Model save
Browse files- README.md +61 -0
- adapter_model.safetensors +1 -1
- all_results.json +8 -0
- runs/May21_07-22-36_gpu4-119-5/events.out.tfevents.1716240318.gpu4-119-5.1396954.0 +2 -2
- train_results.json +8 -0
- trainer_state.json +1346 -0
README.md
ADDED
@@ -0,0 +1,61 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
library_name: peft
|
3 |
+
tags:
|
4 |
+
- trl
|
5 |
+
- dpo
|
6 |
+
- generated_from_trainer
|
7 |
+
base_model: DUAL-GPO/phi-2-irepo-chatml-merged-i0
|
8 |
+
model-index:
|
9 |
+
- name: phi-2-irepo-chatml-v13-i1
|
10 |
+
results: []
|
11 |
+
---
|
12 |
+
|
13 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
14 |
+
should probably proofread and complete it, then remove this comment. -->
|
15 |
+
|
16 |
+
# phi-2-irepo-chatml-v13-i1
|
17 |
+
|
18 |
+
This model is a fine-tuned version of [DUAL-GPO/phi-2-irepo-chatml-merged-i0](https://huggingface.co/DUAL-GPO/phi-2-irepo-chatml-merged-i0) on the None dataset.
|
19 |
+
|
20 |
+
## Model description
|
21 |
+
|
22 |
+
More information needed
|
23 |
+
|
24 |
+
## Intended uses & limitations
|
25 |
+
|
26 |
+
More information needed
|
27 |
+
|
28 |
+
## Training and evaluation data
|
29 |
+
|
30 |
+
More information needed
|
31 |
+
|
32 |
+
## Training procedure
|
33 |
+
|
34 |
+
### Training hyperparameters
|
35 |
+
|
36 |
+
The following hyperparameters were used during training:
|
37 |
+
- learning_rate: 5e-06
|
38 |
+
- train_batch_size: 4
|
39 |
+
- eval_batch_size: 4
|
40 |
+
- seed: 42
|
41 |
+
- distributed_type: multi-GPU
|
42 |
+
- num_devices: 2
|
43 |
+
- gradient_accumulation_steps: 4
|
44 |
+
- total_train_batch_size: 32
|
45 |
+
- total_eval_batch_size: 8
|
46 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
47 |
+
- lr_scheduler_type: cosine
|
48 |
+
- lr_scheduler_warmup_ratio: 0.1
|
49 |
+
- num_epochs: 1
|
50 |
+
|
51 |
+
### Training results
|
52 |
+
|
53 |
+
|
54 |
+
|
55 |
+
### Framework versions
|
56 |
+
|
57 |
+
- PEFT 0.7.1
|
58 |
+
- Transformers 4.36.2
|
59 |
+
- Pytorch 2.1.2+cu121
|
60 |
+
- Datasets 2.14.6
|
61 |
+
- Tokenizers 0.15.2
|
adapter_model.safetensors
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 335579632
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:dda691ae9be5ac08a8df4d8407c1fe4d5870642f5d5344b6151a127d0037633e
|
3 |
size 335579632
|
all_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 1.0,
|
3 |
+
"train_loss": 0.274457171320788,
|
4 |
+
"train_runtime": 7836.9626,
|
5 |
+
"train_samples": 30000,
|
6 |
+
"train_samples_per_second": 3.828,
|
7 |
+
"train_steps_per_second": 0.12
|
8 |
+
}
|
runs/May21_07-22-36_gpu4-119-5/events.out.tfevents.1716240318.gpu4-119-5.1396954.0
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
-
size
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:5a8ef6341b0333588b573f8c9241bdfce50fd1611fe24b23a0e8813ec7d7ad55
|
3 |
+
size 64720
|
train_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 1.0,
|
3 |
+
"train_loss": 0.274457171320788,
|
4 |
+
"train_runtime": 7836.9626,
|
5 |
+
"train_samples": 30000,
|
6 |
+
"train_samples_per_second": 3.828,
|
7 |
+
"train_steps_per_second": 0.12
|
8 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,1346 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 0.9994666666666666,
|
5 |
+
"eval_steps": 500,
|
6 |
+
"global_step": 937,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.0,
|
13 |
+
"learning_rate": 5.319148936170213e-08,
|
14 |
+
"logits/chosen": 0.4053989052772522,
|
15 |
+
"logits/rejected": 0.1312936246395111,
|
16 |
+
"logps/chosen": -434.00537109375,
|
17 |
+
"logps/rejected": -516.5983276367188,
|
18 |
+
"loss": 0.309,
|
19 |
+
"rewards/accuracies": 0.0,
|
20 |
+
"rewards/chosen": 0.0,
|
21 |
+
"rewards/margins": 0.0,
|
22 |
+
"rewards/rejected": 0.0,
|
23 |
+
"step": 1
|
24 |
+
},
|
25 |
+
{
|
26 |
+
"epoch": 0.01,
|
27 |
+
"learning_rate": 5.319148936170213e-07,
|
28 |
+
"logits/chosen": 0.15412023663520813,
|
29 |
+
"logits/rejected": 0.2926546633243561,
|
30 |
+
"logps/chosen": -365.67779541015625,
|
31 |
+
"logps/rejected": -353.0097961425781,
|
32 |
+
"loss": 0.3148,
|
33 |
+
"rewards/accuracies": 0.2777777910232544,
|
34 |
+
"rewards/chosen": -0.0005839762161485851,
|
35 |
+
"rewards/margins": -9.90507487586001e-06,
|
36 |
+
"rewards/rejected": -0.0005740711931139231,
|
37 |
+
"step": 10
|
38 |
+
},
|
39 |
+
{
|
40 |
+
"epoch": 0.02,
|
41 |
+
"learning_rate": 1.0638297872340427e-06,
|
42 |
+
"logits/chosen": 0.12003284692764282,
|
43 |
+
"logits/rejected": 0.2044968158006668,
|
44 |
+
"logps/chosen": -340.22833251953125,
|
45 |
+
"logps/rejected": -348.37213134765625,
|
46 |
+
"loss": 0.3073,
|
47 |
+
"rewards/accuracies": 0.26249998807907104,
|
48 |
+
"rewards/chosen": -0.0007841844344511628,
|
49 |
+
"rewards/margins": -0.00017091934569180012,
|
50 |
+
"rewards/rejected": -0.0006132650305517018,
|
51 |
+
"step": 20
|
52 |
+
},
|
53 |
+
{
|
54 |
+
"epoch": 0.03,
|
55 |
+
"learning_rate": 1.595744680851064e-06,
|
56 |
+
"logits/chosen": 0.2551342844963074,
|
57 |
+
"logits/rejected": 0.2518610954284668,
|
58 |
+
"logps/chosen": -383.241455078125,
|
59 |
+
"logps/rejected": -364.0563049316406,
|
60 |
+
"loss": 0.2998,
|
61 |
+
"rewards/accuracies": 0.33125001192092896,
|
62 |
+
"rewards/chosen": -0.0006543128984048963,
|
63 |
+
"rewards/margins": 3.304009396742913e-06,
|
64 |
+
"rewards/rejected": -0.0006576168816536665,
|
65 |
+
"step": 30
|
66 |
+
},
|
67 |
+
{
|
68 |
+
"epoch": 0.04,
|
69 |
+
"learning_rate": 2.1276595744680853e-06,
|
70 |
+
"logits/chosen": 0.15533123910427094,
|
71 |
+
"logits/rejected": 0.27131038904190063,
|
72 |
+
"logps/chosen": -402.34442138671875,
|
73 |
+
"logps/rejected": -386.5889587402344,
|
74 |
+
"loss": 0.3147,
|
75 |
+
"rewards/accuracies": 0.32499998807907104,
|
76 |
+
"rewards/chosen": -0.001332725747488439,
|
77 |
+
"rewards/margins": -8.493573113810271e-05,
|
78 |
+
"rewards/rejected": -0.0012477899435907602,
|
79 |
+
"step": 40
|
80 |
+
},
|
81 |
+
{
|
82 |
+
"epoch": 0.05,
|
83 |
+
"learning_rate": 2.6595744680851065e-06,
|
84 |
+
"logits/chosen": 0.2343147248029709,
|
85 |
+
"logits/rejected": 0.27498680353164673,
|
86 |
+
"logps/chosen": -442.1997985839844,
|
87 |
+
"logps/rejected": -432.7005920410156,
|
88 |
+
"loss": 0.3086,
|
89 |
+
"rewards/accuracies": 0.4437499940395355,
|
90 |
+
"rewards/chosen": -0.0012903345050290227,
|
91 |
+
"rewards/margins": 0.0002948477049358189,
|
92 |
+
"rewards/rejected": -0.0015851823845878243,
|
93 |
+
"step": 50
|
94 |
+
},
|
95 |
+
{
|
96 |
+
"epoch": 0.06,
|
97 |
+
"learning_rate": 3.191489361702128e-06,
|
98 |
+
"logits/chosen": 0.1715986281633377,
|
99 |
+
"logits/rejected": 0.3192042112350464,
|
100 |
+
"logps/chosen": -409.2780456542969,
|
101 |
+
"logps/rejected": -413.3976135253906,
|
102 |
+
"loss": 0.311,
|
103 |
+
"rewards/accuracies": 0.4312500059604645,
|
104 |
+
"rewards/chosen": -0.00036035641096532345,
|
105 |
+
"rewards/margins": 0.0003330351901240647,
|
106 |
+
"rewards/rejected": -0.0006933916010893881,
|
107 |
+
"step": 60
|
108 |
+
},
|
109 |
+
{
|
110 |
+
"epoch": 0.07,
|
111 |
+
"learning_rate": 3.723404255319149e-06,
|
112 |
+
"logits/chosen": 0.1239662617444992,
|
113 |
+
"logits/rejected": 0.22276969254016876,
|
114 |
+
"logps/chosen": -356.0749206542969,
|
115 |
+
"logps/rejected": -358.4507751464844,
|
116 |
+
"loss": 0.3075,
|
117 |
+
"rewards/accuracies": 0.48750001192092896,
|
118 |
+
"rewards/chosen": -0.0014285400975495577,
|
119 |
+
"rewards/margins": 0.0010666290763765574,
|
120 |
+
"rewards/rejected": -0.002495169173926115,
|
121 |
+
"step": 70
|
122 |
+
},
|
123 |
+
{
|
124 |
+
"epoch": 0.09,
|
125 |
+
"learning_rate": 4.255319148936171e-06,
|
126 |
+
"logits/chosen": 0.14190419018268585,
|
127 |
+
"logits/rejected": 0.2491893321275711,
|
128 |
+
"logps/chosen": -399.1524963378906,
|
129 |
+
"logps/rejected": -387.684326171875,
|
130 |
+
"loss": 0.3052,
|
131 |
+
"rewards/accuracies": 0.48124998807907104,
|
132 |
+
"rewards/chosen": -0.00629749009385705,
|
133 |
+
"rewards/margins": 0.0015292944153770804,
|
134 |
+
"rewards/rejected": -0.007826784625649452,
|
135 |
+
"step": 80
|
136 |
+
},
|
137 |
+
{
|
138 |
+
"epoch": 0.1,
|
139 |
+
"learning_rate": 4.787234042553192e-06,
|
140 |
+
"logits/chosen": 0.23625846207141876,
|
141 |
+
"logits/rejected": 0.17152468860149384,
|
142 |
+
"logps/chosen": -408.34222412109375,
|
143 |
+
"logps/rejected": -431.87689208984375,
|
144 |
+
"loss": 0.3018,
|
145 |
+
"rewards/accuracies": 0.48750001192092896,
|
146 |
+
"rewards/chosen": -0.01540874969214201,
|
147 |
+
"rewards/margins": 0.0023751407861709595,
|
148 |
+
"rewards/rejected": -0.017783891409635544,
|
149 |
+
"step": 90
|
150 |
+
},
|
151 |
+
{
|
152 |
+
"epoch": 0.11,
|
153 |
+
"learning_rate": 4.999375059004058e-06,
|
154 |
+
"logits/chosen": 0.15845081210136414,
|
155 |
+
"logits/rejected": 0.16451677680015564,
|
156 |
+
"logps/chosen": -451.0520935058594,
|
157 |
+
"logps/rejected": -446.60052490234375,
|
158 |
+
"loss": 0.314,
|
159 |
+
"rewards/accuracies": 0.512499988079071,
|
160 |
+
"rewards/chosen": -0.03385753557085991,
|
161 |
+
"rewards/margins": 0.003108574775978923,
|
162 |
+
"rewards/rejected": -0.03696610778570175,
|
163 |
+
"step": 100
|
164 |
+
},
|
165 |
+
{
|
166 |
+
"epoch": 0.12,
|
167 |
+
"learning_rate": 4.9955571065548795e-06,
|
168 |
+
"logits/chosen": 0.2290416657924652,
|
169 |
+
"logits/rejected": 0.15046942234039307,
|
170 |
+
"logps/chosen": -470.06536865234375,
|
171 |
+
"logps/rejected": -457.4624938964844,
|
172 |
+
"loss": 0.3177,
|
173 |
+
"rewards/accuracies": 0.48750001192092896,
|
174 |
+
"rewards/chosen": -0.05096244812011719,
|
175 |
+
"rewards/margins": 0.006424424704164267,
|
176 |
+
"rewards/rejected": -0.057386863976716995,
|
177 |
+
"step": 110
|
178 |
+
},
|
179 |
+
{
|
180 |
+
"epoch": 0.13,
|
181 |
+
"learning_rate": 4.9882736864879e-06,
|
182 |
+
"logits/chosen": 0.0766684040427208,
|
183 |
+
"logits/rejected": 0.24015113711357117,
|
184 |
+
"logps/chosen": -494.6429748535156,
|
185 |
+
"logps/rejected": -534.6770629882812,
|
186 |
+
"loss": 0.2842,
|
187 |
+
"rewards/accuracies": 0.512499988079071,
|
188 |
+
"rewards/chosen": -0.07989172637462616,
|
189 |
+
"rewards/margins": 0.012701849453151226,
|
190 |
+
"rewards/rejected": -0.09259358048439026,
|
191 |
+
"step": 120
|
192 |
+
},
|
193 |
+
{
|
194 |
+
"epoch": 0.14,
|
195 |
+
"learning_rate": 4.977534912960124e-06,
|
196 |
+
"logits/chosen": 0.1341143548488617,
|
197 |
+
"logits/rejected": 0.257051020860672,
|
198 |
+
"logps/chosen": -523.8453369140625,
|
199 |
+
"logps/rejected": -525.6238403320312,
|
200 |
+
"loss": 0.297,
|
201 |
+
"rewards/accuracies": 0.5062500238418579,
|
202 |
+
"rewards/chosen": -0.11127883195877075,
|
203 |
+
"rewards/margins": 0.015445197932422161,
|
204 |
+
"rewards/rejected": -0.1267240345478058,
|
205 |
+
"step": 130
|
206 |
+
},
|
207 |
+
{
|
208 |
+
"epoch": 0.15,
|
209 |
+
"learning_rate": 4.963355698422092e-06,
|
210 |
+
"logits/chosen": 0.1251862347126007,
|
211 |
+
"logits/rejected": 0.18638013303279877,
|
212 |
+
"logps/chosen": -528.09912109375,
|
213 |
+
"logps/rejected": -526.5968627929688,
|
214 |
+
"loss": 0.2885,
|
215 |
+
"rewards/accuracies": 0.39375001192092896,
|
216 |
+
"rewards/chosen": -0.1439468413591385,
|
217 |
+
"rewards/margins": 0.01796194352209568,
|
218 |
+
"rewards/rejected": -0.16190877556800842,
|
219 |
+
"step": 140
|
220 |
+
},
|
221 |
+
{
|
222 |
+
"epoch": 0.16,
|
223 |
+
"learning_rate": 4.945755732909625e-06,
|
224 |
+
"logits/chosen": 0.004607650451362133,
|
225 |
+
"logits/rejected": 0.046932995319366455,
|
226 |
+
"logps/chosen": -519.865478515625,
|
227 |
+
"logps/rejected": -557.3140869140625,
|
228 |
+
"loss": 0.2817,
|
229 |
+
"rewards/accuracies": 0.4375,
|
230 |
+
"rewards/chosen": -0.1371057778596878,
|
231 |
+
"rewards/margins": 0.034668728709220886,
|
232 |
+
"rewards/rejected": -0.1717745065689087,
|
233 |
+
"step": 150
|
234 |
+
},
|
235 |
+
{
|
236 |
+
"epoch": 0.17,
|
237 |
+
"learning_rate": 4.924759456701167e-06,
|
238 |
+
"logits/chosen": 0.07374308258295059,
|
239 |
+
"logits/rejected": 0.1448010951280594,
|
240 |
+
"logps/chosen": -570.7451171875,
|
241 |
+
"logps/rejected": -618.2618408203125,
|
242 |
+
"loss": 0.2727,
|
243 |
+
"rewards/accuracies": 0.5249999761581421,
|
244 |
+
"rewards/chosen": -0.14755412936210632,
|
245 |
+
"rewards/margins": 0.05937732011079788,
|
246 |
+
"rewards/rejected": -0.2069314420223236,
|
247 |
+
"step": 160
|
248 |
+
},
|
249 |
+
{
|
250 |
+
"epoch": 0.18,
|
251 |
+
"learning_rate": 4.900396026378671e-06,
|
252 |
+
"logits/chosen": 0.008384035900235176,
|
253 |
+
"logits/rejected": 0.18068811297416687,
|
254 |
+
"logps/chosen": -607.10693359375,
|
255 |
+
"logps/rejected": -628.9288330078125,
|
256 |
+
"loss": 0.273,
|
257 |
+
"rewards/accuracies": 0.48750001192092896,
|
258 |
+
"rewards/chosen": -0.16640765964984894,
|
259 |
+
"rewards/margins": 0.04880762845277786,
|
260 |
+
"rewards/rejected": -0.2152152955532074,
|
261 |
+
"step": 170
|
262 |
+
},
|
263 |
+
{
|
264 |
+
"epoch": 0.19,
|
265 |
+
"learning_rate": 4.872699274339169e-06,
|
266 |
+
"logits/chosen": 0.11068884283304214,
|
267 |
+
"logits/rejected": 0.1132630854845047,
|
268 |
+
"logps/chosen": -507.98583984375,
|
269 |
+
"logps/rejected": -551.5580444335938,
|
270 |
+
"loss": 0.2848,
|
271 |
+
"rewards/accuracies": 0.48124998807907104,
|
272 |
+
"rewards/chosen": -0.12697137892246246,
|
273 |
+
"rewards/margins": 0.04405537620186806,
|
274 |
+
"rewards/rejected": -0.17102673649787903,
|
275 |
+
"step": 180
|
276 |
+
},
|
277 |
+
{
|
278 |
+
"epoch": 0.2,
|
279 |
+
"learning_rate": 4.8417076618132434e-06,
|
280 |
+
"logits/chosen": 0.033597201108932495,
|
281 |
+
"logits/rejected": 0.10967175662517548,
|
282 |
+
"logps/chosen": -624.1072998046875,
|
283 |
+
"logps/rejected": -641.2291870117188,
|
284 |
+
"loss": 0.2779,
|
285 |
+
"rewards/accuracies": 0.518750011920929,
|
286 |
+
"rewards/chosen": -0.13993802666664124,
|
287 |
+
"rewards/margins": 0.0485340878367424,
|
288 |
+
"rewards/rejected": -0.18847212195396423,
|
289 |
+
"step": 190
|
290 |
+
},
|
291 |
+
{
|
292 |
+
"epoch": 0.21,
|
293 |
+
"learning_rate": 4.807464225455655e-06,
|
294 |
+
"logits/chosen": -0.03184026479721069,
|
295 |
+
"logits/rejected": 0.09308318793773651,
|
296 |
+
"logps/chosen": -545.9130249023438,
|
297 |
+
"logps/rejected": -619.6661376953125,
|
298 |
+
"loss": 0.277,
|
299 |
+
"rewards/accuracies": 0.543749988079071,
|
300 |
+
"rewards/chosen": -0.14176079630851746,
|
301 |
+
"rewards/margins": 0.05656677484512329,
|
302 |
+
"rewards/rejected": -0.19832755625247955,
|
303 |
+
"step": 200
|
304 |
+
},
|
305 |
+
{
|
306 |
+
"epoch": 0.22,
|
307 |
+
"learning_rate": 4.770016517582283e-06,
|
308 |
+
"logits/chosen": 0.009775908663868904,
|
309 |
+
"logits/rejected": -0.003572815330699086,
|
310 |
+
"logps/chosen": -554.5719604492188,
|
311 |
+
"logps/rejected": -616.5828857421875,
|
312 |
+
"loss": 0.261,
|
313 |
+
"rewards/accuracies": 0.550000011920929,
|
314 |
+
"rewards/chosen": -0.15296652913093567,
|
315 |
+
"rewards/margins": 0.05833062529563904,
|
316 |
+
"rewards/rejected": -0.2112971544265747,
|
317 |
+
"step": 210
|
318 |
+
},
|
319 |
+
{
|
320 |
+
"epoch": 0.23,
|
321 |
+
"learning_rate": 4.7294165401363616e-06,
|
322 |
+
"logits/chosen": -0.05359144136309624,
|
323 |
+
"logits/rejected": -0.09271421283483505,
|
324 |
+
"logps/chosen": -588.1046142578125,
|
325 |
+
"logps/rejected": -641.3451538085938,
|
326 |
+
"loss": 0.2567,
|
327 |
+
"rewards/accuracies": 0.5062500238418579,
|
328 |
+
"rewards/chosen": -0.18619795143604279,
|
329 |
+
"rewards/margins": 0.058625251054763794,
|
330 |
+
"rewards/rejected": -0.24482321739196777,
|
331 |
+
"step": 220
|
332 |
+
},
|
333 |
+
{
|
334 |
+
"epoch": 0.25,
|
335 |
+
"learning_rate": 4.68572067247573e-06,
|
336 |
+
"logits/chosen": -0.09103509038686752,
|
337 |
+
"logits/rejected": -0.07337954640388489,
|
338 |
+
"logps/chosen": -569.4376220703125,
|
339 |
+
"logps/rejected": -632.4283447265625,
|
340 |
+
"loss": 0.3067,
|
341 |
+
"rewards/accuracies": 0.4749999940395355,
|
342 |
+
"rewards/chosen": -0.1694985032081604,
|
343 |
+
"rewards/margins": 0.062124162912368774,
|
344 |
+
"rewards/rejected": -0.23162266612052917,
|
345 |
+
"step": 230
|
346 |
+
},
|
347 |
+
{
|
348 |
+
"epoch": 0.26,
|
349 |
+
"learning_rate": 4.638989593081364e-06,
|
350 |
+
"logits/chosen": -0.18483319878578186,
|
351 |
+
"logits/rejected": -0.019147302955389023,
|
352 |
+
"logps/chosen": -496.79150390625,
|
353 |
+
"logps/rejected": -542.9866333007812,
|
354 |
+
"loss": 0.2758,
|
355 |
+
"rewards/accuracies": 0.4937500059604645,
|
356 |
+
"rewards/chosen": -0.1192159652709961,
|
357 |
+
"rewards/margins": 0.04292690008878708,
|
358 |
+
"rewards/rejected": -0.16214287281036377,
|
359 |
+
"step": 240
|
360 |
+
},
|
361 |
+
{
|
362 |
+
"epoch": 0.27,
|
363 |
+
"learning_rate": 4.5892881952959015e-06,
|
364 |
+
"logits/chosen": -0.11058203876018524,
|
365 |
+
"logits/rejected": -0.04781871661543846,
|
366 |
+
"logps/chosen": -514.0642700195312,
|
367 |
+
"logps/rejected": -546.6673583984375,
|
368 |
+
"loss": 0.2877,
|
369 |
+
"rewards/accuracies": 0.5,
|
370 |
+
"rewards/chosen": -0.09990507364273071,
|
371 |
+
"rewards/margins": 0.04675906524062157,
|
372 |
+
"rewards/rejected": -0.14666414260864258,
|
373 |
+
"step": 250
|
374 |
+
},
|
375 |
+
{
|
376 |
+
"epoch": 0.28,
|
377 |
+
"learning_rate": 4.536685497209182e-06,
|
378 |
+
"logits/chosen": -0.08988751471042633,
|
379 |
+
"logits/rejected": -0.06392081081867218,
|
380 |
+
"logps/chosen": -541.5545654296875,
|
381 |
+
"logps/rejected": -603.2689819335938,
|
382 |
+
"loss": 0.2804,
|
383 |
+
"rewards/accuracies": 0.4749999940395355,
|
384 |
+
"rewards/chosen": -0.11030902713537216,
|
385 |
+
"rewards/margins": 0.0489293709397316,
|
386 |
+
"rewards/rejected": -0.15923838317394257,
|
387 |
+
"step": 260
|
388 |
+
},
|
389 |
+
{
|
390 |
+
"epoch": 0.29,
|
391 |
+
"learning_rate": 4.481254545815943e-06,
|
392 |
+
"logits/chosen": -0.14757739007472992,
|
393 |
+
"logits/rejected": -0.11037661880254745,
|
394 |
+
"logps/chosen": -569.8583984375,
|
395 |
+
"logps/rejected": -653.552734375,
|
396 |
+
"loss": 0.2831,
|
397 |
+
"rewards/accuracies": 0.543749988079071,
|
398 |
+
"rewards/chosen": -0.13597574830055237,
|
399 |
+
"rewards/margins": 0.06436828523874283,
|
400 |
+
"rewards/rejected": -0.2003440111875534,
|
401 |
+
"step": 270
|
402 |
+
},
|
403 |
+
{
|
404 |
+
"epoch": 0.3,
|
405 |
+
"learning_rate": 4.42307231557875e-06,
|
406 |
+
"logits/chosen": -0.15325765311717987,
|
407 |
+
"logits/rejected": -0.08648681640625,
|
408 |
+
"logps/chosen": -532.4110107421875,
|
409 |
+
"logps/rejected": -580.8661499023438,
|
410 |
+
"loss": 0.2633,
|
411 |
+
"rewards/accuracies": 0.4937500059604645,
|
412 |
+
"rewards/chosen": -0.1242569088935852,
|
413 |
+
"rewards/margins": 0.06257221847772598,
|
414 |
+
"rewards/rejected": -0.18682913482189178,
|
415 |
+
"step": 280
|
416 |
+
},
|
417 |
+
{
|
418 |
+
"epoch": 0.31,
|
419 |
+
"learning_rate": 4.3622196015370305e-06,
|
420 |
+
"logits/chosen": -0.159153550863266,
|
421 |
+
"logits/rejected": -0.1052144318819046,
|
422 |
+
"logps/chosen": -548.8265380859375,
|
423 |
+
"logps/rejected": -647.3924560546875,
|
424 |
+
"loss": 0.2877,
|
425 |
+
"rewards/accuracies": 0.543749988079071,
|
426 |
+
"rewards/chosen": -0.14587877690792084,
|
427 |
+
"rewards/margins": 0.06854279339313507,
|
428 |
+
"rewards/rejected": -0.2144215852022171,
|
429 |
+
"step": 290
|
430 |
+
},
|
431 |
+
{
|
432 |
+
"epoch": 0.32,
|
433 |
+
"learning_rate": 4.298780907110648e-06,
|
434 |
+
"logits/chosen": -0.13682904839515686,
|
435 |
+
"logits/rejected": -0.15670131146907806,
|
436 |
+
"logps/chosen": -573.5653076171875,
|
437 |
+
"logps/rejected": -609.18896484375,
|
438 |
+
"loss": 0.2689,
|
439 |
+
"rewards/accuracies": 0.46875,
|
440 |
+
"rewards/chosen": -0.15031126141548157,
|
441 |
+
"rewards/margins": 0.07083144038915634,
|
442 |
+
"rewards/rejected": -0.2211426943540573,
|
443 |
+
"step": 300
|
444 |
+
},
|
445 |
+
{
|
446 |
+
"epoch": 0.33,
|
447 |
+
"learning_rate": 4.23284432675381e-06,
|
448 |
+
"logits/chosen": -0.20464687049388885,
|
449 |
+
"logits/rejected": -0.07677153497934341,
|
450 |
+
"logps/chosen": -494.4937438964844,
|
451 |
+
"logps/rejected": -564.7810668945312,
|
452 |
+
"loss": 0.2712,
|
453 |
+
"rewards/accuracies": 0.46875,
|
454 |
+
"rewards/chosen": -0.11489574611186981,
|
455 |
+
"rewards/margins": 0.07067451626062393,
|
456 |
+
"rewards/rejected": -0.18557025492191315,
|
457 |
+
"step": 310
|
458 |
+
},
|
459 |
+
{
|
460 |
+
"epoch": 0.34,
|
461 |
+
"learning_rate": 4.164501423622277e-06,
|
462 |
+
"logits/chosen": -0.15810438990592957,
|
463 |
+
"logits/rejected": -0.12246428430080414,
|
464 |
+
"logps/chosen": -530.6107788085938,
|
465 |
+
"logps/rejected": -583.3417358398438,
|
466 |
+
"loss": 0.2944,
|
467 |
+
"rewards/accuracies": 0.543749988079071,
|
468 |
+
"rewards/chosen": -0.14831142127513885,
|
469 |
+
"rewards/margins": 0.06477416306734085,
|
470 |
+
"rewards/rejected": -0.2130856066942215,
|
471 |
+
"step": 320
|
472 |
+
},
|
473 |
+
{
|
474 |
+
"epoch": 0.35,
|
475 |
+
"learning_rate": 4.0938471024237355e-06,
|
476 |
+
"logits/chosen": -0.13243547081947327,
|
477 |
+
"logits/rejected": -0.12164945900440216,
|
478 |
+
"logps/chosen": -611.0227661132812,
|
479 |
+
"logps/rejected": -638.0067138671875,
|
480 |
+
"loss": 0.2708,
|
481 |
+
"rewards/accuracies": 0.4749999940395355,
|
482 |
+
"rewards/chosen": -0.173904150724411,
|
483 |
+
"rewards/margins": 0.058606814593076706,
|
484 |
+
"rewards/rejected": -0.2325109988451004,
|
485 |
+
"step": 330
|
486 |
+
},
|
487 |
+
{
|
488 |
+
"epoch": 0.36,
|
489 |
+
"learning_rate": 4.020979477627907e-06,
|
490 |
+
"logits/chosen": -0.09038461744785309,
|
491 |
+
"logits/rejected": -0.08104994148015976,
|
492 |
+
"logps/chosen": -577.1747436523438,
|
493 |
+
"logps/rejected": -659.236328125,
|
494 |
+
"loss": 0.2662,
|
495 |
+
"rewards/accuracies": 0.581250011920929,
|
496 |
+
"rewards/chosen": -0.15762028098106384,
|
497 |
+
"rewards/margins": 0.07804208993911743,
|
498 |
+
"rewards/rejected": -0.23566241562366486,
|
499 |
+
"step": 340
|
500 |
+
},
|
501 |
+
{
|
502 |
+
"epoch": 0.37,
|
503 |
+
"learning_rate": 3.9459997372194105e-06,
|
504 |
+
"logits/chosen": -0.15925371646881104,
|
505 |
+
"logits/rejected": -0.02752949297428131,
|
506 |
+
"logps/chosen": -556.4603881835938,
|
507 |
+
"logps/rejected": -596.9464111328125,
|
508 |
+
"loss": 0.2753,
|
509 |
+
"rewards/accuracies": 0.45625001192092896,
|
510 |
+
"rewards/chosen": -0.11931456625461578,
|
511 |
+
"rewards/margins": 0.045837145298719406,
|
512 |
+
"rewards/rejected": -0.1651517152786255,
|
513 |
+
"step": 350
|
514 |
+
},
|
515 |
+
{
|
516 |
+
"epoch": 0.38,
|
517 |
+
"learning_rate": 3.869012002182573e-06,
|
518 |
+
"logits/chosen": -0.2364901602268219,
|
519 |
+
"logits/rejected": -0.1447964608669281,
|
520 |
+
"logps/chosen": -554.9327392578125,
|
521 |
+
"logps/rejected": -575.6502075195312,
|
522 |
+
"loss": 0.2698,
|
523 |
+
"rewards/accuracies": 0.5687500238418579,
|
524 |
+
"rewards/chosen": -0.10456685721874237,
|
525 |
+
"rewards/margins": 0.05687998980283737,
|
526 |
+
"rewards/rejected": -0.16144683957099915,
|
527 |
+
"step": 360
|
528 |
+
},
|
529 |
+
{
|
530 |
+
"epoch": 0.39,
|
531 |
+
"learning_rate": 3.7901231819133104e-06,
|
532 |
+
"logits/chosen": -0.1706862449645996,
|
533 |
+
"logits/rejected": -0.1711297631263733,
|
534 |
+
"logps/chosen": -512.93115234375,
|
535 |
+
"logps/rejected": -578.0069580078125,
|
536 |
+
"loss": 0.2742,
|
537 |
+
"rewards/accuracies": 0.4749999940395355,
|
538 |
+
"rewards/chosen": -0.11323670297861099,
|
539 |
+
"rewards/margins": 0.06127854436635971,
|
540 |
+
"rewards/rejected": -0.1745152473449707,
|
541 |
+
"step": 370
|
542 |
+
},
|
543 |
+
{
|
544 |
+
"epoch": 0.41,
|
545 |
+
"learning_rate": 3.709442825758875e-06,
|
546 |
+
"logits/chosen": -0.2998970150947571,
|
547 |
+
"logits/rejected": -0.14411643147468567,
|
548 |
+
"logps/chosen": -507.00347900390625,
|
549 |
+
"logps/rejected": -539.485595703125,
|
550 |
+
"loss": 0.2566,
|
551 |
+
"rewards/accuracies": 0.4625000059604645,
|
552 |
+
"rewards/chosen": -0.10295897722244263,
|
553 |
+
"rewards/margins": 0.05676443502306938,
|
554 |
+
"rewards/rejected": -0.1597234159708023,
|
555 |
+
"step": 380
|
556 |
+
},
|
557 |
+
{
|
558 |
+
"epoch": 0.42,
|
559 |
+
"learning_rate": 3.6270829708916113e-06,
|
560 |
+
"logits/chosen": -0.2854919731616974,
|
561 |
+
"logits/rejected": -0.21910324692726135,
|
562 |
+
"logps/chosen": -544.7410888671875,
|
563 |
+
"logps/rejected": -592.5999145507812,
|
564 |
+
"loss": 0.2625,
|
565 |
+
"rewards/accuracies": 0.45625001192092896,
|
566 |
+
"rewards/chosen": -0.12320928275585175,
|
567 |
+
"rewards/margins": 0.03932071477174759,
|
568 |
+
"rewards/rejected": -0.16252999007701874,
|
569 |
+
"step": 390
|
570 |
+
},
|
571 |
+
{
|
572 |
+
"epoch": 0.43,
|
573 |
+
"learning_rate": 3.543157986727991e-06,
|
574 |
+
"logits/chosen": -0.17541547119617462,
|
575 |
+
"logits/rejected": -0.17069751024246216,
|
576 |
+
"logps/chosen": -524.9605102539062,
|
577 |
+
"logps/rejected": -575.748046875,
|
578 |
+
"loss": 0.2735,
|
579 |
+
"rewards/accuracies": 0.518750011920929,
|
580 |
+
"rewards/chosen": -0.10180839151144028,
|
581 |
+
"rewards/margins": 0.05412058159708977,
|
582 |
+
"rewards/rejected": -0.15592898428440094,
|
583 |
+
"step": 400
|
584 |
+
},
|
585 |
+
{
|
586 |
+
"epoch": 0.44,
|
587 |
+
"learning_rate": 3.4577844161089614e-06,
|
588 |
+
"logits/chosen": -0.1704898327589035,
|
589 |
+
"logits/rejected": -0.17921459674835205,
|
590 |
+
"logps/chosen": -513.3040161132812,
|
591 |
+
"logps/rejected": -590.165771484375,
|
592 |
+
"loss": 0.2485,
|
593 |
+
"rewards/accuracies": 0.5249999761581421,
|
594 |
+
"rewards/chosen": -0.09149526059627533,
|
595 |
+
"rewards/margins": 0.05739337205886841,
|
596 |
+
"rewards/rejected": -0.14888863265514374,
|
597 |
+
"step": 410
|
598 |
+
},
|
599 |
+
{
|
600 |
+
"epoch": 0.45,
|
601 |
+
"learning_rate": 3.3710808134621577e-06,
|
602 |
+
"logits/chosen": -0.18199101090431213,
|
603 |
+
"logits/rejected": -0.1548750400543213,
|
604 |
+
"logps/chosen": -554.3550415039062,
|
605 |
+
"logps/rejected": -620.1253662109375,
|
606 |
+
"loss": 0.2814,
|
607 |
+
"rewards/accuracies": 0.512499988079071,
|
608 |
+
"rewards/chosen": -0.10769498348236084,
|
609 |
+
"rewards/margins": 0.06901998817920685,
|
610 |
+
"rewards/rejected": -0.1767149716615677,
|
611 |
+
"step": 420
|
612 |
+
},
|
613 |
+
{
|
614 |
+
"epoch": 0.46,
|
615 |
+
"learning_rate": 3.2831675801707126e-06,
|
616 |
+
"logits/chosen": -0.2242196500301361,
|
617 |
+
"logits/rejected": -0.23621216416358948,
|
618 |
+
"logps/chosen": -472.6395568847656,
|
619 |
+
"logps/rejected": -534.2669677734375,
|
620 |
+
"loss": 0.2554,
|
621 |
+
"rewards/accuracies": 0.512499988079071,
|
622 |
+
"rewards/chosen": -0.10465536266565323,
|
623 |
+
"rewards/margins": 0.06357412785291672,
|
624 |
+
"rewards/rejected": -0.16822949051856995,
|
625 |
+
"step": 430
|
626 |
+
},
|
627 |
+
{
|
628 |
+
"epoch": 0.47,
|
629 |
+
"learning_rate": 3.194166797377289e-06,
|
630 |
+
"logits/chosen": -0.24027328193187714,
|
631 |
+
"logits/rejected": -0.22386594116687775,
|
632 |
+
"logps/chosen": -576.7644653320312,
|
633 |
+
"logps/rejected": -606.2843627929688,
|
634 |
+
"loss": 0.2728,
|
635 |
+
"rewards/accuracies": 0.45625001192092896,
|
636 |
+
"rewards/chosen": -0.1252559870481491,
|
637 |
+
"rewards/margins": 0.03980395942926407,
|
638 |
+
"rewards/rejected": -0.1650599241256714,
|
639 |
+
"step": 440
|
640 |
+
},
|
641 |
+
{
|
642 |
+
"epoch": 0.48,
|
643 |
+
"learning_rate": 3.104202056455501e-06,
|
644 |
+
"logits/chosen": -0.24792134761810303,
|
645 |
+
"logits/rejected": -0.20381967723369598,
|
646 |
+
"logps/chosen": -540.4932250976562,
|
647 |
+
"logps/rejected": -590.7255249023438,
|
648 |
+
"loss": 0.2625,
|
649 |
+
"rewards/accuracies": 0.5062500238418579,
|
650 |
+
"rewards/chosen": -0.11483976989984512,
|
651 |
+
"rewards/margins": 0.06688660383224487,
|
652 |
+
"rewards/rejected": -0.1817263662815094,
|
653 |
+
"step": 450
|
654 |
+
},
|
655 |
+
{
|
656 |
+
"epoch": 0.49,
|
657 |
+
"learning_rate": 3.013398287384144e-06,
|
658 |
+
"logits/chosen": -0.22240272164344788,
|
659 |
+
"logits/rejected": -0.16780824959278107,
|
660 |
+
"logps/chosen": -575.5987548828125,
|
661 |
+
"logps/rejected": -615.41259765625,
|
662 |
+
"loss": 0.274,
|
663 |
+
"rewards/accuracies": 0.5062500238418579,
|
664 |
+
"rewards/chosen": -0.1371522694826126,
|
665 |
+
"rewards/margins": 0.05847863107919693,
|
666 |
+
"rewards/rejected": -0.19563090801239014,
|
667 |
+
"step": 460
|
668 |
+
},
|
669 |
+
{
|
670 |
+
"epoch": 0.5,
|
671 |
+
"learning_rate": 2.9218815852625717e-06,
|
672 |
+
"logits/chosen": -0.22179889678955078,
|
673 |
+
"logits/rejected": -0.2187281847000122,
|
674 |
+
"logps/chosen": -551.1289672851562,
|
675 |
+
"logps/rejected": -633.8525390625,
|
676 |
+
"loss": 0.2638,
|
677 |
+
"rewards/accuracies": 0.5687500238418579,
|
678 |
+
"rewards/chosen": -0.13685192167758942,
|
679 |
+
"rewards/margins": 0.08387337625026703,
|
680 |
+
"rewards/rejected": -0.22072529792785645,
|
681 |
+
"step": 470
|
682 |
+
},
|
683 |
+
{
|
684 |
+
"epoch": 0.51,
|
685 |
+
"learning_rate": 2.829779035208113e-06,
|
686 |
+
"logits/chosen": -0.31543129682540894,
|
687 |
+
"logits/rejected": -0.1996282935142517,
|
688 |
+
"logps/chosen": -528.4293212890625,
|
689 |
+
"logps/rejected": -616.8591918945312,
|
690 |
+
"loss": 0.2639,
|
691 |
+
"rewards/accuracies": 0.45625001192092896,
|
692 |
+
"rewards/chosen": -0.14036959409713745,
|
693 |
+
"rewards/margins": 0.08370579779148102,
|
694 |
+
"rewards/rejected": -0.22407540678977966,
|
695 |
+
"step": 480
|
696 |
+
},
|
697 |
+
{
|
698 |
+
"epoch": 0.52,
|
699 |
+
"learning_rate": 2.737218535878705e-06,
|
700 |
+
"logits/chosen": -0.1916017085313797,
|
701 |
+
"logits/rejected": -0.20610889792442322,
|
702 |
+
"logps/chosen": -513.9475708007812,
|
703 |
+
"logps/rejected": -600.5337524414062,
|
704 |
+
"loss": 0.2788,
|
705 |
+
"rewards/accuracies": 0.46875,
|
706 |
+
"rewards/chosen": -0.13624821603298187,
|
707 |
+
"rewards/margins": 0.07949782907962799,
|
708 |
+
"rewards/rejected": -0.21574604511260986,
|
709 |
+
"step": 490
|
710 |
+
},
|
711 |
+
{
|
712 |
+
"epoch": 0.53,
|
713 |
+
"learning_rate": 2.64432862186579e-06,
|
714 |
+
"logits/chosen": -0.25276193022727966,
|
715 |
+
"logits/rejected": -0.27581551671028137,
|
716 |
+
"logps/chosen": -503.4148864746094,
|
717 |
+
"logps/rejected": -561.3239135742188,
|
718 |
+
"loss": 0.2674,
|
719 |
+
"rewards/accuracies": 0.4749999940395355,
|
720 |
+
"rewards/chosen": -0.13355006277561188,
|
721 |
+
"rewards/margins": 0.05300268530845642,
|
722 |
+
"rewards/rejected": -0.1865527629852295,
|
723 |
+
"step": 500
|
724 |
+
},
|
725 |
+
{
|
726 |
+
"epoch": 0.54,
|
727 |
+
"learning_rate": 2.551238285204126e-06,
|
728 |
+
"logits/chosen": -0.24862711131572723,
|
729 |
+
"logits/rejected": -0.2061503380537033,
|
730 |
+
"logps/chosen": -596.1258544921875,
|
731 |
+
"logps/rejected": -645.7933959960938,
|
732 |
+
"loss": 0.2702,
|
733 |
+
"rewards/accuracies": 0.5,
|
734 |
+
"rewards/chosen": -0.15105004608631134,
|
735 |
+
"rewards/margins": 0.06355919688940048,
|
736 |
+
"rewards/rejected": -0.21460923552513123,
|
737 |
+
"step": 510
|
738 |
+
},
|
739 |
+
{
|
740 |
+
"epoch": 0.55,
|
741 |
+
"learning_rate": 2.4580767962463688e-06,
|
742 |
+
"logits/chosen": -0.29833105206489563,
|
743 |
+
"logits/rejected": -0.1915448009967804,
|
744 |
+
"logps/chosen": -537.9877319335938,
|
745 |
+
"logps/rejected": -557.24169921875,
|
746 |
+
"loss": 0.2662,
|
747 |
+
"rewards/accuracies": 0.44999998807907104,
|
748 |
+
"rewards/chosen": -0.12659652531147003,
|
749 |
+
"rewards/margins": 0.051892559975385666,
|
750 |
+
"rewards/rejected": -0.178489089012146,
|
751 |
+
"step": 520
|
752 |
+
},
|
753 |
+
{
|
754 |
+
"epoch": 0.57,
|
755 |
+
"learning_rate": 2.3649735241511546e-06,
|
756 |
+
"logits/chosen": -0.15996907651424408,
|
757 |
+
"logits/rejected": -0.19643227756023407,
|
758 |
+
"logps/chosen": -550.2649536132812,
|
759 |
+
"logps/rejected": -597.158203125,
|
760 |
+
"loss": 0.2658,
|
761 |
+
"rewards/accuracies": 0.5375000238418579,
|
762 |
+
"rewards/chosen": -0.13815242052078247,
|
763 |
+
"rewards/margins": 0.05568670108914375,
|
764 |
+
"rewards/rejected": -0.19383911788463593,
|
765 |
+
"step": 530
|
766 |
+
},
|
767 |
+
{
|
768 |
+
"epoch": 0.58,
|
769 |
+
"learning_rate": 2.2720577572339914e-06,
|
770 |
+
"logits/chosen": -0.2864486575126648,
|
771 |
+
"logits/rejected": -0.20076756179332733,
|
772 |
+
"logps/chosen": -528.1815185546875,
|
773 |
+
"logps/rejected": -566.8001098632812,
|
774 |
+
"loss": 0.2851,
|
775 |
+
"rewards/accuracies": 0.4437499940395355,
|
776 |
+
"rewards/chosen": -0.12378416955471039,
|
777 |
+
"rewards/margins": 0.05396197363734245,
|
778 |
+
"rewards/rejected": -0.17774614691734314,
|
779 |
+
"step": 540
|
780 |
+
},
|
781 |
+
{
|
782 |
+
"epoch": 0.59,
|
783 |
+
"learning_rate": 2.1794585234303995e-06,
|
784 |
+
"logits/chosen": -0.3104463219642639,
|
785 |
+
"logits/rejected": -0.18759891390800476,
|
786 |
+
"logps/chosen": -548.5665283203125,
|
787 |
+
"logps/rejected": -590.8157958984375,
|
788 |
+
"loss": 0.262,
|
789 |
+
"rewards/accuracies": 0.4937500059604645,
|
790 |
+
"rewards/chosen": -0.1302889883518219,
|
791 |
+
"rewards/margins": 0.04864712059497833,
|
792 |
+
"rewards/rejected": -0.17893610894680023,
|
793 |
+
"step": 550
|
794 |
+
},
|
795 |
+
{
|
796 |
+
"epoch": 0.6,
|
797 |
+
"learning_rate": 2.0873044111206407e-06,
|
798 |
+
"logits/chosen": -0.2759536802768707,
|
799 |
+
"logits/rejected": -0.2688484787940979,
|
800 |
+
"logps/chosen": -513.609375,
|
801 |
+
"logps/rejected": -596.7303466796875,
|
802 |
+
"loss": 0.2862,
|
803 |
+
"rewards/accuracies": 0.45625001192092896,
|
804 |
+
"rewards/chosen": -0.13550987839698792,
|
805 |
+
"rewards/margins": 0.053082652390003204,
|
806 |
+
"rewards/rejected": -0.18859252333641052,
|
807 |
+
"step": 560
|
808 |
+
},
|
809 |
+
{
|
810 |
+
"epoch": 0.61,
|
811 |
+
"learning_rate": 1.9957233905648293e-06,
|
812 |
+
"logits/chosen": -0.3110070824623108,
|
813 |
+
"logits/rejected": -0.294691264629364,
|
814 |
+
"logps/chosen": -499.83721923828125,
|
815 |
+
"logps/rejected": -553.9511108398438,
|
816 |
+
"loss": 0.2786,
|
817 |
+
"rewards/accuracies": 0.4937500059604645,
|
818 |
+
"rewards/chosen": -0.1329517960548401,
|
819 |
+
"rewards/margins": 0.05472971126437187,
|
820 |
+
"rewards/rejected": -0.18768151104450226,
|
821 |
+
"step": 570
|
822 |
+
},
|
823 |
+
{
|
824 |
+
"epoch": 0.62,
|
825 |
+
"learning_rate": 1.904842636196402e-06,
|
826 |
+
"logits/chosen": -0.2652963697910309,
|
827 |
+
"logits/rejected": -0.2378832995891571,
|
828 |
+
"logps/chosen": -536.1483154296875,
|
829 |
+
"logps/rejected": -593.2440795898438,
|
830 |
+
"loss": 0.269,
|
831 |
+
"rewards/accuracies": 0.518750011920929,
|
832 |
+
"rewards/chosen": -0.1256466507911682,
|
833 |
+
"rewards/margins": 0.07022566348314285,
|
834 |
+
"rewards/rejected": -0.19587230682373047,
|
835 |
+
"step": 580
|
836 |
+
},
|
837 |
+
{
|
838 |
+
"epoch": 0.63,
|
839 |
+
"learning_rate": 1.814788350020726e-06,
|
840 |
+
"logits/chosen": -0.2916659712791443,
|
841 |
+
"logits/rejected": -0.17712509632110596,
|
842 |
+
"logps/chosen": -560.8721923828125,
|
843 |
+
"logps/rejected": -634.2991943359375,
|
844 |
+
"loss": 0.2612,
|
845 |
+
"rewards/accuracies": 0.581250011920929,
|
846 |
+
"rewards/chosen": -0.1301390826702118,
|
847 |
+
"rewards/margins": 0.08458835631608963,
|
848 |
+
"rewards/rejected": -0.21472744643688202,
|
849 |
+
"step": 590
|
850 |
+
},
|
851 |
+
{
|
852 |
+
"epoch": 0.64,
|
853 |
+
"learning_rate": 1.725685586364051e-06,
|
854 |
+
"logits/chosen": -0.2919694781303406,
|
855 |
+
"logits/rejected": -0.2693978548049927,
|
856 |
+
"logps/chosen": -478.8960876464844,
|
857 |
+
"logps/rejected": -579.1296997070312,
|
858 |
+
"loss": 0.2587,
|
859 |
+
"rewards/accuracies": 0.48124998807907104,
|
860 |
+
"rewards/chosen": -0.1212470531463623,
|
861 |
+
"rewards/margins": 0.07617990672588348,
|
862 |
+
"rewards/rejected": -0.1974269449710846,
|
863 |
+
"step": 600
|
864 |
+
},
|
865 |
+
{
|
866 |
+
"epoch": 0.65,
|
867 |
+
"learning_rate": 1.6376580782162172e-06,
|
868 |
+
"logits/chosen": -0.2872418761253357,
|
869 |
+
"logits/rejected": -0.3003498911857605,
|
870 |
+
"logps/chosen": -541.9686279296875,
|
871 |
+
"logps/rejected": -597.8888549804688,
|
872 |
+
"loss": 0.2804,
|
873 |
+
"rewards/accuracies": 0.4937500059604645,
|
874 |
+
"rewards/chosen": -0.13385756313800812,
|
875 |
+
"rewards/margins": 0.07602544128894806,
|
876 |
+
"rewards/rejected": -0.20988301932811737,
|
877 |
+
"step": 610
|
878 |
+
},
|
879 |
+
{
|
880 |
+
"epoch": 0.66,
|
881 |
+
"learning_rate": 1.550828065408227e-06,
|
882 |
+
"logits/chosen": -0.19217848777770996,
|
883 |
+
"logits/rejected": -0.3079163134098053,
|
884 |
+
"logps/chosen": -527.4349365234375,
|
885 |
+
"logps/rejected": -640.5628051757812,
|
886 |
+
"loss": 0.2827,
|
887 |
+
"rewards/accuracies": 0.5562499761581421,
|
888 |
+
"rewards/chosen": -0.1333102285861969,
|
889 |
+
"rewards/margins": 0.08168620616197586,
|
890 |
+
"rewards/rejected": -0.21499641239643097,
|
891 |
+
"step": 620
|
892 |
+
},
|
893 |
+
{
|
894 |
+
"epoch": 0.67,
|
895 |
+
"learning_rate": 1.4653161248633053e-06,
|
896 |
+
"logits/chosen": -0.3484002649784088,
|
897 |
+
"logits/rejected": -0.33647680282592773,
|
898 |
+
"logps/chosen": -478.8099670410156,
|
899 |
+
"logps/rejected": -542.4144287109375,
|
900 |
+
"loss": 0.2699,
|
901 |
+
"rewards/accuracies": 0.4749999940395355,
|
902 |
+
"rewards/chosen": -0.10638388246297836,
|
903 |
+
"rewards/margins": 0.0811399295926094,
|
904 |
+
"rewards/rejected": -0.18752381205558777,
|
905 |
+
"step": 630
|
906 |
+
},
|
907 |
+
{
|
908 |
+
"epoch": 0.68,
|
909 |
+
"learning_rate": 1.381241003157162e-06,
|
910 |
+
"logits/chosen": -0.3240329623222351,
|
911 |
+
"logits/rejected": -0.28249433636665344,
|
912 |
+
"logps/chosen": -507.2118225097656,
|
913 |
+
"logps/rejected": -577.4991455078125,
|
914 |
+
"loss": 0.2603,
|
915 |
+
"rewards/accuracies": 0.45625001192092896,
|
916 |
+
"rewards/chosen": -0.11796662956476212,
|
917 |
+
"rewards/margins": 0.08652862906455994,
|
918 |
+
"rewards/rejected": -0.20449523627758026,
|
919 |
+
"step": 640
|
920 |
+
},
|
921 |
+
{
|
922 |
+
"epoch": 0.69,
|
923 |
+
"learning_rate": 1.298719451619979e-06,
|
924 |
+
"logits/chosen": -0.31510046124458313,
|
925 |
+
"logits/rejected": -0.13614080846309662,
|
926 |
+
"logps/chosen": -538.2633056640625,
|
927 |
+
"logps/rejected": -618.5576782226562,
|
928 |
+
"loss": 0.2772,
|
929 |
+
"rewards/accuracies": 0.5562499761581421,
|
930 |
+
"rewards/chosen": -0.11485198885202408,
|
931 |
+
"rewards/margins": 0.09617733955383301,
|
932 |
+
"rewards/rejected": -0.2110293209552765,
|
933 |
+
"step": 650
|
934 |
+
},
|
935 |
+
{
|
936 |
+
"epoch": 0.7,
|
937 |
+
"learning_rate": 1.2178660642091036e-06,
|
938 |
+
"logits/chosen": -0.3489423096179962,
|
939 |
+
"logits/rejected": -0.20355625450611115,
|
940 |
+
"logps/chosen": -573.0088500976562,
|
941 |
+
"logps/rejected": -612.1022338867188,
|
942 |
+
"loss": 0.2873,
|
943 |
+
"rewards/accuracies": 0.4749999940395355,
|
944 |
+
"rewards/chosen": -0.14457251131534576,
|
945 |
+
"rewards/margins": 0.057286668568849564,
|
946 |
+
"rewards/rejected": -0.20185919106006622,
|
947 |
+
"step": 660
|
948 |
+
},
|
949 |
+
{
|
950 |
+
"epoch": 0.71,
|
951 |
+
"learning_rate": 1.1387931183775821e-06,
|
952 |
+
"logits/chosen": -0.171179860830307,
|
953 |
+
"logits/rejected": -0.2416304349899292,
|
954 |
+
"logps/chosen": -521.5185546875,
|
955 |
+
"logps/rejected": -576.9376220703125,
|
956 |
+
"loss": 0.276,
|
957 |
+
"rewards/accuracies": 0.5062500238418579,
|
958 |
+
"rewards/chosen": -0.13226917386054993,
|
959 |
+
"rewards/margins": 0.05868411809206009,
|
960 |
+
"rewards/rejected": -0.19095326960086823,
|
961 |
+
"step": 670
|
962 |
+
},
|
963 |
+
{
|
964 |
+
"epoch": 0.73,
|
965 |
+
"learning_rate": 1.061610419159532e-06,
|
966 |
+
"logits/chosen": -0.22147266566753387,
|
967 |
+
"logits/rejected": -0.22640736401081085,
|
968 |
+
"logps/chosen": -479.4178161621094,
|
969 |
+
"logps/rejected": -519.9364013671875,
|
970 |
+
"loss": 0.2594,
|
971 |
+
"rewards/accuracies": 0.42500001192092896,
|
972 |
+
"rewards/chosen": -0.10929121822118759,
|
973 |
+
"rewards/margins": 0.0537792444229126,
|
974 |
+
"rewards/rejected": -0.1630704551935196,
|
975 |
+
"step": 680
|
976 |
+
},
|
977 |
+
{
|
978 |
+
"epoch": 0.74,
|
979 |
+
"learning_rate": 9.864251466888364e-07,
|
980 |
+
"logits/chosen": -0.28457021713256836,
|
981 |
+
"logits/rejected": -0.1854095160961151,
|
982 |
+
"logps/chosen": -514.6656494140625,
|
983 |
+
"logps/rejected": -569.9134521484375,
|
984 |
+
"loss": 0.2667,
|
985 |
+
"rewards/accuracies": 0.48124998807907104,
|
986 |
+
"rewards/chosen": -0.11874970048666,
|
987 |
+
"rewards/margins": 0.060789503157138824,
|
988 |
+
"rewards/rejected": -0.17953920364379883,
|
989 |
+
"step": 690
|
990 |
+
},
|
991 |
+
{
|
992 |
+
"epoch": 0.75,
|
993 |
+
"learning_rate": 9.133417073629288e-07,
|
994 |
+
"logits/chosen": -0.30244481563568115,
|
995 |
+
"logits/rejected": -0.25662168860435486,
|
996 |
+
"logps/chosen": -507.129150390625,
|
997 |
+
"logps/rejected": -575.9667358398438,
|
998 |
+
"loss": 0.2562,
|
999 |
+
"rewards/accuracies": 0.512499988079071,
|
1000 |
+
"rewards/chosen": -0.11172547191381454,
|
1001 |
+
"rewards/margins": 0.06496167182922363,
|
1002 |
+
"rewards/rejected": -0.17668713629245758,
|
1003 |
+
"step": 700
|
1004 |
+
},
|
1005 |
+
{
|
1006 |
+
"epoch": 0.76,
|
1007 |
+
"learning_rate": 8.424615888583332e-07,
|
1008 |
+
"logits/chosen": -0.280193954706192,
|
1009 |
+
"logits/rejected": -0.17224326729774475,
|
1010 |
+
"logps/chosen": -535.2366943359375,
|
1011 |
+
"logps/rejected": -601.5757446289062,
|
1012 |
+
"loss": 0.2549,
|
1013 |
+
"rewards/accuracies": 0.512499988079071,
|
1014 |
+
"rewards/chosen": -0.12534330785274506,
|
1015 |
+
"rewards/margins": 0.06499765813350677,
|
1016 |
+
"rewards/rejected": -0.19034096598625183,
|
1017 |
+
"step": 710
|
1018 |
+
},
|
1019 |
+
{
|
1020 |
+
"epoch": 0.77,
|
1021 |
+
"learning_rate": 7.738832191993092e-07,
|
1022 |
+
"logits/chosen": -0.228287011384964,
|
1023 |
+
"logits/rejected": -0.22289884090423584,
|
1024 |
+
"logps/chosen": -550.7459106445312,
|
1025 |
+
"logps/rejected": -601.8172607421875,
|
1026 |
+
"loss": 0.2557,
|
1027 |
+
"rewards/accuracies": 0.53125,
|
1028 |
+
"rewards/chosen": -0.1304478794336319,
|
1029 |
+
"rewards/margins": 0.06193193793296814,
|
1030 |
+
"rewards/rejected": -0.19237980246543884,
|
1031 |
+
"step": 720
|
1032 |
+
},
|
1033 |
+
{
|
1034 |
+
"epoch": 0.78,
|
1035 |
+
"learning_rate": 7.077018300752917e-07,
|
1036 |
+
"logits/chosen": -0.23304042220115662,
|
1037 |
+
"logits/rejected": -0.2573367953300476,
|
1038 |
+
"logps/chosen": -546.1334838867188,
|
1039 |
+
"logps/rejected": -603.4042358398438,
|
1040 |
+
"loss": 0.253,
|
1041 |
+
"rewards/accuracies": 0.543749988079071,
|
1042 |
+
"rewards/chosen": -0.12139827013015747,
|
1043 |
+
"rewards/margins": 0.07925884425640106,
|
1044 |
+
"rewards/rejected": -0.20065709948539734,
|
1045 |
+
"step": 730
|
1046 |
+
},
|
1047 |
+
{
|
1048 |
+
"epoch": 0.79,
|
1049 |
+
"learning_rate": 6.440093245969342e-07,
|
1050 |
+
"logits/chosen": -0.28443416953086853,
|
1051 |
+
"logits/rejected": -0.18750058114528656,
|
1052 |
+
"logps/chosen": -487.604248046875,
|
1053 |
+
"logps/rejected": -528.2503051757812,
|
1054 |
+
"loss": 0.2834,
|
1055 |
+
"rewards/accuracies": 0.4749999940395355,
|
1056 |
+
"rewards/chosen": -0.10632891952991486,
|
1057 |
+
"rewards/margins": 0.07151490449905396,
|
1058 |
+
"rewards/rejected": -0.17784383893013,
|
1059 |
+
"step": 740
|
1060 |
+
},
|
1061 |
+
{
|
1062 |
+
"epoch": 0.8,
|
1063 |
+
"learning_rate": 5.828941496744075e-07,
|
1064 |
+
"logits/chosen": -0.26037511229515076,
|
1065 |
+
"logits/rejected": -0.2036275863647461,
|
1066 |
+
"logps/chosen": -544.8526611328125,
|
1067 |
+
"logps/rejected": -635.510009765625,
|
1068 |
+
"loss": 0.2576,
|
1069 |
+
"rewards/accuracies": 0.543749988079071,
|
1070 |
+
"rewards/chosen": -0.12171368300914764,
|
1071 |
+
"rewards/margins": 0.08451158553361893,
|
1072 |
+
"rewards/rejected": -0.20622527599334717,
|
1073 |
+
"step": 750
|
1074 |
+
},
|
1075 |
+
{
|
1076 |
+
"epoch": 0.81,
|
1077 |
+
"learning_rate": 5.244411731951671e-07,
|
1078 |
+
"logits/chosen": -0.3115929961204529,
|
1079 |
+
"logits/rejected": -0.104661725461483,
|
1080 |
+
"logps/chosen": -510.882080078125,
|
1081 |
+
"logps/rejected": -545.64501953125,
|
1082 |
+
"loss": 0.2533,
|
1083 |
+
"rewards/accuracies": 0.46875,
|
1084 |
+
"rewards/chosen": -0.12393975257873535,
|
1085 |
+
"rewards/margins": 0.060543425381183624,
|
1086 |
+
"rewards/rejected": -0.18448317050933838,
|
1087 |
+
"step": 760
|
1088 |
+
},
|
1089 |
+
{
|
1090 |
+
"epoch": 0.82,
|
1091 |
+
"learning_rate": 4.6873156617173594e-07,
|
1092 |
+
"logits/chosen": -0.354599267244339,
|
1093 |
+
"logits/rejected": -0.2644585967063904,
|
1094 |
+
"logps/chosen": -556.0206909179688,
|
1095 |
+
"logps/rejected": -593.3365478515625,
|
1096 |
+
"loss": 0.2578,
|
1097 |
+
"rewards/accuracies": 0.5062500238418579,
|
1098 |
+
"rewards/chosen": -0.12226442992687225,
|
1099 |
+
"rewards/margins": 0.06983228772878647,
|
1100 |
+
"rewards/rejected": -0.1920967400074005,
|
1101 |
+
"step": 770
|
1102 |
+
},
|
1103 |
+
{
|
1104 |
+
"epoch": 0.83,
|
1105 |
+
"learning_rate": 4.1584269002318653e-07,
|
1106 |
+
"logits/chosen": -0.29683440923690796,
|
1107 |
+
"logits/rejected": -0.1917039006948471,
|
1108 |
+
"logps/chosen": -565.5325927734375,
|
1109 |
+
"logps/rejected": -639.9913330078125,
|
1110 |
+
"loss": 0.2528,
|
1111 |
+
"rewards/accuracies": 0.543749988079071,
|
1112 |
+
"rewards/chosen": -0.128949835896492,
|
1113 |
+
"rewards/margins": 0.07002034783363342,
|
1114 |
+
"rewards/rejected": -0.19897018373012543,
|
1115 |
+
"step": 780
|
1116 |
+
},
|
1117 |
+
{
|
1118 |
+
"epoch": 0.84,
|
1119 |
+
"learning_rate": 3.658479891468258e-07,
|
1120 |
+
"logits/chosen": -0.20871396362781525,
|
1121 |
+
"logits/rejected": -0.18840913474559784,
|
1122 |
+
"logps/chosen": -515.9395751953125,
|
1123 |
+
"logps/rejected": -580.9120483398438,
|
1124 |
+
"loss": 0.2695,
|
1125 |
+
"rewards/accuracies": 0.5062500238418579,
|
1126 |
+
"rewards/chosen": -0.1138933077454567,
|
1127 |
+
"rewards/margins": 0.06584799289703369,
|
1128 |
+
"rewards/rejected": -0.1797412931919098,
|
1129 |
+
"step": 790
|
1130 |
+
},
|
1131 |
+
{
|
1132 |
+
"epoch": 0.85,
|
1133 |
+
"learning_rate": 3.18816888929272e-07,
|
1134 |
+
"logits/chosen": -0.31950509548187256,
|
1135 |
+
"logits/rejected": -0.20166322588920593,
|
1136 |
+
"logps/chosen": -542.8575439453125,
|
1137 |
+
"logps/rejected": -579.9568481445312,
|
1138 |
+
"loss": 0.2798,
|
1139 |
+
"rewards/accuracies": 0.518750011920929,
|
1140 |
+
"rewards/chosen": -0.12461917102336884,
|
1141 |
+
"rewards/margins": 0.0687582939863205,
|
1142 |
+
"rewards/rejected": -0.19337746500968933,
|
1143 |
+
"step": 800
|
1144 |
+
},
|
1145 |
+
{
|
1146 |
+
"epoch": 0.86,
|
1147 |
+
"learning_rate": 2.748146993385484e-07,
|
1148 |
+
"logits/chosen": -0.24710452556610107,
|
1149 |
+
"logits/rejected": -0.285795658826828,
|
1150 |
+
"logps/chosen": -584.72412109375,
|
1151 |
+
"logps/rejected": -667.7575073242188,
|
1152 |
+
"loss": 0.2646,
|
1153 |
+
"rewards/accuracies": 0.5625,
|
1154 |
+
"rewards/chosen": -0.13876564800739288,
|
1155 |
+
"rewards/margins": 0.08441019803285599,
|
1156 |
+
"rewards/rejected": -0.22317583858966827,
|
1157 |
+
"step": 810
|
1158 |
+
},
|
1159 |
+
{
|
1160 |
+
"epoch": 0.87,
|
1161 |
+
"learning_rate": 2.3390252423108077e-07,
|
1162 |
+
"logits/chosen": -0.34350934624671936,
|
1163 |
+
"logits/rejected": -0.1507629156112671,
|
1164 |
+
"logps/chosen": -555.4998779296875,
|
1165 |
+
"logps/rejected": -603.321533203125,
|
1166 |
+
"loss": 0.2657,
|
1167 |
+
"rewards/accuracies": 0.574999988079071,
|
1168 |
+
"rewards/chosen": -0.11694788932800293,
|
1169 |
+
"rewards/margins": 0.09008406102657318,
|
1170 |
+
"rewards/rejected": -0.20703193545341492,
|
1171 |
+
"step": 820
|
1172 |
+
},
|
1173 |
+
{
|
1174 |
+
"epoch": 0.89,
|
1175 |
+
"learning_rate": 1.961371764995243e-07,
|
1176 |
+
"logits/chosen": -0.2246185541152954,
|
1177 |
+
"logits/rejected": -0.23372571170330048,
|
1178 |
+
"logps/chosen": -545.0233154296875,
|
1179 |
+
"logps/rejected": -591.8538208007812,
|
1180 |
+
"loss": 0.2813,
|
1181 |
+
"rewards/accuracies": 0.46875,
|
1182 |
+
"rewards/chosen": -0.12621915340423584,
|
1183 |
+
"rewards/margins": 0.06708641350269318,
|
1184 |
+
"rewards/rejected": -0.19330556690692902,
|
1185 |
+
"step": 830
|
1186 |
+
},
|
1187 |
+
{
|
1188 |
+
"epoch": 0.9,
|
1189 |
+
"learning_rate": 1.61571099179261e-07,
|
1190 |
+
"logits/chosen": -0.364227294921875,
|
1191 |
+
"logits/rejected": -0.26786571741104126,
|
1192 |
+
"logps/chosen": -481.79718017578125,
|
1193 |
+
"logps/rejected": -542.220947265625,
|
1194 |
+
"loss": 0.2508,
|
1195 |
+
"rewards/accuracies": 0.5,
|
1196 |
+
"rewards/chosen": -0.11298207938671112,
|
1197 |
+
"rewards/margins": 0.07751892507076263,
|
1198 |
+
"rewards/rejected": -0.19050098955631256,
|
1199 |
+
"step": 840
|
1200 |
+
},
|
1201 |
+
{
|
1202 |
+
"epoch": 0.91,
|
1203 |
+
"learning_rate": 1.3025229262312367e-07,
|
1204 |
+
"logits/chosen": -0.33576229214668274,
|
1205 |
+
"logits/rejected": -0.281321257352829,
|
1206 |
+
"logps/chosen": -549.2854614257812,
|
1207 |
+
"logps/rejected": -602.2276611328125,
|
1208 |
+
"loss": 0.2735,
|
1209 |
+
"rewards/accuracies": 0.5062500238418579,
|
1210 |
+
"rewards/chosen": -0.1300746351480484,
|
1211 |
+
"rewards/margins": 0.06319116055965424,
|
1212 |
+
"rewards/rejected": -0.19326578080654144,
|
1213 |
+
"step": 850
|
1214 |
+
},
|
1215 |
+
{
|
1216 |
+
"epoch": 0.92,
|
1217 |
+
"learning_rate": 1.0222424784546853e-07,
|
1218 |
+
"logits/chosen": -0.1330328732728958,
|
1219 |
+
"logits/rejected": -0.2882300913333893,
|
1220 |
+
"logps/chosen": -529.1005859375,
|
1221 |
+
"logps/rejected": -605.9659423828125,
|
1222 |
+
"loss": 0.2781,
|
1223 |
+
"rewards/accuracies": 0.4937500059604645,
|
1224 |
+
"rewards/chosen": -0.12892162799835205,
|
1225 |
+
"rewards/margins": 0.0663539245724678,
|
1226 |
+
"rewards/rejected": -0.19527556002140045,
|
1227 |
+
"step": 860
|
1228 |
+
},
|
1229 |
+
{
|
1230 |
+
"epoch": 0.93,
|
1231 |
+
"learning_rate": 7.752588612816553e-08,
|
1232 |
+
"logits/chosen": -0.24367213249206543,
|
1233 |
+
"logits/rejected": -0.2970259189605713,
|
1234 |
+
"logps/chosen": -579.6039428710938,
|
1235 |
+
"logps/rejected": -616.4037475585938,
|
1236 |
+
"loss": 0.2589,
|
1237 |
+
"rewards/accuracies": 0.5062500238418579,
|
1238 |
+
"rewards/chosen": -0.13028362393379211,
|
1239 |
+
"rewards/margins": 0.061786215752363205,
|
1240 |
+
"rewards/rejected": -0.19206984341144562,
|
1241 |
+
"step": 870
|
1242 |
+
},
|
1243 |
+
{
|
1244 |
+
"epoch": 0.94,
|
1245 |
+
"learning_rate": 5.619150497236991e-08,
|
1246 |
+
"logits/chosen": -0.3084169030189514,
|
1247 |
+
"logits/rejected": -0.2781293988227844,
|
1248 |
+
"logps/chosen": -531.304931640625,
|
1249 |
+
"logps/rejected": -618.241943359375,
|
1250 |
+
"loss": 0.2481,
|
1251 |
+
"rewards/accuracies": 0.5562499761581421,
|
1252 |
+
"rewards/chosen": -0.12809237837791443,
|
1253 |
+
"rewards/margins": 0.09018449485301971,
|
1254 |
+
"rewards/rejected": -0.21827688813209534,
|
1255 |
+
"step": 880
|
1256 |
+
},
|
1257 |
+
{
|
1258 |
+
"epoch": 0.95,
|
1259 |
+
"learning_rate": 3.825073047112743e-08,
|
1260 |
+
"logits/chosen": -0.3312014043331146,
|
1261 |
+
"logits/rejected": -0.25709718465805054,
|
1262 |
+
"logps/chosen": -530.1060791015625,
|
1263 |
+
"logps/rejected": -596.2859497070312,
|
1264 |
+
"loss": 0.2425,
|
1265 |
+
"rewards/accuracies": 0.5,
|
1266 |
+
"rewards/chosen": -0.11436812579631805,
|
1267 |
+
"rewards/margins": 0.08216674625873566,
|
1268 |
+
"rewards/rejected": -0.1965348720550537,
|
1269 |
+
"step": 890
|
1270 |
+
},
|
1271 |
+
{
|
1272 |
+
"epoch": 0.96,
|
1273 |
+
"learning_rate": 2.372847616895685e-08,
|
1274 |
+
"logits/chosen": -0.24755005538463593,
|
1275 |
+
"logits/rejected": -0.3069338798522949,
|
1276 |
+
"logps/chosen": -514.0535888671875,
|
1277 |
+
"logps/rejected": -623.8255615234375,
|
1278 |
+
"loss": 0.2624,
|
1279 |
+
"rewards/accuracies": 0.48750001192092896,
|
1280 |
+
"rewards/chosen": -0.13177482783794403,
|
1281 |
+
"rewards/margins": 0.08524416387081146,
|
1282 |
+
"rewards/rejected": -0.2170190066099167,
|
1283 |
+
"step": 900
|
1284 |
+
},
|
1285 |
+
{
|
1286 |
+
"epoch": 0.97,
|
1287 |
+
"learning_rate": 1.264490846553279e-08,
|
1288 |
+
"logits/chosen": -0.24237994849681854,
|
1289 |
+
"logits/rejected": -0.26542186737060547,
|
1290 |
+
"logps/chosen": -535.3384399414062,
|
1291 |
+
"logps/rejected": -621.1607055664062,
|
1292 |
+
"loss": 0.2496,
|
1293 |
+
"rewards/accuracies": 0.53125,
|
1294 |
+
"rewards/chosen": -0.10499405860900879,
|
1295 |
+
"rewards/margins": 0.08728118240833282,
|
1296 |
+
"rewards/rejected": -0.1922752410173416,
|
1297 |
+
"step": 910
|
1298 |
+
},
|
1299 |
+
{
|
1300 |
+
"epoch": 0.98,
|
1301 |
+
"learning_rate": 5.015418611516165e-09,
|
1302 |
+
"logits/chosen": -0.3247453570365906,
|
1303 |
+
"logits/rejected": -0.34661737084388733,
|
1304 |
+
"logps/chosen": -501.97705078125,
|
1305 |
+
"logps/rejected": -568.5880737304688,
|
1306 |
+
"loss": 0.2714,
|
1307 |
+
"rewards/accuracies": 0.518750011920929,
|
1308 |
+
"rewards/chosen": -0.11833710968494415,
|
1309 |
+
"rewards/margins": 0.08402119576931,
|
1310 |
+
"rewards/rejected": -0.20235827565193176,
|
1311 |
+
"step": 920
|
1312 |
+
},
|
1313 |
+
{
|
1314 |
+
"epoch": 0.99,
|
1315 |
+
"learning_rate": 8.506013354186993e-10,
|
1316 |
+
"logits/chosen": -0.27017658948898315,
|
1317 |
+
"logits/rejected": -0.2859312891960144,
|
1318 |
+
"logps/chosen": -512.9287109375,
|
1319 |
+
"logps/rejected": -591.2950439453125,
|
1320 |
+
"loss": 0.2695,
|
1321 |
+
"rewards/accuracies": 0.48750001192092896,
|
1322 |
+
"rewards/chosen": -0.12850606441497803,
|
1323 |
+
"rewards/margins": 0.07382752746343613,
|
1324 |
+
"rewards/rejected": -0.20233359932899475,
|
1325 |
+
"step": 930
|
1326 |
+
},
|
1327 |
+
{
|
1328 |
+
"epoch": 1.0,
|
1329 |
+
"step": 937,
|
1330 |
+
"total_flos": 0.0,
|
1331 |
+
"train_loss": 0.274457171320788,
|
1332 |
+
"train_runtime": 7836.9626,
|
1333 |
+
"train_samples_per_second": 3.828,
|
1334 |
+
"train_steps_per_second": 0.12
|
1335 |
+
}
|
1336 |
+
],
|
1337 |
+
"logging_steps": 10,
|
1338 |
+
"max_steps": 937,
|
1339 |
+
"num_input_tokens_seen": 0,
|
1340 |
+
"num_train_epochs": 1,
|
1341 |
+
"save_steps": 100,
|
1342 |
+
"total_flos": 0.0,
|
1343 |
+
"train_batch_size": 4,
|
1344 |
+
"trial_name": null,
|
1345 |
+
"trial_params": null
|
1346 |
+
}
|