616
33
- - Logps/chosen: -86.2666
34
- - Logits/rejected: -3.0452
35
- - Logits/chosen: -2.9819
36
 
37
  ## Model description
38
 
 
2
  license: mit
3
  library_name: peft
4
  tags:
 
 
 
 
5
  - trl
6
  - dpo
7
  - generated_from_trainer
8
  base_model: HuggingFaceH4/mistral-7b-sft-beta
 
 
9
  model-index:
10
  - name: mistral-sft-7b-dpo-qlora
11
  results: []
 
16
 
17
  # mistral-sft-7b-dpo-qlora
18
 
19
+ This model is a fine-tuned version of [HuggingFaceH4/mistral-7b-sft-beta](https://huggingface.co/HuggingFaceH4/mistral-7b-sft-beta) on the None dataset.
 
 
 
 
 
 
 
 
 
 
20
 
21
  ## Model description
22
 
adapter_config.json CHANGED
@@ -19,13 +19,13 @@
19
  "rank_pattern": {},
20
  "revision": null,
21
  "target_modules": [
22
- "o_proj",
23
  "q_proj",
 
24
  "up_proj",
25
  "v_proj",
26
  "k_proj",
27
- "gate_proj",
28
- "down_proj"
29
  ],
30
  "task_type": "CAUSAL_LM"
31
  }
 
19
  "rank_pattern": {},
20
  "revision": null,
21
  "target_modules": [
 
22
  "q_proj",
23
+ "down_proj",
24
  "up_proj",
25
  "v_proj",
26
  "k_proj",
27
+ "o_proj",
28
+ "gate_proj"
29
  ],
30
  "task_type": "CAUSAL_LM"
31
  }
adapter_model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:fc0793e924d0764c4d8614e0b1426ed310afc447165daf63c4f1609395b2d9b3
3
  size 671150064
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:daa5b1cc234baf49e019603995aa95d66d64ffeb5c49dcbbc359f20967c66145
3
  size 671150064
all_results.json CHANGED
@@ -1,21 +1,8 @@
1
  {
2
  "epoch": 5.0,
3
- "eval_logits/chosen": -2.9818739891052246,
4
- "eval_logits/rejected": -3.045213222503662,
5
- "eval_logps/chosen": -86.26663208007812,
6
- "eval_logps/rejected": -122.861572265625,
7
- "eval_loss": 0.6932176947593689,
8
- "eval_rewards/accuracies": 0.5625,
9
- "eval_rewards/chosen": 0.0022567124105989933,
10
- "eval_rewards/margins": 0.000948949484154582,
11
- "eval_rewards/rejected": 0.0013077628100290895,
12
- "eval_runtime": 11.9466,
13
- "eval_samples": 20,
14
- "eval_samples_per_second": 1.674,
15
- "eval_steps_per_second": 0.167,
16
- "train_loss": 0.34364445209503175,
17
- "train_runtime": 77.9014,
18
- "train_samples": 10,
19
- "train_samples_per_second": 0.642,
20
- "train_steps_per_second": 0.064
21
  }
 
1
  {
2
  "epoch": 5.0,
3
+ "train_loss": 0.17302747964859008,
4
+ "train_runtime": 39.8084,
5
+ "train_samples": 5,
6
+ "train_samples_per_second": 0.628,
7
+ "train_steps_per_second": 0.126
 
 
 
 
 
 
 
 
 
 
 
 
 
8
  }
train_results.json CHANGED
@@ -1,8 +1,8 @@
1
  {
2
  "epoch": 5.0,
3
- "train_loss": 0.34364445209503175,
4
- "train_runtime": 77.9014,
5
- "train_samples": 10,
6
- "train_samples_per_second": 0.642,
7
- "train_steps_per_second": 0.064
8
  }
 
1
  {
2
  "epoch": 5.0,
3
+ "train_loss": 0.17302747964859008,
4
+ "train_runtime": 39.8084,
5
+ "train_samples": 5,
6
+ "train_samples_per_second": 0.628,
7
+ "train_steps_per_second": 0.126
8
  }
trainer_state.json CHANGED
@@ -11,11 +11,11 @@
11
  {
12
  "epoch": 1.0,
13
  "learning_rate": 5e-06,
14
- "logits/chosen": -2.811583995819092,
15
- "logits/rejected": -2.8277242183685303,
16
- "logps/chosen": -114.9925765991211,
17
- "logps/rejected": -127.65025329589844,
18
- "loss": 0.3466,
19
  "rewards/accuracies": 0.0,
20
  "rewards/chosen": 0.0,
21
  "rewards/margins": 0.0,
@@ -26,10 +26,10 @@
26
  "epoch": 5.0,
27
  "step": 5,
28
  "total_flos": 0.0,
29
- "train_loss": 0.34364445209503175,
30
- "train_runtime": 77.9014,
31
- "train_samples_per_second": 0.642,
32
- "train_steps_per_second": 0.064
33
  }
34
  ],
35
  "logging_steps": 10,
 
11
  {
12
  "epoch": 1.0,
13
  "learning_rate": 5e-06,
14
+ "logits/chosen": -2.7210071086883545,
15
+ "logits/rejected": -2.6927733421325684,
16
+ "logps/chosen": -90.6630859375,
17
+ "logps/rejected": -77.25151062011719,
18
+ "loss": 0.1733,
19
  "rewards/accuracies": 0.0,
20
  "rewards/chosen": 0.0,
21
  "rewards/margins": 0.0,
 
26
  "epoch": 5.0,
27
  "step": 5,
28
  "total_flos": 0.0,
29
+ "train_loss": 0.17302747964859008,
30
+ "train_runtime": 39.8084,
31
+ "train_samples_per_second": 0.628,
32
+ "train_steps_per_second": 0.126
33
  }
34
  ],
35
  "logging_steps": 10,
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:42394665981968fa2b0c9bacc2ad019c23bfacc4431c755e80b9bb7affd7d253
3
  size 6200
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e5d09a67ee422860ce5a206b24a613c73101c711253665adc16f2aa44af0b0fe
3
  size 6200