Model save
Browse files- README.md +14 -12
- all_results.json +5 -5
- train_results.json +5 -5
- trainer_state.json +0 -0
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 [alignment-handbook/zephyr-7b-sft-full](https://huggingface.co/alignment-handbook/zephyr-7b-sft-full) on the None dataset.
|
20 |
It achieves the following results on the evaluation set:
|
21 |
-
- Loss: 0.
|
22 |
-
- Rewards/chosen:
|
23 |
-
- Rewards/rejected: -
|
24 |
-
- Rewards/accuracies: 0.
|
25 |
-
- Rewards/margins:
|
26 |
-
- Logps/rejected: -
|
27 |
-
- Logps/chosen: -
|
28 |
-
- Logits/rejected: -2.
|
29 |
-
- Logits/chosen: -2.
|
30 |
|
31 |
## Model description
|
32 |
|
@@ -45,13 +45,15 @@ More information needed
|
|
45 |
### Training hyperparameters
|
46 |
|
47 |
The following hyperparameters were used during training:
|
48 |
-
- learning_rate: 5e-
|
49 |
-
- train_batch_size:
|
50 |
- eval_batch_size: 8
|
51 |
- seed: 42
|
52 |
- distributed_type: multi-GPU
|
|
|
53 |
- gradient_accumulation_steps: 4
|
54 |
- total_train_batch_size: 16
|
|
|
55 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
56 |
- lr_scheduler_type: cosine
|
57 |
- lr_scheduler_warmup_ratio: 0.1
|
@@ -61,7 +63,7 @@ The following hyperparameters were used during training:
|
|
61 |
|
62 |
| Training Loss | Epoch | Step | Validation Loss | Rewards/chosen | Rewards/rejected | Rewards/accuracies | Rewards/margins | Logps/rejected | Logps/chosen | Logits/rejected | Logits/chosen |
|
63 |
|:-------------:|:-----:|:----:|:---------------:|:--------------:|:----------------:|:------------------:|:---------------:|:--------------:|:------------:|:---------------:|:-------------:|
|
64 |
-
| 0.
|
65 |
|
66 |
|
67 |
### Framework versions
|
|
|
18 |
|
19 |
This model is a fine-tuned version of [alignment-handbook/zephyr-7b-sft-full](https://huggingface.co/alignment-handbook/zephyr-7b-sft-full) on the None dataset.
|
20 |
It achieves the following results on the evaluation set:
|
21 |
+
- Loss: 0.3553
|
22 |
+
- Rewards/chosen: -6.3145
|
23 |
+
- Rewards/rejected: -9.0890
|
24 |
+
- Rewards/accuracies: 0.8223
|
25 |
+
- Rewards/margins: 2.7746
|
26 |
+
- Logps/rejected: -1238.5492
|
27 |
+
- Logps/chosen: -1017.7702
|
28 |
+
- Logits/rejected: -2.0087
|
29 |
+
- Logits/chosen: -2.0557
|
30 |
|
31 |
## Model description
|
32 |
|
|
|
45 |
### Training hyperparameters
|
46 |
|
47 |
The following hyperparameters were used during training:
|
48 |
+
- learning_rate: 5e-06
|
49 |
+
- train_batch_size: 2
|
50 |
- eval_batch_size: 8
|
51 |
- seed: 42
|
52 |
- distributed_type: multi-GPU
|
53 |
+
- num_devices: 2
|
54 |
- gradient_accumulation_steps: 4
|
55 |
- total_train_batch_size: 16
|
56 |
+
- total_eval_batch_size: 16
|
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
|
|
|
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 |
+
| 0.3948 | 1.0 | 9121 | 0.3553 | -6.3145 | -9.0890 | 0.8223 | 2.7746 | -1238.5492 | -1017.7702 | -2.0087 | -2.0557 |
|
67 |
|
68 |
|
69 |
### Framework versions
|
all_results.json
CHANGED
@@ -1,8 +1,8 @@
|
|
1 |
{
|
2 |
"epoch": 1.0,
|
3 |
-
"train_loss": 0.
|
4 |
-
"train_runtime":
|
5 |
-
"train_samples":
|
6 |
-
"train_samples_per_second": 1.
|
7 |
-
"train_steps_per_second": 0.
|
8 |
}
|
|
|
1 |
{
|
2 |
"epoch": 1.0,
|
3 |
+
"train_loss": 0.35941912531329934,
|
4 |
+
"train_runtime": 141935.9761,
|
5 |
+
"train_samples": 145937,
|
6 |
+
"train_samples_per_second": 1.028,
|
7 |
+
"train_steps_per_second": 0.064
|
8 |
}
|
train_results.json
CHANGED
@@ -1,8 +1,8 @@
|
|
1 |
{
|
2 |
"epoch": 1.0,
|
3 |
-
"train_loss": 0.
|
4 |
-
"train_runtime":
|
5 |
-
"train_samples":
|
6 |
-
"train_samples_per_second": 1.
|
7 |
-
"train_steps_per_second": 0.
|
8 |
}
|
|
|
1 |
{
|
2 |
"epoch": 1.0,
|
3 |
+
"train_loss": 0.35941912531329934,
|
4 |
+
"train_runtime": 141935.9761,
|
5 |
+
"train_samples": 145937,
|
6 |
+
"train_samples_per_second": 1.028,
|
7 |
+
"train_steps_per_second": 0.064
|
8 |
}
|
trainer_state.json
CHANGED
The diff for this file is too large to render.
See raw diff
|
|