Hugging Face's logo

Litzy619
/
O0428B1

Safetensors
Generated from Trainer
Model card Files
xet
Community
Litzy619 commited on
Commit
addd19b
·
verified ·
1 Parent(s): 32636a8

End of training

Browse files
Files changed (5) hide show
  1. README.md +16 -38
  2. adapter_config.json +1 -1
  3. adapter_model.bin +1 -1
  4. model.safetensors +1 -1
  5. training_args.bin +1 -1
README.md CHANGED
@@ -15,7 +15,7 @@ should probably proofread and complete it, then remove this comment. -->
15
 
16
  This model is a fine-tuned version of [allenai/OLMo-1B](https://huggingface.co/allenai/OLMo-1B) on an unknown dataset.
17
  It achieves the following results on the evaluation set:
18
- - Loss: 0.1445
19
 
20
  ## Model description
21
 
@@ -35,54 +35,32 @@ More information needed
35
 
36
  The following hyperparameters were used during training:
37
  - learning_rate: 0.0003
38
- - train_batch_size: 8
39
  - eval_batch_size: 8
40
  - seed: 42
41
- - gradient_accumulation_steps: 16
42
  - total_train_batch_size: 128
43
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
44
  - lr_scheduler_type: cosine_with_restarts
45
- - lr_scheduler_warmup_steps: 80
46
- - num_epochs: 3
47
  - mixed_precision_training: Native AMP
48
 
49
  ### Training results
50
 
51
  | Training Loss | Epoch | Step | Validation Loss |
52
  |:-------------:|:-----:|:----:|:---------------:|
53
- | 1.1326 | 0.09 | 10 | 0.1682 |
54
- | 0.1661 | 0.18 | 20 | 0.1573 |
55
- | 0.1515 | 0.27 | 30 | 0.1546 |
56
- | 0.1535 | 0.36 | 40 | 0.1488 |
57
- | 0.1482 | 0.45 | 50 | 0.1474 |
58
- | 0.151 | 0.54 | 60 | 0.1485 |
59
- | 0.149 | 0.63 | 70 | 0.1463 |
60
- | 0.1493 | 0.73 | 80 | 0.1528 |
61
- | 0.1468 | 0.82 | 90 | 0.1505 |
62
- | 0.1506 | 0.91 | 100 | 0.1519 |
63
- | 0.1547 | 1.0 | 110 | 0.1511 |
64
- | 0.1469 | 1.09 | 120 | 0.1509 |
65
- | 0.1469 | 1.18 | 130 | 0.1503 |
66
- | 0.148 | 1.27 | 140 | 0.1475 |
67
- | 0.1506 | 1.36 | 150 | 0.1513 |
68
- | 0.1464 | 1.45 | 160 | 0.1491 |
69
- | 0.1463 | 1.54 | 170 | 0.1475 |
70
- | 0.1487 | 1.63 | 180 | 0.1471 |
71
- | 0.148 | 1.72 | 190 | 0.1499 |
72
- | 0.147 | 1.81 | 200 | 0.1559 |
73
- | 0.1509 | 1.9 | 210 | 0.1478 |
74
- | 0.1471 | 1.99 | 220 | 0.1493 |
75
- | 0.1471 | 2.08 | 230 | 0.1472 |
76
- | 0.1417 | 2.18 | 240 | 0.1478 |
77
- | 0.1436 | 2.27 | 250 | 0.1467 |
78
- | 0.1439 | 2.36 | 260 | 0.1486 |
79
- | 0.1423 | 2.45 | 270 | 0.1466 |
80
- | 0.1411 | 2.54 | 280 | 0.1462 |
81
- | 0.1404 | 2.63 | 290 | 0.1476 |
82
- | 0.1454 | 2.72 | 300 | 0.1445 |
83
- | 0.1437 | 2.81 | 310 | 0.1445 |
84
- | 0.143 | 2.9 | 320 | 0.1445 |
85
- | 0.1437 | 2.99 | 330 | 0.1445 |
86
 
87
 
88
  ### Framework versions
 
15
 
16
  This model is a fine-tuned version of [allenai/OLMo-1B](https://huggingface.co/allenai/OLMo-1B) on an unknown dataset.
17
  It achieves the following results on the evaluation set:
18
+ - Loss: 0.1469
19
 
20
  ## Model description
21
 
 
35
 
36
  The following hyperparameters were used during training:
37
  - learning_rate: 0.0003
38
+ - train_batch_size: 4
39
  - eval_batch_size: 8
40
  - seed: 42
41
+ - gradient_accumulation_steps: 32
42
  - total_train_batch_size: 128
43
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
44
  - lr_scheduler_type: cosine_with_restarts
45
+ - lr_scheduler_warmup_steps: 60
46
+ - num_epochs: 1
47
  - mixed_precision_training: Native AMP
48
 
49
  ### Training results
50
 
51
  | Training Loss | Epoch | Step | Validation Loss |
52
  |:-------------:|:-----:|:----:|:---------------:|
53
+ | 2.7159 | 0.09 | 10 | 1.9907 |
54
+ | 1.3938 | 0.18 | 20 | 0.5917 |
55
+ | 0.2804 | 0.27 | 30 | 0.1637 |
56
+ | 0.164 | 0.36 | 40 | 0.1546 |
57
+ | 0.1511 | 0.45 | 50 | 0.1503 |
58
+ | 0.1892 | 0.54 | 60 | 0.1511 |
59
+ | 0.1598 | 0.63 | 70 | 0.1480 |
60
+ | 0.1618 | 0.73 | 80 | 0.1511 |
61
+ | 0.1434 | 0.82 | 90 | 0.1488 |
62
+ | 0.1541 | 0.91 | 100 | 0.1474 |
63
+ | 0.1569 | 1.0 | 110 | 0.1469 |
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
64
 
65
 
66
  ### Framework versions
adapter_config.json CHANGED
@@ -10,7 +10,7 @@
10
  "merge_weights": false,
11
  "modules_to_save": null,
12
  "peft_type": "LORA",
13
- "r": 16,
14
  "target_modules": [
15
  "attn_out"
16
  ],
 
10
  "merge_weights": false,
11
  "modules_to_save": null,
12
  "peft_type": "LORA",
13
+ "r": 4,
14
  "target_modules": [
15
  "attn_out"
16
  ],
adapter_model.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:cc6f2f89b32b7f988941d9c166ce8e934b5bc65bfead7cb1c761460046df86e2
3
  size 2108942
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3314f3cf4c79b57cfeb24879832312048861adf577b0a429f1e773cd8990e514
3
  size 2108942
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:37e86e662663a14d75585bd327454652c0a07fb7be4a51526602035a45eb0bd3
3
  size 4709168176
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2d748465bf361ece564e0c9a4037abd5e0164639d348e9ca39de9cd37d884033
3
  size 4709168176
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:dc91770ebce394f275c10b6cae6cedc49b8a7828ddb51981aa26d86ce2e436bb
3
  size 5112
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:07fa8f30930fe61200434a5770d4ec8b32762775a0ef329e1d05f01043d2918a
3
  size 5112