Commit
•
e2ce8fb
1
Parent(s):
27a95c7
End of training
Browse files
README.md
CHANGED
@@ -18,15 +18,15 @@ should probably proofread and complete it, then remove this comment. -->
|
|
18 |
|
19 |
This model is a fine-tuned version of [meta-llama/Llama-2-7b-hf](https://huggingface.co/meta-llama/Llama-2-7b-hf) on an unknown dataset.
|
20 |
It achieves the following results on the evaluation set:
|
21 |
-
- Loss: 0.
|
22 |
-
- Rewards/chosen: 0.
|
23 |
-
- Rewards/rejected: -
|
24 |
- Rewards/accuracies: 1.0
|
25 |
-
- Rewards/margins: 2.
|
26 |
-
- Logps/rejected: -
|
27 |
-
- Logps/chosen: -156.
|
28 |
-
- Logits/rejected: -1.
|
29 |
-
- Logits/chosen: -0.
|
30 |
|
31 |
## Model description
|
32 |
|
@@ -45,7 +45,7 @@ More information needed
|
|
45 |
### Training hyperparameters
|
46 |
|
47 |
The following hyperparameters were used during training:
|
48 |
-
- learning_rate:
|
49 |
- train_batch_size: 1
|
50 |
- eval_batch_size: 1
|
51 |
- seed: 42
|
@@ -53,42 +53,32 @@ The following hyperparameters were used during training:
|
|
53 |
- total_train_batch_size: 8
|
54 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
55 |
- lr_scheduler_type: linear
|
56 |
-
- num_epochs:
|
57 |
|
58 |
### Training results
|
59 |
|
60 |
| Training Loss | Epoch | Step | Validation Loss | Rewards/chosen | Rewards/rejected | Rewards/accuracies | Rewards/margins | Logps/rejected | Logps/chosen | Logits/rejected | Logits/chosen |
|
61 |
|:-------------:|:-----:|:----:|:---------------:|:--------------:|:----------------:|:------------------:|:---------------:|:--------------:|:------------:|:---------------:|:-------------:|
|
62 |
-
| 0.
|
63 |
-
| 0.
|
64 |
-
| 0.
|
65 |
-
| 0.
|
66 |
-
| 0.
|
67 |
-
| 0.
|
68 |
-
| 0.
|
69 |
-
| 0.
|
70 |
-
| 0.
|
71 |
-
| 0.
|
72 |
-
| 0.
|
73 |
-
| 0.
|
74 |
-
| 0.
|
75 |
-
| 0.
|
76 |
-
| 0.
|
77 |
-
| 0.
|
78 |
-
| 0.
|
79 |
-
| 0.
|
80 |
-
| 0.
|
81 |
-
| 0.
|
82 |
-
| 0.0961 | 2.09 | 525 | 0.1002 | 0.4227 | -1.9798 | 1.0 | 2.4025 | -204.6080 | -156.4070 | -1.0518 | -0.8587 |
|
83 |
-
| 0.0932 | 2.19 | 550 | 0.1000 | 0.4237 | -1.9796 | 1.0 | 2.4033 | -204.6052 | -156.3964 | -1.0518 | -0.8597 |
|
84 |
-
| 0.0901 | 2.29 | 575 | 0.1002 | 0.4231 | -1.9785 | 1.0 | 2.4015 | -204.5942 | -156.4030 | -1.0514 | -0.8594 |
|
85 |
-
| 0.1033 | 2.38 | 600 | 0.1003 | 0.4248 | -1.9780 | 1.0 | 2.4028 | -204.5901 | -156.3859 | -1.0517 | -0.8616 |
|
86 |
-
| 0.1108 | 2.48 | 625 | 0.0999 | 0.4262 | -1.9796 | 1.0 | 2.4057 | -204.6053 | -156.3723 | -1.0517 | -0.8583 |
|
87 |
-
| 0.1026 | 2.58 | 650 | 0.0998 | 0.4208 | -1.9879 | 1.0 | 2.4088 | -204.6889 | -156.4255 | -1.0522 | -0.8594 |
|
88 |
-
| 0.0956 | 2.68 | 675 | 0.1001 | 0.4227 | -1.9818 | 1.0 | 2.4045 | -204.6279 | -156.4070 | -1.0517 | -0.8588 |
|
89 |
-
| 0.1003 | 2.78 | 700 | 0.0996 | 0.4241 | -1.9817 | 1.0 | 2.4058 | -204.6262 | -156.3926 | -1.0516 | -0.8584 |
|
90 |
-
| 0.0874 | 2.88 | 725 | 0.0997 | 0.4228 | -1.9835 | 1.0 | 2.4064 | -204.6450 | -156.4057 | -1.0519 | -0.8609 |
|
91 |
-
| 0.1001 | 2.98 | 750 | 0.1001 | 0.4226 | -1.9804 | 1.0 | 2.4030 | -204.6132 | -156.4080 | -1.0519 | -0.8585 |
|
92 |
|
93 |
|
94 |
### Framework versions
|
|
|
18 |
|
19 |
This model is a fine-tuned version of [meta-llama/Llama-2-7b-hf](https://huggingface.co/meta-llama/Llama-2-7b-hf) on an unknown dataset.
|
20 |
It achieves the following results on the evaluation set:
|
21 |
+
- Loss: 0.0896
|
22 |
+
- Rewards/chosen: 0.4401
|
23 |
+
- Rewards/rejected: -2.0930
|
24 |
- Rewards/accuracies: 1.0
|
25 |
+
- Rewards/margins: 2.5330
|
26 |
+
- Logps/rejected: -205.7391
|
27 |
+
- Logps/chosen: -156.2334
|
28 |
+
- Logits/rejected: -1.0514
|
29 |
+
- Logits/chosen: -0.8587
|
30 |
|
31 |
## Model description
|
32 |
|
|
|
45 |
### Training hyperparameters
|
46 |
|
47 |
The following hyperparameters were used during training:
|
48 |
+
- learning_rate: 8e-07
|
49 |
- train_batch_size: 1
|
50 |
- eval_batch_size: 1
|
51 |
- seed: 42
|
|
|
53 |
- total_train_batch_size: 8
|
54 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
55 |
- lr_scheduler_type: linear
|
56 |
+
- num_epochs: 2
|
57 |
|
58 |
### Training results
|
59 |
|
60 |
| Training Loss | Epoch | Step | Validation Loss | Rewards/chosen | Rewards/rejected | Rewards/accuracies | Rewards/margins | Logps/rejected | Logps/chosen | Logits/rejected | Logits/chosen |
|
61 |
|:-------------:|:-----:|:----:|:---------------:|:--------------:|:----------------:|:------------------:|:---------------:|:--------------:|:------------:|:---------------:|:-------------:|
|
62 |
+
| 0.6699 | 0.1 | 25 | 0.6428 | 0.0307 | -0.0744 | 0.9033 | 0.1051 | -185.5532 | -160.3267 | -1.0520 | -0.8550 |
|
63 |
+
| 0.5702 | 0.2 | 50 | 0.5471 | 0.0866 | -0.2359 | 0.9933 | 0.3225 | -187.1690 | -159.7680 | -1.0514 | -0.8544 |
|
64 |
+
| 0.488 | 0.3 | 75 | 0.4456 | 0.1502 | -0.4424 | 1.0 | 0.5926 | -189.2334 | -159.1314 | -1.0527 | -0.8555 |
|
65 |
+
| 0.3957 | 0.4 | 100 | 0.3600 | 0.2054 | -0.6615 | 1.0 | 0.8669 | -191.4245 | -158.5795 | -1.0530 | -0.8577 |
|
66 |
+
| 0.3338 | 0.5 | 125 | 0.2865 | 0.2569 | -0.8933 | 1.0 | 1.1502 | -193.7425 | -158.0646 | -1.0524 | -0.8564 |
|
67 |
+
| 0.253 | 0.6 | 150 | 0.2257 | 0.3043 | -1.1373 | 1.0 | 1.4416 | -196.1830 | -157.5914 | -1.0523 | -0.8570 |
|
68 |
+
| 0.2134 | 0.7 | 175 | 0.1819 | 0.3496 | -1.3537 | 1.0 | 1.7033 | -198.3466 | -157.1379 | -1.0530 | -0.8584 |
|
69 |
+
| 0.1613 | 0.79 | 200 | 0.1473 | 0.3842 | -1.5693 | 1.0 | 1.9535 | -200.5027 | -156.7917 | -1.0525 | -0.8591 |
|
70 |
+
| 0.1358 | 0.89 | 225 | 0.1231 | 0.4031 | -1.7582 | 1.0 | 2.1614 | -202.3919 | -156.6024 | -1.0523 | -0.8593 |
|
71 |
+
| 0.115 | 0.99 | 250 | 0.1076 | 0.4205 | -1.8980 | 1.0 | 2.3185 | -203.7897 | -156.4292 | -1.0521 | -0.8590 |
|
72 |
+
| 0.1111 | 1.09 | 275 | 0.0989 | 0.4291 | -1.9856 | 1.0 | 2.4148 | -204.6660 | -156.3426 | -1.0515 | -0.8591 |
|
73 |
+
| 0.0902 | 1.19 | 300 | 0.0949 | 0.4280 | -2.0337 | 1.0 | 2.4617 | -205.1465 | -156.3540 | -1.0507 | -0.8576 |
|
74 |
+
| 0.0867 | 1.29 | 325 | 0.0920 | 0.4325 | -2.0705 | 1.0 | 2.5030 | -205.5146 | -156.3087 | -1.0510 | -0.8576 |
|
75 |
+
| 0.0973 | 1.39 | 350 | 0.0905 | 0.4357 | -2.0839 | 1.0 | 2.5196 | -205.6485 | -156.2766 | -1.0506 | -0.8576 |
|
76 |
+
| 0.0942 | 1.49 | 375 | 0.0897 | 0.4422 | -2.0838 | 1.0 | 2.5260 | -205.6476 | -156.2122 | -1.0515 | -0.8578 |
|
77 |
+
| 0.0858 | 1.59 | 400 | 0.0897 | 0.4392 | -2.0903 | 1.0 | 2.5295 | -205.7121 | -156.2415 | -1.0515 | -0.8587 |
|
78 |
+
| 0.083 | 1.69 | 425 | 0.0893 | 0.4401 | -2.0972 | 1.0 | 2.5373 | -205.7811 | -156.2327 | -1.0511 | -0.8584 |
|
79 |
+
| 0.0964 | 1.79 | 450 | 0.0897 | 0.4368 | -2.0947 | 1.0 | 2.5315 | -205.7564 | -156.2662 | -1.0511 | -0.8577 |
|
80 |
+
| 0.0931 | 1.89 | 475 | 0.0890 | 0.4406 | -2.0970 | 1.0 | 2.5376 | -205.7794 | -156.2282 | -1.0512 | -0.8585 |
|
81 |
+
| 0.0915 | 1.99 | 500 | 0.0896 | 0.4401 | -2.0930 | 1.0 | 2.5330 | -205.7391 | -156.2334 | -1.0514 | -0.8587 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
82 |
|
83 |
|
84 |
### Framework versions
|