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
3ccf294
•
1 Parent(s): 4a53a2d

Training in progress, step 500

Browse files
Files changed (8) hide show
  1. README.md +15 -27
  2. adapter_config.json +5 -5
  3. adapter_model.safetensors +1 -1
  4. all_results.json +16 -16
  5. eval_results.json +12 -12
  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: 0.4730
24
- - Rewards/chosen: 0.2464
25
- - Rewards/rejected: 0.1175
26
- - Rewards/accuracies: 0.5947
27
- - Rewards/margins: 0.1288
28
- - Logps/rejected: -37.3207
29
- - Logps/chosen: -33.6239
30
- - Logits/rejected: -2.1470
31
- - Logits/chosen: -2.1517
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
- | 0.4799 | 0.26 | 100 | -2.2349 | -2.2301 | -33.9029 | -37.3943 | 0.4987 | 0.5104 | 0.0790 | 0.0056 | 0.0734 |
67
- | 0.4548 | 0.52 | 200 | -2.2335 | -2.2287 | -33.7696 | -37.2846 | 0.4956 | 0.5341 | 0.1590 | 0.0198 | 0.1392 |
68
- | 0.41 | 0.78 | 300 | -2.2309 | -2.2261 | -33.7614 | -37.2848 | 0.4937 | 0.5361 | 0.1639 | 0.0248 | 0.1391 |
69
- | 0.3497 | 1.04 | 400 | 0.4927 | 0.2171 | 0.1863 | 0.5652 | 0.0309 | -37.2062 | -33.6727 | -2.2113 | -2.2162 |
70
- | 0.2906 | 1.3 | 500 | 0.4870 | 0.2484 | 0.1921 | 0.5922 | 0.0563 | -37.1964 | -33.6205 | -2.1834 | -2.1881 |
71
- | 0.3014 | 1.56 | 600 | 0.4796 | 0.2630 | 0.1719 | 0.5797 | 0.0911 | -37.2301 | -33.5962 | -2.1694 | -2.1741 |
72
- | 0.2776 | 1.82 | 700 | 0.4825 | 0.2341 | 0.1554 | 0.5768 | 0.0787 | -37.2576 | -33.6444 | -2.1625 | -2.1672 |
73
- | 0.201 | 2.08 | 800 | 0.4766 | 0.2639 | 0.1595 | 0.5914 | 0.1043 | -37.2507 | -33.5948 | -2.1641 | -2.1689 |
74
- | 0.1721 | 2.34 | 900 | 0.4749 | 0.2446 | 0.1298 | 0.5860 | 0.1148 | -37.3003 | -33.6269 | -2.1516 | -2.1563 |
75
- | 0.2259 | 2.6 | 1000 | 0.4736 | 0.2483 | 0.1257 | 0.5860 | 0.1226 | -37.3072 | -33.6207 | -2.1481 | -2.1528 |
76
- | 0.2405 | 2.86 | 1100 | 0.4740 | 0.2438 | 0.1229 | 0.5860 | 0.1209 | -37.3118 | -33.6283 | -2.1475 | -2.1522 |
77
- | 0.1793 | 3.12 | 1200 | 0.4746 | 0.2441 | 0.1249 | 0.5685 | 0.1192 | -37.3085 | -33.6277 | -2.1469 | -2.1516 |
78
- | 0.1633 | 3.38 | 1300 | 0.4744 | 0.2433 | 0.1235 | 0.6009 | 0.1198 | -37.3107 | -33.6290 | -2.1471 | -2.1518 |
79
- | 0.202 | 3.64 | 1400 | 0.4748 | 0.2450 | 0.1279 | 0.5831 | 0.1170 | -37.3034 | -33.6263 | -2.1472 | -2.1519 |
80
- | 0.1889 | 3.9 | 1500 | 0.4727 | 0.2480 | 0.1188 | 0.6005 | 0.1292 | -37.3186 | -33.6212 | -2.1470 | -2.1517 |
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: 0.7598
24
+ - Rewards/chosen: -0.0165
25
+ - Rewards/rejected: -0.0300
26
+ - Rewards/accuracies: 0.5307
27
+ - Rewards/margins: 0.0135
28
+ - Logps/rejected: -37.5666
29
+ - Logps/chosen: -34.0620
30
+ - Logits/rejected: -2.2283
31
+ - Logits/chosen: -2.2332
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
+ | 0.5903 | 0.26 | 100 | 0.7450 | 0.0123 | 0.0047 | 0.5241 | 0.0075 | -37.5087 | -34.0141 | -2.2318 | -2.2366 |
67
+ | 0.59 | 0.52 | 200 | 0.7500 | -0.0208 | -0.0352 | 0.5365 | 0.0144 | -37.5752 | -34.0691 | -2.2284 | -2.2332 |
68
+ | 0.3871 | 0.78 | 300 | 0.7655 | -0.0166 | -0.0295 | 0.5071 | 0.0129 | -37.5657 | -34.0622 | -2.2285 | -2.2333 |
 
 
 
 
 
 
 
 
 
 
 
 
69
 
70
 
71
  ### Framework versions
adapter_config.json CHANGED
@@ -19,13 +19,13 @@
19
  "rank_pattern": {},
20
  "revision": null,
21
  "target_modules": [
22
- "gate_proj",
 
23
  "up_proj",
 
 
24
  "k_proj",
25
- "down_proj",
26
- "o_proj",
27
- "v_proj",
28
- "q_proj"
29
  ],
30
  "task_type": "CAUSAL_LM",
31
  "use_rslora": false
 
19
  "rank_pattern": {},
20
  "revision": null,
21
  "target_modules": [
22
+ "v_proj",
23
+ "down_proj",
24
  "up_proj",
25
+ "gate_proj",
26
+ "q_proj",
27
  "k_proj",
28
+ "o_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 sha256:055ffd1fcf3c15abaeb1e658da1ca3ca33d4856bb69ba7776358c3487a3e5b13
3
  size 176183216
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:647a30127ff87e6873207e0e6886687457d6421bb849937f43670e0dc30d2513
3
  size 176183216
all_results.json CHANGED
@@ -1,21 +1,21 @@
1
  {
2
- "epoch": 4.0,
3
- "eval_logits/chosen": -2.1517372131347656,
4
- "eval_logits/rejected": -2.1470396518707275,
5
- "eval_logps/chosen": -33.62392044067383,
6
- "eval_logps/rejected": -37.32072830200195,
7
- "eval_loss": 0.47301676869392395,
8
- "eval_rewards/accuracies": 0.594684362411499,
9
- "eval_rewards/chosen": 0.24637830257415771,
10
- "eval_rewards/margins": 0.1288420557975769,
11
- "eval_rewards/rejected": 0.11753623187541962,
12
- "eval_runtime": 145.7991,
13
  "eval_samples": 343,
14
- "eval_samples_per_second": 2.353,
15
  "eval_steps_per_second": 0.295,
16
- "train_loss": 0.19402528018146367,
17
- "train_runtime": 10806.8766,
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": -2.2331926822662354,
4
+ "eval_logits/rejected": -2.2283315658569336,
5
+ "eval_logps/chosen": -34.061973571777344,
6
+ "eval_logps/rejected": -37.566593170166016,
7
+ "eval_loss": 0.7597638368606567,
8
+ "eval_rewards/accuracies": 0.5307309031486511,
9
+ "eval_rewards/chosen": -0.016451681032776833,
10
+ "eval_rewards/margins": 0.013532036915421486,
11
+ "eval_rewards/rejected": -0.029983719810843468,
12
+ "eval_runtime": 145.5395,
13
  "eval_samples": 343,
14
+ "eval_samples_per_second": 2.357,
15
  "eval_steps_per_second": 0.295,
16
+ "train_loss": 0.5965218513042896,
17
+ "train_runtime": 3252.2803,
18
  "train_samples": 3079,
19
+ "train_samples_per_second": 0.947,
20
+ "train_steps_per_second": 0.118
21
  }
eval_results.json CHANGED
@@ -1,16 +1,16 @@
1
  {
2
- "epoch": 4.0,
3
- "eval_logits/chosen": -2.1517372131347656,
4
- "eval_logits/rejected": -2.1470396518707275,
5
- "eval_logps/chosen": -33.62392044067383,
6
- "eval_logps/rejected": -37.32072830200195,
7
- "eval_loss": 0.47301676869392395,
8
- "eval_rewards/accuracies": 0.594684362411499,
9
- "eval_rewards/chosen": 0.24637830257415771,
10
- "eval_rewards/margins": 0.1288420557975769,
11
- "eval_rewards/rejected": 0.11753623187541962,
12
- "eval_runtime": 145.7991,
13
  "eval_samples": 343,
14
- "eval_samples_per_second": 2.353,
15
  "eval_steps_per_second": 0.295
16
  }
 
1
  {
2
+ "epoch": 1.0,
3
+ "eval_logits/chosen": -2.2331926822662354,
4
+ "eval_logits/rejected": -2.2283315658569336,
5
+ "eval_logps/chosen": -34.061973571777344,
6
+ "eval_logps/rejected": -37.566593170166016,
7
+ "eval_loss": 0.7597638368606567,
8
+ "eval_rewards/accuracies": 0.5307309031486511,
9
+ "eval_rewards/chosen": -0.016451681032776833,
10
+ "eval_rewards/margins": 0.013532036915421486,
11
+ "eval_rewards/rejected": -0.029983719810843468,
12
+ "eval_runtime": 145.5395,
13
  "eval_samples": 343,
14
+ "eval_samples_per_second": 2.357,
15
  "eval_steps_per_second": 0.295
16
  }
train_results.json CHANGED
@@ -1,8 +1,8 @@
1
  {
2
- "epoch": 4.0,
3
- "train_loss": 0.19402528018146367,
4
- "train_runtime": 10806.8766,
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": 0.5965218513042896,
4
+ "train_runtime": 3252.2803,
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:a61cf4921e25061d4673043facf2e8060cc927f0981742727fad132386961410
3
  size 4984
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:eacae65050e9f803beda9fe6763ea7afd8ff2656b8fcda6cbf8438d16e92bb88
3
  size 4984