Hugging Face's logo

hugodk-sch
/
aftonposten-6b-align-scan

PEFT
Safetensors
gptj
alignment-handbook
trl
dpo
Generated from Trainer
4-bit precision
bitsandbytes
Model card Files Community
hugodk-sch commited on
Commit
672138a
•
1 Parent(s): 64cc945

Training in progress, step 400

Browse files
Files changed (8) hide show
  1. README.md +15 -27
  2. adapter_config.json +3 -3
  3. adapter_model.safetensors +1 -1
  4. all_results.json +15 -15
  5. eval_results.json +11 -11
  6. train_results.json +5 -5
  7. trainer_state.json +0 -0
  8. training_args.bin +1 -1
README.md CHANGED
@@ -20,15 +20,15 @@ should probably proofread and complete it, then remove this comment. -->
20
 
21
  This model is a fine-tuned version of [data/ap-gpt-j-6b-sft-qlora-04-08](https://huggingface.co/data/ap-gpt-j-6b-sft-qlora-04-08) on the hugodk-sch/aftonposten_title_prefs dataset.
22
  It achieves the following results on the evaluation set:
23
- - Loss: 6.0779
24
- - Rewards/chosen: -0.0133
25
- - Rewards/rejected: -0.0383
26
- - Rewards/accuracies: 0.5714
27
- - Rewards/margins: 0.0250
28
- - Logps/rejected: -37.7083
29
- - Logps/chosen: -34.1011
30
- - Logits/rejected: -2.2004
31
- - Logits/chosen: -2.2052
32
 
33
  ## Model description
34
 
@@ -57,27 +57,15 @@ The following hyperparameters were used during training:
57
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
58
  - lr_scheduler_type: cosine
59
  - lr_scheduler_warmup_ratio: 0.1
60
- - num_epochs: 4
61
 
62
  ### Training results
63
 
64
- | Training Loss | Epoch | Step | Logits/chosen | Logits/rejected | Logps/chosen | Logps/rejected | Validation Loss | Rewards/accuracies | Rewards/chosen | Rewards/margins | Rewards/rejected |
65
- |:-------------:|:-----:|:----:|:-------------:|:---------------:|:------------:|:--------------:|:---------------:|:------------------:|:--------------:|:---------------:|:----------------:|
66
- | 5.6745 | 0.26 | 100 | -2.2338 | -2.2290 | -34.0153 | -37.5181 | 6.2508 | 0.5461 | 0.0038 | 0.0041 | -0.0003 |
67
- | 5.2135 | 0.52 | 200 | -2.2315 | -2.2267 | -34.0014 | -37.5042 | 6.2881 | 0.5403 | 0.0066 | 0.0041 | 0.0025 |
68
- | 4.3883 | 0.78 | 300 | -2.2291 | -2.2243 | -34.0193 | -37.5325 | 6.2382 | 0.5166 | 0.0031 | 0.0062 | -0.0032 |
69
- | 2.9753 | 1.04 | 400 | 6.0369 | 0.0069 | -0.0106 | 0.6034 | 0.0176 | -37.5698 | -33.9999 | -2.2093 | -2.2141 |
70
- | 2.4163 | 1.3 | 500 | 6.0677 | -0.0149 | -0.0375 | 0.5801 | 0.0225 | -37.7039 | -34.1092 | -2.1858 | -2.1907 |
71
- | 2.52 | 1.56 | 600 | 5.9990 | -0.0097 | -0.0348 | 0.5748 | 0.0251 | -37.6905 | -34.0832 | -2.1951 | -2.1999 |
72
- | 2.9186 | 1.82 | 700 | 6.1696 | -0.0176 | -0.0364 | 0.5598 | 0.0188 | -37.6988 | -34.1227 | -2.2048 | -2.2097 |
73
- | 1.2867 | 2.08 | 800 | 6.0594 | -0.0122 | -0.0361 | 0.5777 | 0.0239 | -37.6970 | -34.0957 | -2.2060 | -2.2109 |
74
- | 0.8862 | 2.34 | 900 | 6.0621 | -0.0165 | -0.0403 | 0.5918 | 0.0237 | -37.7179 | -34.1172 | -2.2027 | -2.2076 |
75
- | 1.2395 | 2.6 | 1000 | 6.0000 | -0.0163 | -0.0418 | 0.5864 | 0.0255 | -37.7257 | -34.1161 | -2.2002 | -2.2050 |
76
- | 1.4312 | 2.86 | 1100 | 5.9905 | -0.0144 | -0.0409 | 0.5860 | 0.0264 | -37.7210 | -34.1067 | -2.1989 | -2.2038 |
77
- | 1.0133 | 3.12 | 1200 | 6.1103 | -0.0167 | -0.0396 | 0.5889 | 0.0229 | -37.7146 | -34.1182 | -2.2000 | -2.2048 |
78
- | 0.5152 | 3.38 | 1300 | 6.0578 | -0.0132 | -0.0383 | 0.5544 | 0.0251 | -37.7080 | -34.1004 | -2.2003 | -2.2051 |
79
- | 0.8378 | 3.64 | 1400 | 6.0572 | -0.0138 | -0.0389 | 0.5748 | 0.0251 | -37.7113 | -34.1035 | -2.2004 | -2.2052 |
80
- | 0.9599 | 3.9 | 1500 | 6.0348 | -0.0125 | -0.0385 | 0.5835 | 0.0260 | -37.7091 | -34.0972 | -2.2004 | -2.2052 |
81
 
82
 
83
  ### Framework versions
 
20
 
21
  This model is a fine-tuned version of [data/ap-gpt-j-6b-sft-qlora-04-08](https://huggingface.co/data/ap-gpt-j-6b-sft-qlora-04-08) on the hugodk-sch/aftonposten_title_prefs dataset.
22
  It achieves the following results on the evaluation set:
23
+ - Loss: 0.9927
24
+ - Rewards/chosen: -0.0011
25
+ - Rewards/rejected: -0.0084
26
+ - Rewards/accuracies: 0.5536
27
+ - Rewards/margins: 0.0073
28
+ - Logps/rejected: -37.5587
29
+ - Logps/chosen: -34.0400
30
+ - Logits/rejected: -2.2193
31
+ - Logits/chosen: -2.2241
32
 
33
  ## Model description
34
 
 
57
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
58
  - lr_scheduler_type: cosine
59
  - lr_scheduler_warmup_ratio: 0.1
60
+ - num_epochs: 1
61
 
62
  ### Training results
63
 
64
+ | Training Loss | Epoch | Step | Validation Loss | Rewards/chosen | Rewards/rejected | Rewards/accuracies | Rewards/margins | Logps/rejected | Logps/chosen | Logits/rejected | Logits/chosen |
65
+ |:-------------:|:-----:|:----:|:---------------:|:--------------:|:----------------:|:------------------:|:---------------:|:--------------:|:------------:|:---------------:|:-------------:|
66
+ | 0.9701 | 0.26 | 100 | 0.9947 | 0.0025 | -0.0029 | 0.5311 | 0.0054 | -37.5310 | -34.0218 | -2.2257 | -2.2305 |
67
+ | 0.9192 | 0.52 | 200 | 0.9888 | 0.0055 | -0.0057 | 0.5336 | 0.0112 | -37.5450 | -34.0069 | -2.2237 | -2.2285 |
68
+ | 0.8593 | 0.78 | 300 | 0.9896 | -0.0002 | -0.0105 | 0.5652 | 0.0103 | -37.5692 | -34.0354 | -2.2191 | -2.2239 |
 
 
 
 
 
 
 
 
 
 
 
 
69
 
70
 
71
  ### Framework versions
adapter_config.json CHANGED
@@ -19,13 +19,13 @@
19
  "rank_pattern": {},
20
  "revision": null,
21
  "target_modules": [
 
22
  "v_proj",
23
- "down_proj",
24
  "o_proj",
25
  "q_proj",
26
  "k_proj",
27
- "gate_proj",
28
- "up_proj"
29
  ],
30
  "task_type": "CAUSAL_LM",
31
  "use_rslora": false
 
19
  "rank_pattern": {},
20
  "revision": null,
21
  "target_modules": [
22
+ "gate_proj",
23
  "v_proj",
 
24
  "o_proj",
25
  "q_proj",
26
  "k_proj",
27
+ "up_proj",
28
+ "down_proj"
29
  ],
30
  "task_type": "CAUSAL_LM",
31
  "use_rslora": false
adapter_model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:5fe0989cf18557a7daa207439d8382e361ceff43a9e66402269ceda7d71f2a3e
3
  size 176183216
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:32d56e589aacd9a87f0a6816e32fcc42fac42b3f4a716d9e492440497cd2cac1
3
  size 176183216
all_results.json CHANGED
@@ -1,21 +1,21 @@
1
  {
2
- "epoch": 4.0,
3
- "eval_logits/chosen": -2.205192804336548,
4
- "eval_logits/rejected": -2.20035719871521,
5
- "eval_logps/chosen": -34.1010856628418,
6
- "eval_logps/rejected": -37.70829772949219,
7
- "eval_loss": 6.077866077423096,
8
- "eval_rewards/accuracies": 0.5714285373687744,
9
- "eval_rewards/chosen": -0.013306654058396816,
10
- "eval_rewards/margins": 0.025028489530086517,
11
- "eval_rewards/rejected": -0.03833514451980591,
12
- "eval_runtime": 145.6887,
13
  "eval_samples": 343,
14
  "eval_samples_per_second": 2.354,
15
  "eval_steps_per_second": 0.295,
16
- "train_loss": 1.3887401007986688,
17
- "train_runtime": 10806.9485,
18
  "train_samples": 3079,
19
- "train_samples_per_second": 1.14,
20
- "train_steps_per_second": 0.143
21
  }
 
1
  {
2
+ "epoch": 1.0,
3
+ "eval_logits/chosen": -2.2240829467773438,
4
+ "eval_logits/rejected": -2.2192695140838623,
5
+ "eval_logps/chosen": -34.03996658325195,
6
+ "eval_logps/rejected": -37.558677673339844,
7
+ "eval_loss": 0.9926707744598389,
8
+ "eval_rewards/accuracies": 0.5535714626312256,
9
+ "eval_rewards/chosen": -0.0010830374667420983,
10
+ "eval_rewards/margins": 0.0073289512656629086,
11
+ "eval_rewards/rejected": -0.008411988615989685,
12
+ "eval_runtime": 145.7353,
13
  "eval_samples": 343,
14
  "eval_samples_per_second": 2.354,
15
  "eval_steps_per_second": 0.295,
16
+ "train_loss": 0.9474717462217653,
17
+ "train_runtime": 3250.5705,
18
  "train_samples": 3079,
19
+ "train_samples_per_second": 0.947,
20
+ "train_steps_per_second": 0.118
21
  }
eval_results.json CHANGED
@@ -1,15 +1,15 @@
1
  {
2
- "epoch": 4.0,
3
- "eval_logits/chosen": -2.205192804336548,
4
- "eval_logits/rejected": -2.20035719871521,
5
- "eval_logps/chosen": -34.1010856628418,
6
- "eval_logps/rejected": -37.70829772949219,
7
- "eval_loss": 6.077866077423096,
8
- "eval_rewards/accuracies": 0.5714285373687744,
9
- "eval_rewards/chosen": -0.013306654058396816,
10
- "eval_rewards/margins": 0.025028489530086517,
11
- "eval_rewards/rejected": -0.03833514451980591,
12
- "eval_runtime": 145.6887,
13
  "eval_samples": 343,
14
  "eval_samples_per_second": 2.354,
15
  "eval_steps_per_second": 0.295
 
1
  {
2
+ "epoch": 1.0,
3
+ "eval_logits/chosen": -2.2240829467773438,
4
+ "eval_logits/rejected": -2.2192695140838623,
5
+ "eval_logps/chosen": -34.03996658325195,
6
+ "eval_logps/rejected": -37.558677673339844,
7
+ "eval_loss": 0.9926707744598389,
8
+ "eval_rewards/accuracies": 0.5535714626312256,
9
+ "eval_rewards/chosen": -0.0010830374667420983,
10
+ "eval_rewards/margins": 0.0073289512656629086,
11
+ "eval_rewards/rejected": -0.008411988615989685,
12
+ "eval_runtime": 145.7353,
13
  "eval_samples": 343,
14
  "eval_samples_per_second": 2.354,
15
  "eval_steps_per_second": 0.295
train_results.json CHANGED
@@ -1,8 +1,8 @@
1
  {
2
- "epoch": 4.0,
3
- "train_loss": 1.3887401007986688,
4
- "train_runtime": 10806.9485,
5
  "train_samples": 3079,
6
- "train_samples_per_second": 1.14,
7
- "train_steps_per_second": 0.143
8
  }
 
1
  {
2
+ "epoch": 1.0,
3
+ "train_loss": 0.9474717462217653,
4
+ "train_runtime": 3250.5705,
5
  "train_samples": 3079,
6
+ "train_samples_per_second": 0.947,
7
+ "train_steps_per_second": 0.118
8
  }
trainer_state.json CHANGED
The diff for this file is too large to render. See raw diff
 
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:7a62740126001c4e37753bce979b61832119e1036635abceee907dccff0b6baa
3
  size 4984
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:14d8d82d267f6d519e0b2cb4934be36e24c3477dcb3ba98069a4b273db07577a
3
  size 4984