Model save
Browse files- README.md +61 -0
- adapter_model.safetensors +1 -1
- all_results.json +8 -0
- runs/May20_23-59-22_gpu4-119-5/events.out.tfevents.1716213724.gpu4-119-5.1093839.0 +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-v10-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-v10-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 sha256:013e246e692f3708d5627e2aa217f46b35764ec9628f3e6b416fee5b65022826
|
3 |
size 335579632
|
all_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 1.0,
|
3 |
+
"train_loss": 0.1881250925163322,
|
4 |
+
"train_runtime": 7837.4153,
|
5 |
+
"train_samples": 30000,
|
6 |
+
"train_samples_per_second": 3.828,
|
7 |
+
"train_steps_per_second": 0.12
|
8 |
+
}
|
runs/May20_23-59-22_gpu4-119-5/events.out.tfevents.1716213724.gpu4-119-5.1093839.0
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 sha256:51b9861bafae2f8aaca8edc38e12dbb3c948afc76e524b16325a1d4d387f229b
|
3 |
+
size 64720
|
train_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 1.0,
|
3 |
+
"train_loss": 0.1881250925163322,
|
4 |
+
"train_runtime": 7837.4153,
|
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": 0.1853,
|
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": 0.15297521650791168,
|
29 |
+
"logits/rejected": 0.29175662994384766,
|
30 |
+
"logps/chosen": -365.80181884765625,
|
31 |
+
"logps/rejected": -353.0853271484375,
|
32 |
+
"loss": 0.2099,
|
33 |
+
"rewards/accuracies": 0.25,
|
34 |
+
"rewards/chosen": -0.0007080123177729547,
|
35 |
+
"rewards/margins": -5.8396861277287826e-05,
|
36 |
+
"rewards/rejected": -0.0006496154237538576,
|
37 |
+
"step": 10
|
38 |
+
},
|
39 |
+
{
|
40 |
+
"epoch": 0.02,
|
41 |
+
"learning_rate": 1.0638297872340427e-06,
|
42 |
+
"logits/chosen": 0.11968117952346802,
|
43 |
+
"logits/rejected": 0.2041483372449875,
|
44 |
+
"logps/chosen": -340.0993347167969,
|
45 |
+
"logps/rejected": -348.33087158203125,
|
46 |
+
"loss": 0.2094,
|
47 |
+
"rewards/accuracies": 0.26249998807907104,
|
48 |
+
"rewards/chosen": -0.000655159296002239,
|
49 |
+
"rewards/margins": -8.313418220495805e-05,
|
50 |
+
"rewards/rejected": -0.0005720251356251538,
|
51 |
+
"step": 20
|
52 |
+
},
|
53 |
+
{
|
54 |
+
"epoch": 0.03,
|
55 |
+
"learning_rate": 1.595744680851064e-06,
|
56 |
+
"logits/chosen": 0.2551038861274719,
|
57 |
+
"logits/rejected": 0.25183868408203125,
|
58 |
+
"logps/chosen": -383.1521301269531,
|
59 |
+
"logps/rejected": -364.0672302246094,
|
60 |
+
"loss": 0.2012,
|
61 |
+
"rewards/accuracies": 0.38749998807907104,
|
62 |
+
"rewards/chosen": -0.0005650260718539357,
|
63 |
+
"rewards/margins": 0.00010353984544053674,
|
64 |
+
"rewards/rejected": -0.0006685658590868115,
|
65 |
+
"step": 30
|
66 |
+
},
|
67 |
+
{
|
68 |
+
"epoch": 0.04,
|
69 |
+
"learning_rate": 2.1276595744680853e-06,
|
70 |
+
"logits/chosen": 0.1547292321920395,
|
71 |
+
"logits/rejected": 0.27106207609176636,
|
72 |
+
"logps/chosen": -401.61614990234375,
|
73 |
+
"logps/rejected": -385.8863220214844,
|
74 |
+
"loss": 0.2099,
|
75 |
+
"rewards/accuracies": 0.3375000059604645,
|
76 |
+
"rewards/chosen": -0.0006045111804269254,
|
77 |
+
"rewards/margins": -5.9384223277447745e-05,
|
78 |
+
"rewards/rejected": -0.0005451269680634141,
|
79 |
+
"step": 40
|
80 |
+
},
|
81 |
+
{
|
82 |
+
"epoch": 0.05,
|
83 |
+
"learning_rate": 2.6595744680851065e-06,
|
84 |
+
"logits/chosen": 0.23326897621154785,
|
85 |
+
"logits/rejected": 0.27433687448501587,
|
86 |
+
"logps/chosen": -441.8401794433594,
|
87 |
+
"logps/rejected": -432.41485595703125,
|
88 |
+
"loss": 0.2047,
|
89 |
+
"rewards/accuracies": 0.48750001192092896,
|
90 |
+
"rewards/chosen": -0.000930719543248415,
|
91 |
+
"rewards/margins": 0.000368706532754004,
|
92 |
+
"rewards/rejected": -0.0012994259595870972,
|
93 |
+
"step": 50
|
94 |
+
},
|
95 |
+
{
|
96 |
+
"epoch": 0.06,
|
97 |
+
"learning_rate": 3.191489361702128e-06,
|
98 |
+
"logits/chosen": 0.17064206302165985,
|
99 |
+
"logits/rejected": 0.3185887336730957,
|
100 |
+
"logps/chosen": -410.41473388671875,
|
101 |
+
"logps/rejected": -414.3666076660156,
|
102 |
+
"loss": 0.2182,
|
103 |
+
"rewards/accuracies": 0.41874998807907104,
|
104 |
+
"rewards/chosen": -0.001497046323493123,
|
105 |
+
"rewards/margins": 0.00016530933498870581,
|
106 |
+
"rewards/rejected": -0.0016623556148260832,
|
107 |
+
"step": 60
|
108 |
+
},
|
109 |
+
{
|
110 |
+
"epoch": 0.07,
|
111 |
+
"learning_rate": 3.723404255319149e-06,
|
112 |
+
"logits/chosen": 0.12393184751272202,
|
113 |
+
"logits/rejected": 0.2235107123851776,
|
114 |
+
"logps/chosen": -354.70562744140625,
|
115 |
+
"logps/rejected": -356.94586181640625,
|
116 |
+
"loss": 0.2086,
|
117 |
+
"rewards/accuracies": 0.5,
|
118 |
+
"rewards/chosen": -5.924403740209527e-05,
|
119 |
+
"rewards/margins": 0.0009310436435043812,
|
120 |
+
"rewards/rejected": -0.000990287633612752,
|
121 |
+
"step": 70
|
122 |
+
},
|
123 |
+
{
|
124 |
+
"epoch": 0.09,
|
125 |
+
"learning_rate": 4.255319148936171e-06,
|
126 |
+
"logits/chosen": 0.14353762567043304,
|
127 |
+
"logits/rejected": 0.2516772449016571,
|
128 |
+
"logps/chosen": -392.6264343261719,
|
129 |
+
"logps/rejected": -380.66351318359375,
|
130 |
+
"loss": 0.208,
|
131 |
+
"rewards/accuracies": 0.40625,
|
132 |
+
"rewards/chosen": 0.0002285484952153638,
|
133 |
+
"rewards/margins": 0.001034508110024035,
|
134 |
+
"rewards/rejected": -0.0008059596875682473,
|
135 |
+
"step": 80
|
136 |
+
},
|
137 |
+
{
|
138 |
+
"epoch": 0.1,
|
139 |
+
"learning_rate": 4.787234042553192e-06,
|
140 |
+
"logits/chosen": 0.24103212356567383,
|
141 |
+
"logits/rejected": 0.1776101142168045,
|
142 |
+
"logps/chosen": -393.3184509277344,
|
143 |
+
"logps/rejected": -416.2762145996094,
|
144 |
+
"loss": 0.1992,
|
145 |
+
"rewards/accuracies": 0.48124998807907104,
|
146 |
+
"rewards/chosen": -0.0003849788336083293,
|
147 |
+
"rewards/margins": 0.0017982361605390906,
|
148 |
+
"rewards/rejected": -0.0021832147613167763,
|
149 |
+
"step": 90
|
150 |
+
},
|
151 |
+
{
|
152 |
+
"epoch": 0.11,
|
153 |
+
"learning_rate": 4.999375059004058e-06,
|
154 |
+
"logits/chosen": 0.16443544626235962,
|
155 |
+
"logits/rejected": 0.17112873494625092,
|
156 |
+
"logps/chosen": -416.6537170410156,
|
157 |
+
"logps/rejected": -411.6963806152344,
|
158 |
+
"loss": 0.2064,
|
159 |
+
"rewards/accuracies": 0.45625001192092896,
|
160 |
+
"rewards/chosen": 0.0005408526631072164,
|
161 |
+
"rewards/margins": 0.0026028361171483994,
|
162 |
+
"rewards/rejected": -0.002061983570456505,
|
163 |
+
"step": 100
|
164 |
+
},
|
165 |
+
{
|
166 |
+
"epoch": 0.12,
|
167 |
+
"learning_rate": 4.9955571065548795e-06,
|
168 |
+
"logits/chosen": 0.2384149730205536,
|
169 |
+
"logits/rejected": 0.1614537537097931,
|
170 |
+
"logps/chosen": -406.7789306640625,
|
171 |
+
"logps/rejected": -391.0703430175781,
|
172 |
+
"loss": 0.2008,
|
173 |
+
"rewards/accuracies": 0.5,
|
174 |
+
"rewards/chosen": 0.012324010953307152,
|
175 |
+
"rewards/margins": 0.0033186424989253283,
|
176 |
+
"rewards/rejected": 0.009005369618535042,
|
177 |
+
"step": 110
|
178 |
+
},
|
179 |
+
{
|
180 |
+
"epoch": 0.13,
|
181 |
+
"learning_rate": 4.9882736864879e-06,
|
182 |
+
"logits/chosen": 0.08936997503042221,
|
183 |
+
"logits/rejected": 0.25732293725013733,
|
184 |
+
"logps/chosen": -397.0160827636719,
|
185 |
+
"logps/rejected": -431.9867248535156,
|
186 |
+
"loss": 0.2064,
|
187 |
+
"rewards/accuracies": 0.53125,
|
188 |
+
"rewards/chosen": 0.01773521490395069,
|
189 |
+
"rewards/margins": 0.007638473063707352,
|
190 |
+
"rewards/rejected": 0.01009674184024334,
|
191 |
+
"step": 120
|
192 |
+
},
|
193 |
+
{
|
194 |
+
"epoch": 0.14,
|
195 |
+
"learning_rate": 4.977534912960124e-06,
|
196 |
+
"logits/chosen": 0.14923642575740814,
|
197 |
+
"logits/rejected": 0.27579236030578613,
|
198 |
+
"logps/chosen": -407.21258544921875,
|
199 |
+
"logps/rejected": -401.8697204589844,
|
200 |
+
"loss": 0.2048,
|
201 |
+
"rewards/accuracies": 0.512499988079071,
|
202 |
+
"rewards/chosen": 0.005354008637368679,
|
203 |
+
"rewards/margins": 0.00832393579185009,
|
204 |
+
"rewards/rejected": -0.0029699269216507673,
|
205 |
+
"step": 130
|
206 |
+
},
|
207 |
+
{
|
208 |
+
"epoch": 0.15,
|
209 |
+
"learning_rate": 4.963355698422092e-06,
|
210 |
+
"logits/chosen": 0.13965365290641785,
|
211 |
+
"logits/rejected": 0.20428553223609924,
|
212 |
+
"logps/chosen": -396.0818786621094,
|
213 |
+
"logps/rejected": -384.4440612792969,
|
214 |
+
"loss": 0.2016,
|
215 |
+
"rewards/accuracies": 0.41874998807907104,
|
216 |
+
"rewards/chosen": -0.011929613538086414,
|
217 |
+
"rewards/margins": 0.00782632827758789,
|
218 |
+
"rewards/rejected": -0.01975594088435173,
|
219 |
+
"step": 140
|
220 |
+
},
|
221 |
+
{
|
222 |
+
"epoch": 0.16,
|
223 |
+
"learning_rate": 4.945755732909625e-06,
|
224 |
+
"logits/chosen": 0.0017524458235129714,
|
225 |
+
"logits/rejected": 0.048104483634233475,
|
226 |
+
"logps/chosen": -403.7103576660156,
|
227 |
+
"logps/rejected": -421.3060607910156,
|
228 |
+
"loss": 0.1918,
|
229 |
+
"rewards/accuracies": 0.44999998807907104,
|
230 |
+
"rewards/chosen": -0.020950669422745705,
|
231 |
+
"rewards/margins": 0.01481578964740038,
|
232 |
+
"rewards/rejected": -0.03576646000146866,
|
233 |
+
"step": 150
|
234 |
+
},
|
235 |
+
{
|
236 |
+
"epoch": 0.17,
|
237 |
+
"learning_rate": 4.924759456701167e-06,
|
238 |
+
"logits/chosen": 0.050279758870601654,
|
239 |
+
"logits/rejected": 0.12556883692741394,
|
240 |
+
"logps/chosen": -467.9580993652344,
|
241 |
+
"logps/rejected": -487.71844482421875,
|
242 |
+
"loss": 0.1868,
|
243 |
+
"rewards/accuracies": 0.550000011920929,
|
244 |
+
"rewards/chosen": -0.044767118990421295,
|
245 |
+
"rewards/margins": 0.03162100166082382,
|
246 |
+
"rewards/rejected": -0.07638812065124512,
|
247 |
+
"step": 160
|
248 |
+
},
|
249 |
+
{
|
250 |
+
"epoch": 0.18,
|
251 |
+
"learning_rate": 4.900396026378671e-06,
|
252 |
+
"logits/chosen": -0.020991306751966476,
|
253 |
+
"logits/rejected": 0.15817420184612274,
|
254 |
+
"logps/chosen": -522.8843383789062,
|
255 |
+
"logps/rejected": -518.1360473632812,
|
256 |
+
"loss": 0.2105,
|
257 |
+
"rewards/accuracies": 0.4749999940395355,
|
258 |
+
"rewards/chosen": -0.08218502998352051,
|
259 |
+
"rewards/margins": 0.02223752811551094,
|
260 |
+
"rewards/rejected": -0.10442256927490234,
|
261 |
+
"step": 170
|
262 |
+
},
|
263 |
+
{
|
264 |
+
"epoch": 0.19,
|
265 |
+
"learning_rate": 4.872699274339169e-06,
|
266 |
+
"logits/chosen": 0.08929436653852463,
|
267 |
+
"logits/rejected": 0.09290768206119537,
|
268 |
+
"logps/chosen": -470.04296875,
|
269 |
+
"logps/rejected": -501.46661376953125,
|
270 |
+
"loss": 0.1886,
|
271 |
+
"rewards/accuracies": 0.48750001192092896,
|
272 |
+
"rewards/chosen": -0.08902844041585922,
|
273 |
+
"rewards/margins": 0.03190689533948898,
|
274 |
+
"rewards/rejected": -0.12093535810709,
|
275 |
+
"step": 180
|
276 |
+
},
|
277 |
+
{
|
278 |
+
"epoch": 0.2,
|
279 |
+
"learning_rate": 4.8417076618132434e-06,
|
280 |
+
"logits/chosen": 0.017135417088866234,
|
281 |
+
"logits/rejected": 0.09486501663923264,
|
282 |
+
"logps/chosen": -600.1754760742188,
|
283 |
+
"logps/rejected": -609.9652709960938,
|
284 |
+
"loss": 0.1887,
|
285 |
+
"rewards/accuracies": 0.543749988079071,
|
286 |
+
"rewards/chosen": -0.11600615829229355,
|
287 |
+
"rewards/margins": 0.041202057152986526,
|
288 |
+
"rewards/rejected": -0.15720821917057037,
|
289 |
+
"step": 190
|
290 |
+
},
|
291 |
+
{
|
292 |
+
"epoch": 0.21,
|
293 |
+
"learning_rate": 4.807464225455655e-06,
|
294 |
+
"logits/chosen": -0.032647065818309784,
|
295 |
+
"logits/rejected": 0.09240031987428665,
|
296 |
+
"logps/chosen": -527.2655029296875,
|
297 |
+
"logps/rejected": -595.9906005859375,
|
298 |
+
"loss": 0.1949,
|
299 |
+
"rewards/accuracies": 0.543749988079071,
|
300 |
+
"rewards/chosen": -0.1231132298707962,
|
301 |
+
"rewards/margins": 0.051538724452257156,
|
302 |
+
"rewards/rejected": -0.17465195059776306,
|
303 |
+
"step": 200
|
304 |
+
},
|
305 |
+
{
|
306 |
+
"epoch": 0.22,
|
307 |
+
"learning_rate": 4.770016517582283e-06,
|
308 |
+
"logits/chosen": 0.03914088383316994,
|
309 |
+
"logits/rejected": 0.028707262128591537,
|
310 |
+
"logps/chosen": -524.7379760742188,
|
311 |
+
"logps/rejected": -570.7955322265625,
|
312 |
+
"loss": 0.1902,
|
313 |
+
"rewards/accuracies": 0.53125,
|
314 |
+
"rewards/chosen": -0.12313251197338104,
|
315 |
+
"rewards/margins": 0.04237721115350723,
|
316 |
+
"rewards/rejected": -0.16550973057746887,
|
317 |
+
"step": 210
|
318 |
+
},
|
319 |
+
{
|
320 |
+
"epoch": 0.23,
|
321 |
+
"learning_rate": 4.7294165401363616e-06,
|
322 |
+
"logits/chosen": 0.010909264907240868,
|
323 |
+
"logits/rejected": -0.024190250784158707,
|
324 |
+
"logps/chosen": -549.97607421875,
|
325 |
+
"logps/rejected": -590.9778442382812,
|
326 |
+
"loss": 0.1843,
|
327 |
+
"rewards/accuracies": 0.5062500238418579,
|
328 |
+
"rewards/chosen": -0.14806947112083435,
|
329 |
+
"rewards/margins": 0.04638643562793732,
|
330 |
+
"rewards/rejected": -0.19445592164993286,
|
331 |
+
"step": 220
|
332 |
+
},
|
333 |
+
{
|
334 |
+
"epoch": 0.25,
|
335 |
+
"learning_rate": 4.68572067247573e-06,
|
336 |
+
"logits/chosen": -0.018162641674280167,
|
337 |
+
"logits/rejected": 0.000972352921962738,
|
338 |
+
"logps/chosen": -549.0392456054688,
|
339 |
+
"logps/rejected": -598.3811645507812,
|
340 |
+
"loss": 0.2025,
|
341 |
+
"rewards/accuracies": 0.5062500238418579,
|
342 |
+
"rewards/chosen": -0.1491001546382904,
|
343 |
+
"rewards/margins": 0.04847537726163864,
|
344 |
+
"rewards/rejected": -0.19757553935050964,
|
345 |
+
"step": 230
|
346 |
+
},
|
347 |
+
{
|
348 |
+
"epoch": 0.26,
|
349 |
+
"learning_rate": 4.638989593081364e-06,
|
350 |
+
"logits/chosen": -0.12062356621026993,
|
351 |
+
"logits/rejected": 0.04868536815047264,
|
352 |
+
"logps/chosen": -484.58685302734375,
|
353 |
+
"logps/rejected": -516.2865600585938,
|
354 |
+
"loss": 0.1908,
|
355 |
+
"rewards/accuracies": 0.46875,
|
356 |
+
"rewards/chosen": -0.1070113405585289,
|
357 |
+
"rewards/margins": 0.028431424871087074,
|
358 |
+
"rewards/rejected": -0.13544276356697083,
|
359 |
+
"step": 240
|
360 |
+
},
|
361 |
+
{
|
362 |
+
"epoch": 0.27,
|
363 |
+
"learning_rate": 4.5892881952959015e-06,
|
364 |
+
"logits/chosen": -0.041638366878032684,
|
365 |
+
"logits/rejected": 0.0221172496676445,
|
366 |
+
"logps/chosen": -507.4730529785156,
|
367 |
+
"logps/rejected": -527.9345703125,
|
368 |
+
"loss": 0.2052,
|
369 |
+
"rewards/accuracies": 0.4937500059604645,
|
370 |
+
"rewards/chosen": -0.09331385791301727,
|
371 |
+
"rewards/margins": 0.03461749479174614,
|
372 |
+
"rewards/rejected": -0.1279313564300537,
|
373 |
+
"step": 250
|
374 |
+
},
|
375 |
+
{
|
376 |
+
"epoch": 0.28,
|
377 |
+
"learning_rate": 4.536685497209182e-06,
|
378 |
+
"logits/chosen": -0.05273251608014107,
|
379 |
+
"logits/rejected": -0.022044766694307327,
|
380 |
+
"logps/chosen": -538.548583984375,
|
381 |
+
"logps/rejected": -590.6500854492188,
|
382 |
+
"loss": 0.182,
|
383 |
+
"rewards/accuracies": 0.5,
|
384 |
+
"rewards/chosen": -0.10730306804180145,
|
385 |
+
"rewards/margins": 0.039316385984420776,
|
386 |
+
"rewards/rejected": -0.14661946892738342,
|
387 |
+
"step": 260
|
388 |
+
},
|
389 |
+
{
|
390 |
+
"epoch": 0.29,
|
391 |
+
"learning_rate": 4.481254545815943e-06,
|
392 |
+
"logits/chosen": -0.12279339134693146,
|
393 |
+
"logits/rejected": -0.079288050532341,
|
394 |
+
"logps/chosen": -560.711181640625,
|
395 |
+
"logps/rejected": -635.7985229492188,
|
396 |
+
"loss": 0.1845,
|
397 |
+
"rewards/accuracies": 0.5562499761581421,
|
398 |
+
"rewards/chosen": -0.12682856619358063,
|
399 |
+
"rewards/margins": 0.0557611808180809,
|
400 |
+
"rewards/rejected": -0.18258973956108093,
|
401 |
+
"step": 270
|
402 |
+
},
|
403 |
+
{
|
404 |
+
"epoch": 0.3,
|
405 |
+
"learning_rate": 4.42307231557875e-06,
|
406 |
+
"logits/chosen": -0.12978403270244598,
|
407 |
+
"logits/rejected": -0.05718718096613884,
|
408 |
+
"logps/chosen": -534.7046508789062,
|
409 |
+
"logps/rejected": -573.9546508789062,
|
410 |
+
"loss": 0.1872,
|
411 |
+
"rewards/accuracies": 0.4749999940395355,
|
412 |
+
"rewards/chosen": -0.12655052542686462,
|
413 |
+
"rewards/margins": 0.05336705967783928,
|
414 |
+
"rewards/rejected": -0.1799176186323166,
|
415 |
+
"step": 280
|
416 |
+
},
|
417 |
+
{
|
418 |
+
"epoch": 0.31,
|
419 |
+
"learning_rate": 4.3622196015370305e-06,
|
420 |
+
"logits/chosen": -0.13656684756278992,
|
421 |
+
"logits/rejected": -0.07923261821269989,
|
422 |
+
"logps/chosen": -537.6559448242188,
|
423 |
+
"logps/rejected": -627.1619873046875,
|
424 |
+
"loss": 0.1952,
|
425 |
+
"rewards/accuracies": 0.5562499761581421,
|
426 |
+
"rewards/chosen": -0.13470812141895294,
|
427 |
+
"rewards/margins": 0.05948293209075928,
|
428 |
+
"rewards/rejected": -0.1941910684108734,
|
429 |
+
"step": 290
|
430 |
+
},
|
431 |
+
{
|
432 |
+
"epoch": 0.32,
|
433 |
+
"learning_rate": 4.298780907110648e-06,
|
434 |
+
"logits/chosen": -0.11429516226053238,
|
435 |
+
"logits/rejected": -0.12869636714458466,
|
436 |
+
"logps/chosen": -543.2788696289062,
|
437 |
+
"logps/rejected": -558.6578369140625,
|
438 |
+
"loss": 0.1847,
|
439 |
+
"rewards/accuracies": 0.48750001192092896,
|
440 |
+
"rewards/chosen": -0.12002478539943695,
|
441 |
+
"rewards/margins": 0.050586897879838943,
|
442 |
+
"rewards/rejected": -0.1706116795539856,
|
443 |
+
"step": 300
|
444 |
+
},
|
445 |
+
{
|
446 |
+
"epoch": 0.33,
|
447 |
+
"learning_rate": 4.23284432675381e-06,
|
448 |
+
"logits/chosen": -0.18651030957698822,
|
449 |
+
"logits/rejected": -0.052459727972745895,
|
450 |
+
"logps/chosen": -461.1847229003906,
|
451 |
+
"logps/rejected": -506.2823181152344,
|
452 |
+
"loss": 0.1928,
|
453 |
+
"rewards/accuracies": 0.45625001192092896,
|
454 |
+
"rewards/chosen": -0.08158674091100693,
|
455 |
+
"rewards/margins": 0.045484792441129684,
|
456 |
+
"rewards/rejected": -0.1270715296268463,
|
457 |
+
"step": 310
|
458 |
+
},
|
459 |
+
{
|
460 |
+
"epoch": 0.34,
|
461 |
+
"learning_rate": 4.164501423622277e-06,
|
462 |
+
"logits/chosen": -0.1330818384885788,
|
463 |
+
"logits/rejected": -0.09265539795160294,
|
464 |
+
"logps/chosen": -483.45599365234375,
|
465 |
+
"logps/rejected": -515.7194213867188,
|
466 |
+
"loss": 0.1797,
|
467 |
+
"rewards/accuracies": 0.53125,
|
468 |
+
"rewards/chosen": -0.10115663707256317,
|
469 |
+
"rewards/margins": 0.04430658370256424,
|
470 |
+
"rewards/rejected": -0.14546321332454681,
|
471 |
+
"step": 320
|
472 |
+
},
|
473 |
+
{
|
474 |
+
"epoch": 0.35,
|
475 |
+
"learning_rate": 4.0938471024237355e-06,
|
476 |
+
"logits/chosen": -0.11196194589138031,
|
477 |
+
"logits/rejected": -0.09686783701181412,
|
478 |
+
"logps/chosen": -557.8707885742188,
|
479 |
+
"logps/rejected": -571.39794921875,
|
480 |
+
"loss": 0.1958,
|
481 |
+
"rewards/accuracies": 0.4937500059604645,
|
482 |
+
"rewards/chosen": -0.12075225263834,
|
483 |
+
"rewards/margins": 0.04514995589852333,
|
484 |
+
"rewards/rejected": -0.16590221226215363,
|
485 |
+
"step": 330
|
486 |
+
},
|
487 |
+
{
|
488 |
+
"epoch": 0.36,
|
489 |
+
"learning_rate": 4.020979477627907e-06,
|
490 |
+
"logits/chosen": -0.08174435794353485,
|
491 |
+
"logits/rejected": -0.06923134624958038,
|
492 |
+
"logps/chosen": -531.8570556640625,
|
493 |
+
"logps/rejected": -605.0074462890625,
|
494 |
+
"loss": 0.1889,
|
495 |
+
"rewards/accuracies": 0.581250011920929,
|
496 |
+
"rewards/chosen": -0.11230266094207764,
|
497 |
+
"rewards/margins": 0.06913084536790848,
|
498 |
+
"rewards/rejected": -0.18143349885940552,
|
499 |
+
"step": 340
|
500 |
+
},
|
501 |
+
{
|
502 |
+
"epoch": 0.37,
|
503 |
+
"learning_rate": 3.9459997372194105e-06,
|
504 |
+
"logits/chosen": -0.16061343252658844,
|
505 |
+
"logits/rejected": -0.027816006913781166,
|
506 |
+
"logps/chosen": -534.1594848632812,
|
507 |
+
"logps/rejected": -573.8477783203125,
|
508 |
+
"loss": 0.1926,
|
509 |
+
"rewards/accuracies": 0.46875,
|
510 |
+
"rewards/chosen": -0.0970136970281601,
|
511 |
+
"rewards/margins": 0.045039448887109756,
|
512 |
+
"rewards/rejected": -0.14205312728881836,
|
513 |
+
"step": 350
|
514 |
+
},
|
515 |
+
{
|
516 |
+
"epoch": 0.38,
|
517 |
+
"learning_rate": 3.869012002182573e-06,
|
518 |
+
"logits/chosen": -0.24527081847190857,
|
519 |
+
"logits/rejected": -0.1484527587890625,
|
520 |
+
"logps/chosen": -544.539306640625,
|
521 |
+
"logps/rejected": -564.9341430664062,
|
522 |
+
"loss": 0.1859,
|
523 |
+
"rewards/accuracies": 0.5687500238418579,
|
524 |
+
"rewards/chosen": -0.09417351335287094,
|
525 |
+
"rewards/margins": 0.056557249277830124,
|
526 |
+
"rewards/rejected": -0.15073075890541077,
|
527 |
+
"step": 360
|
528 |
+
},
|
529 |
+
{
|
530 |
+
"epoch": 0.39,
|
531 |
+
"learning_rate": 3.7901231819133104e-06,
|
532 |
+
"logits/chosen": -0.1722763478755951,
|
533 |
+
"logits/rejected": -0.17130622267723083,
|
534 |
+
"logps/chosen": -496.27313232421875,
|
535 |
+
"logps/rejected": -557.1398315429688,
|
536 |
+
"loss": 0.1877,
|
537 |
+
"rewards/accuracies": 0.48750001192092896,
|
538 |
+
"rewards/chosen": -0.0965786799788475,
|
539 |
+
"rewards/margins": 0.057069409638643265,
|
540 |
+
"rewards/rejected": -0.15364809334278107,
|
541 |
+
"step": 370
|
542 |
+
},
|
543 |
+
{
|
544 |
+
"epoch": 0.41,
|
545 |
+
"learning_rate": 3.709442825758875e-06,
|
546 |
+
"logits/chosen": -0.286950945854187,
|
547 |
+
"logits/rejected": -0.12660877406597137,
|
548 |
+
"logps/chosen": -487.8304138183594,
|
549 |
+
"logps/rejected": -506.80267333984375,
|
550 |
+
"loss": 0.1784,
|
551 |
+
"rewards/accuracies": 0.48124998807907104,
|
552 |
+
"rewards/chosen": -0.08378596603870392,
|
553 |
+
"rewards/margins": 0.04325443506240845,
|
554 |
+
"rewards/rejected": -0.12704041600227356,
|
555 |
+
"step": 380
|
556 |
+
},
|
557 |
+
{
|
558 |
+
"epoch": 0.42,
|
559 |
+
"learning_rate": 3.6270829708916113e-06,
|
560 |
+
"logits/chosen": -0.2721463441848755,
|
561 |
+
"logits/rejected": -0.19791728258132935,
|
562 |
+
"logps/chosen": -525.049560546875,
|
563 |
+
"logps/rejected": -564.6629028320312,
|
564 |
+
"loss": 0.1924,
|
565 |
+
"rewards/accuracies": 0.4749999940395355,
|
566 |
+
"rewards/chosen": -0.1035178154706955,
|
567 |
+
"rewards/margins": 0.03107512556016445,
|
568 |
+
"rewards/rejected": -0.1345929503440857,
|
569 |
+
"step": 390
|
570 |
+
},
|
571 |
+
{
|
572 |
+
"epoch": 0.43,
|
573 |
+
"learning_rate": 3.543157986727991e-06,
|
574 |
+
"logits/chosen": -0.17590856552124023,
|
575 |
+
"logits/rejected": -0.16738948225975037,
|
576 |
+
"logps/chosen": -520.5001831054688,
|
577 |
+
"logps/rejected": -564.5961303710938,
|
578 |
+
"loss": 0.1854,
|
579 |
+
"rewards/accuracies": 0.518750011920929,
|
580 |
+
"rewards/chosen": -0.09734812378883362,
|
581 |
+
"rewards/margins": 0.0474289208650589,
|
582 |
+
"rewards/rejected": -0.14477702975273132,
|
583 |
+
"step": 400
|
584 |
+
},
|
585 |
+
{
|
586 |
+
"epoch": 0.44,
|
587 |
+
"learning_rate": 3.4577844161089614e-06,
|
588 |
+
"logits/chosen": -0.17745746672153473,
|
589 |
+
"logits/rejected": -0.18353696167469025,
|
590 |
+
"logps/chosen": -508.34637451171875,
|
591 |
+
"logps/rejected": -577.1897583007812,
|
592 |
+
"loss": 0.1804,
|
593 |
+
"rewards/accuracies": 0.5375000238418579,
|
594 |
+
"rewards/chosen": -0.0865376815199852,
|
595 |
+
"rewards/margins": 0.049374908208847046,
|
596 |
+
"rewards/rejected": -0.13591258227825165,
|
597 |
+
"step": 410
|
598 |
+
},
|
599 |
+
{
|
600 |
+
"epoch": 0.45,
|
601 |
+
"learning_rate": 3.3710808134621577e-06,
|
602 |
+
"logits/chosen": -0.17098669707775116,
|
603 |
+
"logits/rejected": -0.13703958690166473,
|
604 |
+
"logps/chosen": -539.40087890625,
|
605 |
+
"logps/rejected": -593.8014526367188,
|
606 |
+
"loss": 0.1851,
|
607 |
+
"rewards/accuracies": 0.543749988079071,
|
608 |
+
"rewards/chosen": -0.09274087101221085,
|
609 |
+
"rewards/margins": 0.05765017122030258,
|
610 |
+
"rewards/rejected": -0.15039105713367462,
|
611 |
+
"step": 420
|
612 |
+
},
|
613 |
+
{
|
614 |
+
"epoch": 0.46,
|
615 |
+
"learning_rate": 3.2831675801707126e-06,
|
616 |
+
"logits/chosen": -0.20213007926940918,
|
617 |
+
"logits/rejected": -0.20745894312858582,
|
618 |
+
"logps/chosen": -453.65478515625,
|
619 |
+
"logps/rejected": -497.0008850097656,
|
620 |
+
"loss": 0.1824,
|
621 |
+
"rewards/accuracies": 0.512499988079071,
|
622 |
+
"rewards/chosen": -0.08567062765359879,
|
623 |
+
"rewards/margins": 0.04529280215501785,
|
624 |
+
"rewards/rejected": -0.13096341490745544,
|
625 |
+
"step": 430
|
626 |
+
},
|
627 |
+
{
|
628 |
+
"epoch": 0.47,
|
629 |
+
"learning_rate": 3.194166797377289e-06,
|
630 |
+
"logits/chosen": -0.21449732780456543,
|
631 |
+
"logits/rejected": -0.19523288309574127,
|
632 |
+
"logps/chosen": -547.9935302734375,
|
633 |
+
"logps/rejected": -572.2437744140625,
|
634 |
+
"loss": 0.1901,
|
635 |
+
"rewards/accuracies": 0.4625000059604645,
|
636 |
+
"rewards/chosen": -0.09648506343364716,
|
637 |
+
"rewards/margins": 0.03453432396054268,
|
638 |
+
"rewards/rejected": -0.13101938366889954,
|
639 |
+
"step": 440
|
640 |
+
},
|
641 |
+
{
|
642 |
+
"epoch": 0.48,
|
643 |
+
"learning_rate": 3.104202056455501e-06,
|
644 |
+
"logits/chosen": -0.22678379714488983,
|
645 |
+
"logits/rejected": -0.18668214976787567,
|
646 |
+
"logps/chosen": -519.3316650390625,
|
647 |
+
"logps/rejected": -561.0910034179688,
|
648 |
+
"loss": 0.1896,
|
649 |
+
"rewards/accuracies": 0.518750011920929,
|
650 |
+
"rewards/chosen": -0.09367823600769043,
|
651 |
+
"rewards/margins": 0.058413583785295486,
|
652 |
+
"rewards/rejected": -0.15209180116653442,
|
653 |
+
"step": 450
|
654 |
+
},
|
655 |
+
{
|
656 |
+
"epoch": 0.49,
|
657 |
+
"learning_rate": 3.013398287384144e-06,
|
658 |
+
"logits/chosen": -0.20922398567199707,
|
659 |
+
"logits/rejected": -0.15190599858760834,
|
660 |
+
"logps/chosen": -554.7764892578125,
|
661 |
+
"logps/rejected": -584.9015502929688,
|
662 |
+
"loss": 0.1777,
|
663 |
+
"rewards/accuracies": 0.5375000238418579,
|
664 |
+
"rewards/chosen": -0.11632993072271347,
|
665 |
+
"rewards/margins": 0.048789944499731064,
|
666 |
+
"rewards/rejected": -0.16511985659599304,
|
667 |
+
"step": 460
|
668 |
+
},
|
669 |
+
{
|
670 |
+
"epoch": 0.5,
|
671 |
+
"learning_rate": 2.9218815852625717e-06,
|
672 |
+
"logits/chosen": -0.2042142152786255,
|
673 |
+
"logits/rejected": -0.19644713401794434,
|
674 |
+
"logps/chosen": -522.6699829101562,
|
675 |
+
"logps/rejected": -589.4488525390625,
|
676 |
+
"loss": 0.189,
|
677 |
+
"rewards/accuracies": 0.5687500238418579,
|
678 |
+
"rewards/chosen": -0.10839296877384186,
|
679 |
+
"rewards/margins": 0.06792866438627243,
|
680 |
+
"rewards/rejected": -0.17632164061069489,
|
681 |
+
"step": 470
|
682 |
+
},
|
683 |
+
{
|
684 |
+
"epoch": 0.51,
|
685 |
+
"learning_rate": 2.829779035208113e-06,
|
686 |
+
"logits/chosen": -0.29581087827682495,
|
687 |
+
"logits/rejected": -0.17288121581077576,
|
688 |
+
"logps/chosen": -492.73297119140625,
|
689 |
+
"logps/rejected": -565.6483764648438,
|
690 |
+
"loss": 0.1819,
|
691 |
+
"rewards/accuracies": 0.46875,
|
692 |
+
"rewards/chosen": -0.10467328131198883,
|
693 |
+
"rewards/margins": 0.0681912824511528,
|
694 |
+
"rewards/rejected": -0.17286454141139984,
|
695 |
+
"step": 480
|
696 |
+
},
|
697 |
+
{
|
698 |
+
"epoch": 0.52,
|
699 |
+
"learning_rate": 2.737218535878705e-06,
|
700 |
+
"logits/chosen": -0.1768864393234253,
|
701 |
+
"logits/rejected": -0.19145308434963226,
|
702 |
+
"logps/chosen": -481.3701171875,
|
703 |
+
"logps/rejected": -552.0697021484375,
|
704 |
+
"loss": 0.1861,
|
705 |
+
"rewards/accuracies": 0.46875,
|
706 |
+
"rewards/chosen": -0.10367073863744736,
|
707 |
+
"rewards/margins": 0.0636112317442894,
|
708 |
+
"rewards/rejected": -0.16728197038173676,
|
709 |
+
"step": 490
|
710 |
+
},
|
711 |
+
{
|
712 |
+
"epoch": 0.53,
|
713 |
+
"learning_rate": 2.64432862186579e-06,
|
714 |
+
"logits/chosen": -0.25040799379348755,
|
715 |
+
"logits/rejected": -0.2705633044242859,
|
716 |
+
"logps/chosen": -473.15777587890625,
|
717 |
+
"logps/rejected": -521.9263916015625,
|
718 |
+
"loss": 0.1845,
|
719 |
+
"rewards/accuracies": 0.4625000059604645,
|
720 |
+
"rewards/chosen": -0.10329292714595795,
|
721 |
+
"rewards/margins": 0.04386230558156967,
|
722 |
+
"rewards/rejected": -0.1471552550792694,
|
723 |
+
"step": 500
|
724 |
+
},
|
725 |
+
{
|
726 |
+
"epoch": 0.54,
|
727 |
+
"learning_rate": 2.551238285204126e-06,
|
728 |
+
"logits/chosen": -0.22839005291461945,
|
729 |
+
"logits/rejected": -0.18522998690605164,
|
730 |
+
"logps/chosen": -562.2581176757812,
|
731 |
+
"logps/rejected": -602.7523193359375,
|
732 |
+
"loss": 0.1852,
|
733 |
+
"rewards/accuracies": 0.5,
|
734 |
+
"rewards/chosen": -0.11718226969242096,
|
735 |
+
"rewards/margins": 0.054385870695114136,
|
736 |
+
"rewards/rejected": -0.1715681403875351,
|
737 |
+
"step": 510
|
738 |
+
},
|
739 |
+
{
|
740 |
+
"epoch": 0.55,
|
741 |
+
"learning_rate": 2.4580767962463688e-06,
|
742 |
+
"logits/chosen": -0.28231528401374817,
|
743 |
+
"logits/rejected": -0.1746218502521515,
|
744 |
+
"logps/chosen": -508.0462951660156,
|
745 |
+
"logps/rejected": -521.466552734375,
|
746 |
+
"loss": 0.1887,
|
747 |
+
"rewards/accuracies": 0.45625001192092896,
|
748 |
+
"rewards/chosen": -0.0966549962759018,
|
749 |
+
"rewards/margins": 0.04605900123715401,
|
750 |
+
"rewards/rejected": -0.1427139937877655,
|
751 |
+
"step": 520
|
752 |
+
},
|
753 |
+
{
|
754 |
+
"epoch": 0.57,
|
755 |
+
"learning_rate": 2.3649735241511546e-06,
|
756 |
+
"logits/chosen": -0.14483687281608582,
|
757 |
+
"logits/rejected": -0.18159925937652588,
|
758 |
+
"logps/chosen": -519.6622314453125,
|
759 |
+
"logps/rejected": -554.4771728515625,
|
760 |
+
"loss": 0.1881,
|
761 |
+
"rewards/accuracies": 0.5062500238418579,
|
762 |
+
"rewards/chosen": -0.10754968971014023,
|
763 |
+
"rewards/margins": 0.043608419597148895,
|
764 |
+
"rewards/rejected": -0.15115809440612793,
|
765 |
+
"step": 530
|
766 |
+
},
|
767 |
+
{
|
768 |
+
"epoch": 0.58,
|
769 |
+
"learning_rate": 2.2720577572339914e-06,
|
770 |
+
"logits/chosen": -0.27724790573120117,
|
771 |
+
"logits/rejected": -0.18303519487380981,
|
772 |
+
"logps/chosen": -502.09747314453125,
|
773 |
+
"logps/rejected": -529.7732543945312,
|
774 |
+
"loss": 0.1902,
|
775 |
+
"rewards/accuracies": 0.4312500059604645,
|
776 |
+
"rewards/chosen": -0.0977000966668129,
|
777 |
+
"rewards/margins": 0.04301925003528595,
|
778 |
+
"rewards/rejected": -0.14071933925151825,
|
779 |
+
"step": 540
|
780 |
+
},
|
781 |
+
{
|
782 |
+
"epoch": 0.59,
|
783 |
+
"learning_rate": 2.1794585234303995e-06,
|
784 |
+
"logits/chosen": -0.2885403633117676,
|
785 |
+
"logits/rejected": -0.16289584338665009,
|
786 |
+
"logps/chosen": -519.3963012695312,
|
787 |
+
"logps/rejected": -553.4032592773438,
|
788 |
+
"loss": 0.1871,
|
789 |
+
"rewards/accuracies": 0.5062500238418579,
|
790 |
+
"rewards/chosen": -0.10111876577138901,
|
791 |
+
"rewards/margins": 0.040404774248600006,
|
792 |
+
"rewards/rejected": -0.1415235549211502,
|
793 |
+
"step": 550
|
794 |
+
},
|
795 |
+
{
|
796 |
+
"epoch": 0.6,
|
797 |
+
"learning_rate": 2.0873044111206407e-06,
|
798 |
+
"logits/chosen": -0.23527821898460388,
|
799 |
+
"logits/rejected": -0.2247372567653656,
|
800 |
+
"logps/chosen": -481.41552734375,
|
801 |
+
"logps/rejected": -552.4132080078125,
|
802 |
+
"loss": 0.2026,
|
803 |
+
"rewards/accuracies": 0.45625001192092896,
|
804 |
+
"rewards/chosen": -0.10331599414348602,
|
805 |
+
"rewards/margins": 0.040959432721138,
|
806 |
+
"rewards/rejected": -0.14427544176578522,
|
807 |
+
"step": 560
|
808 |
+
},
|
809 |
+
{
|
810 |
+
"epoch": 0.61,
|
811 |
+
"learning_rate": 1.9957233905648293e-06,
|
812 |
+
"logits/chosen": -0.28348255157470703,
|
813 |
+
"logits/rejected": -0.26194503903388977,
|
814 |
+
"logps/chosen": -467.77740478515625,
|
815 |
+
"logps/rejected": -507.6676330566406,
|
816 |
+
"loss": 0.1887,
|
817 |
+
"rewards/accuracies": 0.4749999940395355,
|
818 |
+
"rewards/chosen": -0.10089198499917984,
|
819 |
+
"rewards/margins": 0.0405060276389122,
|
820 |
+
"rewards/rejected": -0.14139802753925323,
|
821 |
+
"step": 570
|
822 |
+
},
|
823 |
+
{
|
824 |
+
"epoch": 0.62,
|
825 |
+
"learning_rate": 1.904842636196402e-06,
|
826 |
+
"logits/chosen": -0.22403912246227264,
|
827 |
+
"logits/rejected": -0.19076624512672424,
|
828 |
+
"logps/chosen": -500.50982666015625,
|
829 |
+
"logps/rejected": -544.9527587890625,
|
830 |
+
"loss": 0.1793,
|
831 |
+
"rewards/accuracies": 0.550000011920929,
|
832 |
+
"rewards/chosen": -0.09000807255506516,
|
833 |
+
"rewards/margins": 0.05757290869951248,
|
834 |
+
"rewards/rejected": -0.14758098125457764,
|
835 |
+
"step": 580
|
836 |
+
},
|
837 |
+
{
|
838 |
+
"epoch": 0.63,
|
839 |
+
"learning_rate": 1.814788350020726e-06,
|
840 |
+
"logits/chosen": -0.25425633788108826,
|
841 |
+
"logits/rejected": -0.13311608135700226,
|
842 |
+
"logps/chosen": -523.557373046875,
|
843 |
+
"logps/rejected": -576.8714599609375,
|
844 |
+
"loss": 0.1667,
|
845 |
+
"rewards/accuracies": 0.5874999761581421,
|
846 |
+
"rewards/chosen": -0.09282426536083221,
|
847 |
+
"rewards/margins": 0.06447537243366241,
|
848 |
+
"rewards/rejected": -0.15729963779449463,
|
849 |
+
"step": 590
|
850 |
+
},
|
851 |
+
{
|
852 |
+
"epoch": 0.64,
|
853 |
+
"learning_rate": 1.725685586364051e-06,
|
854 |
+
"logits/chosen": -0.25314217805862427,
|
855 |
+
"logits/rejected": -0.2236749678850174,
|
856 |
+
"logps/chosen": -442.9320373535156,
|
857 |
+
"logps/rejected": -521.5167236328125,
|
858 |
+
"loss": 0.1816,
|
859 |
+
"rewards/accuracies": 0.45625001192092896,
|
860 |
+
"rewards/chosen": -0.08528304100036621,
|
861 |
+
"rewards/margins": 0.05453087016940117,
|
862 |
+
"rewards/rejected": -0.13981391489505768,
|
863 |
+
"step": 600
|
864 |
+
},
|
865 |
+
{
|
866 |
+
"epoch": 0.65,
|
867 |
+
"learning_rate": 1.6376580782162172e-06,
|
868 |
+
"logits/chosen": -0.2589682936668396,
|
869 |
+
"logits/rejected": -0.2686694264411926,
|
870 |
+
"logps/chosen": -501.1578674316406,
|
871 |
+
"logps/rejected": -545.9219970703125,
|
872 |
+
"loss": 0.1949,
|
873 |
+
"rewards/accuracies": 0.46875,
|
874 |
+
"rewards/chosen": -0.09304684400558472,
|
875 |
+
"rewards/margins": 0.06486930698156357,
|
876 |
+
"rewards/rejected": -0.1579161435365677,
|
877 |
+
"step": 610
|
878 |
+
},
|
879 |
+
{
|
880 |
+
"epoch": 0.66,
|
881 |
+
"learning_rate": 1.550828065408227e-06,
|
882 |
+
"logits/chosen": -0.15998974442481995,
|
883 |
+
"logits/rejected": -0.26897841691970825,
|
884 |
+
"logps/chosen": -483.9093322753906,
|
885 |
+
"logps/rejected": -574.64990234375,
|
886 |
+
"loss": 0.1784,
|
887 |
+
"rewards/accuracies": 0.5625,
|
888 |
+
"rewards/chosen": -0.0897846445441246,
|
889 |
+
"rewards/margins": 0.059298910200595856,
|
890 |
+
"rewards/rejected": -0.14908355474472046,
|
891 |
+
"step": 620
|
892 |
+
},
|
893 |
+
{
|
894 |
+
"epoch": 0.67,
|
895 |
+
"learning_rate": 1.4653161248633053e-06,
|
896 |
+
"logits/chosen": -0.30697402358055115,
|
897 |
+
"logits/rejected": -0.2922336459159851,
|
898 |
+
"logps/chosen": -446.08642578125,
|
899 |
+
"logps/rejected": -485.38311767578125,
|
900 |
+
"loss": 0.1835,
|
901 |
+
"rewards/accuracies": 0.4625000059604645,
|
902 |
+
"rewards/chosen": -0.07366035133600235,
|
903 |
+
"rewards/margins": 0.05683215707540512,
|
904 |
+
"rewards/rejected": -0.13049249351024628,
|
905 |
+
"step": 630
|
906 |
+
},
|
907 |
+
{
|
908 |
+
"epoch": 0.68,
|
909 |
+
"learning_rate": 1.381241003157162e-06,
|
910 |
+
"logits/chosen": -0.27867692708969116,
|
911 |
+
"logits/rejected": -0.23723456263542175,
|
912 |
+
"logps/chosen": -472.146240234375,
|
913 |
+
"logps/rejected": -522.3912963867188,
|
914 |
+
"loss": 0.1886,
|
915 |
+
"rewards/accuracies": 0.45625001192092896,
|
916 |
+
"rewards/chosen": -0.08290112018585205,
|
917 |
+
"rewards/margins": 0.06648631393909454,
|
918 |
+
"rewards/rejected": -0.1493874490261078,
|
919 |
+
"step": 640
|
920 |
+
},
|
921 |
+
{
|
922 |
+
"epoch": 0.69,
|
923 |
+
"learning_rate": 1.298719451619979e-06,
|
924 |
+
"logits/chosen": -0.27338069677352905,
|
925 |
+
"logits/rejected": -0.0849432423710823,
|
926 |
+
"logps/chosen": -505.39404296875,
|
927 |
+
"logps/rejected": -564.4884643554688,
|
928 |
+
"loss": 0.1769,
|
929 |
+
"rewards/accuracies": 0.5562499761581421,
|
930 |
+
"rewards/chosen": -0.0819827988743782,
|
931 |
+
"rewards/margins": 0.07497727125883102,
|
932 |
+
"rewards/rejected": -0.15696007013320923,
|
933 |
+
"step": 650
|
934 |
+
},
|
935 |
+
{
|
936 |
+
"epoch": 0.7,
|
937 |
+
"learning_rate": 1.2178660642091036e-06,
|
938 |
+
"logits/chosen": -0.31306496262550354,
|
939 |
+
"logits/rejected": -0.15988986194133759,
|
940 |
+
"logps/chosen": -536.2052001953125,
|
941 |
+
"logps/rejected": -565.6595458984375,
|
942 |
+
"loss": 0.1917,
|
943 |
+
"rewards/accuracies": 0.4749999940395355,
|
944 |
+
"rewards/chosen": -0.10776883363723755,
|
945 |
+
"rewards/margins": 0.04764767736196518,
|
946 |
+
"rewards/rejected": -0.15541651844978333,
|
947 |
+
"step": 660
|
948 |
+
},
|
949 |
+
{
|
950 |
+
"epoch": 0.71,
|
951 |
+
"learning_rate": 1.1387931183775821e-06,
|
952 |
+
"logits/chosen": -0.1312873661518097,
|
953 |
+
"logits/rejected": -0.1946374773979187,
|
954 |
+
"logps/chosen": -489.32037353515625,
|
955 |
+
"logps/rejected": -532.713623046875,
|
956 |
+
"loss": 0.1923,
|
957 |
+
"rewards/accuracies": 0.5062500238418579,
|
958 |
+
"rewards/chosen": -0.10007087886333466,
|
959 |
+
"rewards/margins": 0.0466584786772728,
|
960 |
+
"rewards/rejected": -0.14672937989234924,
|
961 |
+
"step": 670
|
962 |
+
},
|
963 |
+
{
|
964 |
+
"epoch": 0.73,
|
965 |
+
"learning_rate": 1.061610419159532e-06,
|
966 |
+
"logits/chosen": -0.18406830728054047,
|
967 |
+
"logits/rejected": -0.18264801800251007,
|
968 |
+
"logps/chosen": -455.0455627441406,
|
969 |
+
"logps/rejected": -483.98748779296875,
|
970 |
+
"loss": 0.187,
|
971 |
+
"rewards/accuracies": 0.41874998807907104,
|
972 |
+
"rewards/chosen": -0.08491896092891693,
|
973 |
+
"rewards/margins": 0.042202599346637726,
|
974 |
+
"rewards/rejected": -0.12712153792381287,
|
975 |
+
"step": 680
|
976 |
+
},
|
977 |
+
{
|
978 |
+
"epoch": 0.74,
|
979 |
+
"learning_rate": 9.864251466888364e-07,
|
980 |
+
"logits/chosen": -0.2591504454612732,
|
981 |
+
"logits/rejected": -0.1554795801639557,
|
982 |
+
"logps/chosen": -488.55328369140625,
|
983 |
+
"logps/rejected": -532.9073486328125,
|
984 |
+
"loss": 0.1807,
|
985 |
+
"rewards/accuracies": 0.4749999940395355,
|
986 |
+
"rewards/chosen": -0.09263734519481659,
|
987 |
+
"rewards/margins": 0.04989578202366829,
|
988 |
+
"rewards/rejected": -0.1425331085920334,
|
989 |
+
"step": 690
|
990 |
+
},
|
991 |
+
{
|
992 |
+
"epoch": 0.75,
|
993 |
+
"learning_rate": 9.133417073629288e-07,
|
994 |
+
"logits/chosen": -0.28501999378204346,
|
995 |
+
"logits/rejected": -0.23185932636260986,
|
996 |
+
"logps/chosen": -485.5430603027344,
|
997 |
+
"logps/rejected": -541.1561889648438,
|
998 |
+
"loss": 0.1604,
|
999 |
+
"rewards/accuracies": 0.512499988079071,
|
1000 |
+
"rewards/chosen": -0.09013941138982773,
|
1001 |
+
"rewards/margins": 0.051737189292907715,
|
1002 |
+
"rewards/rejected": -0.14187659323215485,
|
1003 |
+
"step": 700
|
1004 |
+
},
|
1005 |
+
{
|
1006 |
+
"epoch": 0.76,
|
1007 |
+
"learning_rate": 8.424615888583332e-07,
|
1008 |
+
"logits/chosen": -0.25448185205459595,
|
1009 |
+
"logits/rejected": -0.13845598697662354,
|
1010 |
+
"logps/chosen": -504.44085693359375,
|
1011 |
+
"logps/rejected": -557.7171630859375,
|
1012 |
+
"loss": 0.1875,
|
1013 |
+
"rewards/accuracies": 0.518750011920929,
|
1014 |
+
"rewards/chosen": -0.09454745799303055,
|
1015 |
+
"rewards/margins": 0.0519348680973053,
|
1016 |
+
"rewards/rejected": -0.14648231863975525,
|
1017 |
+
"step": 710
|
1018 |
+
},
|
1019 |
+
{
|
1020 |
+
"epoch": 0.77,
|
1021 |
+
"learning_rate": 7.738832191993092e-07,
|
1022 |
+
"logits/chosen": -0.20559599995613098,
|
1023 |
+
"logits/rejected": -0.1910923421382904,
|
1024 |
+
"logps/chosen": -518.5410766601562,
|
1025 |
+
"logps/rejected": -559.4114379882812,
|
1026 |
+
"loss": 0.1781,
|
1027 |
+
"rewards/accuracies": 0.512499988079071,
|
1028 |
+
"rewards/chosen": -0.09824297577142715,
|
1029 |
+
"rewards/margins": 0.051730893552303314,
|
1030 |
+
"rewards/rejected": -0.14997386932373047,
|
1031 |
+
"step": 720
|
1032 |
+
},
|
1033 |
+
{
|
1034 |
+
"epoch": 0.78,
|
1035 |
+
"learning_rate": 7.077018300752917e-07,
|
1036 |
+
"logits/chosen": -0.20453593134880066,
|
1037 |
+
"logits/rejected": -0.22350621223449707,
|
1038 |
+
"logps/chosen": -517.8855590820312,
|
1039 |
+
"logps/rejected": -554.9312744140625,
|
1040 |
+
"loss": 0.173,
|
1041 |
+
"rewards/accuracies": 0.53125,
|
1042 |
+
"rewards/chosen": -0.0931503102183342,
|
1043 |
+
"rewards/margins": 0.059033893048763275,
|
1044 |
+
"rewards/rejected": -0.15218421816825867,
|
1045 |
+
"step": 730
|
1046 |
+
},
|
1047 |
+
{
|
1048 |
+
"epoch": 0.79,
|
1049 |
+
"learning_rate": 6.440093245969342e-07,
|
1050 |
+
"logits/chosen": -0.24915683269500732,
|
1051 |
+
"logits/rejected": -0.14556431770324707,
|
1052 |
+
"logps/chosen": -458.88006591796875,
|
1053 |
+
"logps/rejected": -483.50872802734375,
|
1054 |
+
"loss": 0.1791,
|
1055 |
+
"rewards/accuracies": 0.48124998807907104,
|
1056 |
+
"rewards/chosen": -0.0776047632098198,
|
1057 |
+
"rewards/margins": 0.055497486144304276,
|
1058 |
+
"rewards/rejected": -0.13310226798057556,
|
1059 |
+
"step": 740
|
1060 |
+
},
|
1061 |
+
{
|
1062 |
+
"epoch": 0.8,
|
1063 |
+
"learning_rate": 5.828941496744075e-07,
|
1064 |
+
"logits/chosen": -0.22538790106773376,
|
1065 |
+
"logits/rejected": -0.16318151354789734,
|
1066 |
+
"logps/chosen": -516.1492309570312,
|
1067 |
+
"logps/rejected": -586.5814208984375,
|
1068 |
+
"loss": 0.1829,
|
1069 |
+
"rewards/accuracies": 0.5625,
|
1070 |
+
"rewards/chosen": -0.09301020205020905,
|
1071 |
+
"rewards/margins": 0.06428654491901398,
|
1072 |
+
"rewards/rejected": -0.15729674696922302,
|
1073 |
+
"step": 750
|
1074 |
+
},
|
1075 |
+
{
|
1076 |
+
"epoch": 0.81,
|
1077 |
+
"learning_rate": 5.244411731951671e-07,
|
1078 |
+
"logits/chosen": -0.2806158661842346,
|
1079 |
+
"logits/rejected": -0.066395103931427,
|
1080 |
+
"logps/chosen": -479.8614807128906,
|
1081 |
+
"logps/rejected": -503.1717224121094,
|
1082 |
+
"loss": 0.1733,
|
1083 |
+
"rewards/accuracies": 0.4625000059604645,
|
1084 |
+
"rewards/chosen": -0.09291915595531464,
|
1085 |
+
"rewards/margins": 0.04909076914191246,
|
1086 |
+
"rewards/rejected": -0.1420099288225174,
|
1087 |
+
"step": 760
|
1088 |
+
},
|
1089 |
+
{
|
1090 |
+
"epoch": 0.82,
|
1091 |
+
"learning_rate": 4.6873156617173594e-07,
|
1092 |
+
"logits/chosen": -0.325612872838974,
|
1093 |
+
"logits/rejected": -0.22937624156475067,
|
1094 |
+
"logps/chosen": -526.00341796875,
|
1095 |
+
"logps/rejected": -547.8123168945312,
|
1096 |
+
"loss": 0.1746,
|
1097 |
+
"rewards/accuracies": 0.5,
|
1098 |
+
"rewards/chosen": -0.09224705398082733,
|
1099 |
+
"rewards/margins": 0.054325349628925323,
|
1100 |
+
"rewards/rejected": -0.14657239615917206,
|
1101 |
+
"step": 770
|
1102 |
+
},
|
1103 |
+
{
|
1104 |
+
"epoch": 0.83,
|
1105 |
+
"learning_rate": 4.1584269002318653e-07,
|
1106 |
+
"logits/chosen": -0.2622816264629364,
|
1107 |
+
"logits/rejected": -0.15280409157276154,
|
1108 |
+
"logps/chosen": -533.4638061523438,
|
1109 |
+
"logps/rejected": -595.7601318359375,
|
1110 |
+
"loss": 0.1815,
|
1111 |
+
"rewards/accuracies": 0.518750011920929,
|
1112 |
+
"rewards/chosen": -0.09688106924295425,
|
1113 |
+
"rewards/margins": 0.05785801261663437,
|
1114 |
+
"rewards/rejected": -0.15473909676074982,
|
1115 |
+
"step": 780
|
1116 |
+
},
|
1117 |
+
{
|
1118 |
+
"epoch": 0.84,
|
1119 |
+
"learning_rate": 3.658479891468258e-07,
|
1120 |
+
"logits/chosen": -0.18345573544502258,
|
1121 |
+
"logits/rejected": -0.16111025214195251,
|
1122 |
+
"logps/chosen": -487.77886962890625,
|
1123 |
+
"logps/rejected": -538.51171875,
|
1124 |
+
"loss": 0.184,
|
1125 |
+
"rewards/accuracies": 0.48750001192092896,
|
1126 |
+
"rewards/chosen": -0.08573255687952042,
|
1127 |
+
"rewards/margins": 0.0516083724796772,
|
1128 |
+
"rewards/rejected": -0.13734093308448792,
|
1129 |
+
"step": 790
|
1130 |
+
},
|
1131 |
+
{
|
1132 |
+
"epoch": 0.85,
|
1133 |
+
"learning_rate": 3.18816888929272e-07,
|
1134 |
+
"logits/chosen": -0.2816532254219055,
|
1135 |
+
"logits/rejected": -0.15862765908241272,
|
1136 |
+
"logps/chosen": -512.4444580078125,
|
1137 |
+
"logps/rejected": -536.9342651367188,
|
1138 |
+
"loss": 0.1844,
|
1139 |
+
"rewards/accuracies": 0.512499988079071,
|
1140 |
+
"rewards/chosen": -0.09420600533485413,
|
1141 |
+
"rewards/margins": 0.05614888668060303,
|
1142 |
+
"rewards/rejected": -0.15035490691661835,
|
1143 |
+
"step": 800
|
1144 |
+
},
|
1145 |
+
{
|
1146 |
+
"epoch": 0.86,
|
1147 |
+
"learning_rate": 2.748146993385484e-07,
|
1148 |
+
"logits/chosen": -0.21904349327087402,
|
1149 |
+
"logits/rejected": -0.258176326751709,
|
1150 |
+
"logps/chosen": -549.0933837890625,
|
1151 |
+
"logps/rejected": -618.8422241210938,
|
1152 |
+
"loss": 0.18,
|
1153 |
+
"rewards/accuracies": 0.5874999761581421,
|
1154 |
+
"rewards/chosen": -0.10313485562801361,
|
1155 |
+
"rewards/margins": 0.07112576067447662,
|
1156 |
+
"rewards/rejected": -0.17426061630249023,
|
1157 |
+
"step": 810
|
1158 |
+
},
|
1159 |
+
{
|
1160 |
+
"epoch": 0.87,
|
1161 |
+
"learning_rate": 2.3390252423108077e-07,
|
1162 |
+
"logits/chosen": -0.3128640353679657,
|
1163 |
+
"logits/rejected": -0.10742131620645523,
|
1164 |
+
"logps/chosen": -525.4251098632812,
|
1165 |
+
"logps/rejected": -553.0595092773438,
|
1166 |
+
"loss": 0.1734,
|
1167 |
+
"rewards/accuracies": 0.574999988079071,
|
1168 |
+
"rewards/chosen": -0.08687300980091095,
|
1169 |
+
"rewards/margins": 0.06989692151546478,
|
1170 |
+
"rewards/rejected": -0.15676993131637573,
|
1171 |
+
"step": 820
|
1172 |
+
},
|
1173 |
+
{
|
1174 |
+
"epoch": 0.89,
|
1175 |
+
"learning_rate": 1.961371764995243e-07,
|
1176 |
+
"logits/chosen": -0.1870919167995453,
|
1177 |
+
"logits/rejected": -0.18910066783428192,
|
1178 |
+
"logps/chosen": -514.9647827148438,
|
1179 |
+
"logps/rejected": -548.2384643554688,
|
1180 |
+
"loss": 0.182,
|
1181 |
+
"rewards/accuracies": 0.4749999940395355,
|
1182 |
+
"rewards/chosen": -0.0961606353521347,
|
1183 |
+
"rewards/margins": 0.05352962762117386,
|
1184 |
+
"rewards/rejected": -0.14969027042388916,
|
1185 |
+
"step": 830
|
1186 |
+
},
|
1187 |
+
{
|
1188 |
+
"epoch": 0.9,
|
1189 |
+
"learning_rate": 1.61571099179261e-07,
|
1190 |
+
"logits/chosen": -0.31641727685928345,
|
1191 |
+
"logits/rejected": -0.21929411590099335,
|
1192 |
+
"logps/chosen": -452.8780212402344,
|
1193 |
+
"logps/rejected": -497.7808532714844,
|
1194 |
+
"loss": 0.1887,
|
1195 |
+
"rewards/accuracies": 0.5062500238418579,
|
1196 |
+
"rewards/chosen": -0.08406294882297516,
|
1197 |
+
"rewards/margins": 0.06199796125292778,
|
1198 |
+
"rewards/rejected": -0.14606089890003204,
|
1199 |
+
"step": 840
|
1200 |
+
},
|
1201 |
+
{
|
1202 |
+
"epoch": 0.91,
|
1203 |
+
"learning_rate": 1.3025229262312367e-07,
|
1204 |
+
"logits/chosen": -0.2824193239212036,
|
1205 |
+
"logits/rejected": -0.2219020426273346,
|
1206 |
+
"logps/chosen": -516.4136962890625,
|
1207 |
+
"logps/rejected": -553.1406860351562,
|
1208 |
+
"loss": 0.1761,
|
1209 |
+
"rewards/accuracies": 0.48124998807907104,
|
1210 |
+
"rewards/chosen": -0.09720293432474136,
|
1211 |
+
"rewards/margins": 0.046975888311862946,
|
1212 |
+
"rewards/rejected": -0.1441788375377655,
|
1213 |
+
"step": 850
|
1214 |
+
},
|
1215 |
+
{
|
1216 |
+
"epoch": 0.92,
|
1217 |
+
"learning_rate": 1.0222424784546853e-07,
|
1218 |
+
"logits/chosen": -0.09483526647090912,
|
1219 |
+
"logits/rejected": -0.24578902125358582,
|
1220 |
+
"logps/chosen": -496.60028076171875,
|
1221 |
+
"logps/rejected": -558.0816650390625,
|
1222 |
+
"loss": 0.175,
|
1223 |
+
"rewards/accuracies": 0.4937500059604645,
|
1224 |
+
"rewards/chosen": -0.0964212641119957,
|
1225 |
+
"rewards/margins": 0.050969939678907394,
|
1226 |
+
"rewards/rejected": -0.14739122986793518,
|
1227 |
+
"step": 860
|
1228 |
+
},
|
1229 |
+
{
|
1230 |
+
"epoch": 0.93,
|
1231 |
+
"learning_rate": 7.752588612816553e-08,
|
1232 |
+
"logits/chosen": -0.20135729014873505,
|
1233 |
+
"logits/rejected": -0.25128036737442017,
|
1234 |
+
"logps/chosen": -545.1806640625,
|
1235 |
+
"logps/rejected": -568.894775390625,
|
1236 |
+
"loss": 0.1815,
|
1237 |
+
"rewards/accuracies": 0.5,
|
1238 |
+
"rewards/chosen": -0.09586036950349808,
|
1239 |
+
"rewards/margins": 0.04870045185089111,
|
1240 |
+
"rewards/rejected": -0.1445608288049698,
|
1241 |
+
"step": 870
|
1242 |
+
},
|
1243 |
+
{
|
1244 |
+
"epoch": 0.94,
|
1245 |
+
"learning_rate": 5.619150497236991e-08,
|
1246 |
+
"logits/chosen": -0.2596682012081146,
|
1247 |
+
"logits/rejected": -0.22516381740570068,
|
1248 |
+
"logps/chosen": -498.10028076171875,
|
1249 |
+
"logps/rejected": -564.5962524414062,
|
1250 |
+
"loss": 0.1708,
|
1251 |
+
"rewards/accuracies": 0.574999988079071,
|
1252 |
+
"rewards/chosen": -0.09488777071237564,
|
1253 |
+
"rewards/margins": 0.06974340975284576,
|
1254 |
+
"rewards/rejected": -0.164631187915802,
|
1255 |
+
"step": 880
|
1256 |
+
},
|
1257 |
+
{
|
1258 |
+
"epoch": 0.95,
|
1259 |
+
"learning_rate": 3.825073047112743e-08,
|
1260 |
+
"logits/chosen": -0.2944473624229431,
|
1261 |
+
"logits/rejected": -0.2125546932220459,
|
1262 |
+
"logps/chosen": -502.841552734375,
|
1263 |
+
"logps/rejected": -546.7924194335938,
|
1264 |
+
"loss": 0.1826,
|
1265 |
+
"rewards/accuracies": 0.5,
|
1266 |
+
"rewards/chosen": -0.08710362762212753,
|
1267 |
+
"rewards/margins": 0.059937745332717896,
|
1268 |
+
"rewards/rejected": -0.14704139530658722,
|
1269 |
+
"step": 890
|
1270 |
+
},
|
1271 |
+
{
|
1272 |
+
"epoch": 0.96,
|
1273 |
+
"learning_rate": 2.372847616895685e-08,
|
1274 |
+
"logits/chosen": -0.2098701447248459,
|
1275 |
+
"logits/rejected": -0.2635635733604431,
|
1276 |
+
"logps/chosen": -481.09478759765625,
|
1277 |
+
"logps/rejected": -569.34326171875,
|
1278 |
+
"loss": 0.1763,
|
1279 |
+
"rewards/accuracies": 0.5,
|
1280 |
+
"rewards/chosen": -0.09881605207920074,
|
1281 |
+
"rewards/margins": 0.06372065842151642,
|
1282 |
+
"rewards/rejected": -0.16253669559955597,
|
1283 |
+
"step": 900
|
1284 |
+
},
|
1285 |
+
{
|
1286 |
+
"epoch": 0.97,
|
1287 |
+
"learning_rate": 1.264490846553279e-08,
|
1288 |
+
"logits/chosen": -0.20950980484485626,
|
1289 |
+
"logits/rejected": -0.2304944545030594,
|
1290 |
+
"logps/chosen": -508.41241455078125,
|
1291 |
+
"logps/rejected": -574.2259521484375,
|
1292 |
+
"loss": 0.1727,
|
1293 |
+
"rewards/accuracies": 0.5375000238418579,
|
1294 |
+
"rewards/chosen": -0.0780680924654007,
|
1295 |
+
"rewards/margins": 0.06727245450019836,
|
1296 |
+
"rewards/rejected": -0.14534054696559906,
|
1297 |
+
"step": 910
|
1298 |
+
},
|
1299 |
+
{
|
1300 |
+
"epoch": 0.98,
|
1301 |
+
"learning_rate": 5.015418611516165e-09,
|
1302 |
+
"logits/chosen": -0.2829793095588684,
|
1303 |
+
"logits/rejected": -0.2987596392631531,
|
1304 |
+
"logps/chosen": -475.51031494140625,
|
1305 |
+
"logps/rejected": -521.2598876953125,
|
1306 |
+
"loss": 0.1865,
|
1307 |
+
"rewards/accuracies": 0.5249999761581421,
|
1308 |
+
"rewards/chosen": -0.09187036752700806,
|
1309 |
+
"rewards/margins": 0.06315977871417999,
|
1310 |
+
"rewards/rejected": -0.15503014624118805,
|
1311 |
+
"step": 920
|
1312 |
+
},
|
1313 |
+
{
|
1314 |
+
"epoch": 0.99,
|
1315 |
+
"learning_rate": 8.506013354186993e-10,
|
1316 |
+
"logits/chosen": -0.23163847625255585,
|
1317 |
+
"logits/rejected": -0.24427077174186707,
|
1318 |
+
"logps/chosen": -477.51507568359375,
|
1319 |
+
"logps/rejected": -542.298828125,
|
1320 |
+
"loss": 0.1848,
|
1321 |
+
"rewards/accuracies": 0.5249999761581421,
|
1322 |
+
"rewards/chosen": -0.0930924192070961,
|
1323 |
+
"rewards/margins": 0.06024498865008354,
|
1324 |
+
"rewards/rejected": -0.15333738923072815,
|
1325 |
+
"step": 930
|
1326 |
+
},
|
1327 |
+
{
|
1328 |
+
"epoch": 1.0,
|
1329 |
+
"step": 937,
|
1330 |
+
"total_flos": 0.0,
|
1331 |
+
"train_loss": 0.1881250925163322,
|
1332 |
+
"train_runtime": 7837.4153,
|
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 |
+
}
|