Model save
Browse files- README.md +62 -0
- adapter_model.safetensors +1 -1
- all_results.json +8 -0
- runs/May22_15-21-06_gpu4-119-5/events.out.tfevents.1716355436.gpu4-119-5.2823951.0 +2 -2
- train_results.json +8 -0
- trainer_state.json +1346 -0
README.md
ADDED
@@ -0,0 +1,62 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: apache-2.0
|
3 |
+
library_name: peft
|
4 |
+
tags:
|
5 |
+
- trl
|
6 |
+
- dpo
|
7 |
+
- generated_from_trainer
|
8 |
+
base_model: DUAL-GPO/phi-2-irepo-chatml-merged-i1
|
9 |
+
model-index:
|
10 |
+
- name: phi-2-irepo-chatml-v6-i2
|
11 |
+
results: []
|
12 |
+
---
|
13 |
+
|
14 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
15 |
+
should probably proofread and complete it, then remove this comment. -->
|
16 |
+
|
17 |
+
# phi-2-irepo-chatml-v6-i2
|
18 |
+
|
19 |
+
This model is a fine-tuned version of [DUAL-GPO/phi-2-irepo-chatml-merged-i1](https://huggingface.co/DUAL-GPO/phi-2-irepo-chatml-merged-i1) on the None dataset.
|
20 |
+
|
21 |
+
## Model description
|
22 |
+
|
23 |
+
More information needed
|
24 |
+
|
25 |
+
## Intended uses & limitations
|
26 |
+
|
27 |
+
More information needed
|
28 |
+
|
29 |
+
## Training and evaluation data
|
30 |
+
|
31 |
+
More information needed
|
32 |
+
|
33 |
+
## Training procedure
|
34 |
+
|
35 |
+
### Training hyperparameters
|
36 |
+
|
37 |
+
The following hyperparameters were used during training:
|
38 |
+
- learning_rate: 5e-06
|
39 |
+
- train_batch_size: 4
|
40 |
+
- eval_batch_size: 4
|
41 |
+
- seed: 42
|
42 |
+
- distributed_type: multi-GPU
|
43 |
+
- num_devices: 2
|
44 |
+
- gradient_accumulation_steps: 4
|
45 |
+
- total_train_batch_size: 32
|
46 |
+
- total_eval_batch_size: 8
|
47 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
48 |
+
- lr_scheduler_type: cosine
|
49 |
+
- lr_scheduler_warmup_ratio: 0.1
|
50 |
+
- num_epochs: 1
|
51 |
+
|
52 |
+
### Training results
|
53 |
+
|
54 |
+
|
55 |
+
|
56 |
+
### Framework versions
|
57 |
+
|
58 |
+
- PEFT 0.7.1
|
59 |
+
- Transformers 4.36.2
|
60 |
+
- Pytorch 2.1.2+cu121
|
61 |
+
- Datasets 2.14.6
|
62 |
+
- 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 sha256:45413e3c0a1f35e6e3689e6d91f7b484bf189e63b3e3a61ede8008314bdfa230
|
3 |
size 335579632
|
all_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 1.0,
|
3 |
+
"train_loss": 0.08862862108611247,
|
4 |
+
"train_runtime": 3263.0882,
|
5 |
+
"train_samples": 30000,
|
6 |
+
"train_samples_per_second": 9.194,
|
7 |
+
"train_steps_per_second": 0.287
|
8 |
+
}
|
runs/May22_15-21-06_gpu4-119-5/events.out.tfevents.1716355436.gpu4-119-5.2823951.0
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:b8754d0855a3490d53e9729f3c338a9f30ff9a097606de8214a7cd81df7b0918
|
3 |
+
size 32525
|
train_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 1.0,
|
3 |
+
"train_loss": 0.08862862108611247,
|
4 |
+
"train_runtime": 3263.0882,
|
5 |
+
"train_samples": 30000,
|
6 |
+
"train_samples_per_second": 9.194,
|
7 |
+
"train_steps_per_second": 0.287
|
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": -0.1885092854499817,
|
15 |
+
"logits/rejected": -0.3158565163612366,
|
16 |
+
"logps/chosen": -579.0175170898438,
|
17 |
+
"logps/rejected": -485.2366638183594,
|
18 |
+
"loss": 0.2285,
|
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": -0.1450551152229309,
|
29 |
+
"logits/rejected": -0.07828879356384277,
|
30 |
+
"logps/chosen": -490.4548034667969,
|
31 |
+
"logps/rejected": -480.7243347167969,
|
32 |
+
"loss": 0.21,
|
33 |
+
"rewards/accuracies": 0.375,
|
34 |
+
"rewards/chosen": -5.791801231680438e-05,
|
35 |
+
"rewards/margins": 4.912323129246943e-05,
|
36 |
+
"rewards/rejected": -0.00010704125452321023,
|
37 |
+
"step": 10
|
38 |
+
},
|
39 |
+
{
|
40 |
+
"epoch": 0.02,
|
41 |
+
"learning_rate": 1.0638297872340427e-06,
|
42 |
+
"logits/chosen": -0.17377130687236786,
|
43 |
+
"logits/rejected": -0.2042241394519806,
|
44 |
+
"logps/chosen": -491.5914001464844,
|
45 |
+
"logps/rejected": -509.80047607421875,
|
46 |
+
"loss": 0.2024,
|
47 |
+
"rewards/accuracies": 0.40625,
|
48 |
+
"rewards/chosen": 2.5502964490442537e-05,
|
49 |
+
"rewards/margins": 0.00020013886387459934,
|
50 |
+
"rewards/rejected": -0.00017463587573729455,
|
51 |
+
"step": 20
|
52 |
+
},
|
53 |
+
{
|
54 |
+
"epoch": 0.03,
|
55 |
+
"learning_rate": 1.595744680851064e-06,
|
56 |
+
"logits/chosen": -0.13665731251239777,
|
57 |
+
"logits/rejected": -0.13284753262996674,
|
58 |
+
"logps/chosen": -530.023681640625,
|
59 |
+
"logps/rejected": -542.8099975585938,
|
60 |
+
"loss": 0.2154,
|
61 |
+
"rewards/accuracies": 0.4312500059604645,
|
62 |
+
"rewards/chosen": -0.0006009428761899471,
|
63 |
+
"rewards/margins": 7.30833999114111e-05,
|
64 |
+
"rewards/rejected": -0.0006740263779647648,
|
65 |
+
"step": 30
|
66 |
+
},
|
67 |
+
{
|
68 |
+
"epoch": 0.04,
|
69 |
+
"learning_rate": 2.1276595744680853e-06,
|
70 |
+
"logits/chosen": -0.10384176671504974,
|
71 |
+
"logits/rejected": -0.09949521720409393,
|
72 |
+
"logps/chosen": -578.4327392578125,
|
73 |
+
"logps/rejected": -594.6671752929688,
|
74 |
+
"loss": 0.2087,
|
75 |
+
"rewards/accuracies": 0.5062500238418579,
|
76 |
+
"rewards/chosen": -0.0006206859834492207,
|
77 |
+
"rewards/margins": 0.0007750510121695697,
|
78 |
+
"rewards/rejected": -0.0013957370538264513,
|
79 |
+
"step": 40
|
80 |
+
},
|
81 |
+
{
|
82 |
+
"epoch": 0.05,
|
83 |
+
"learning_rate": 2.6595744680851065e-06,
|
84 |
+
"logits/chosen": -0.10368921607732773,
|
85 |
+
"logits/rejected": -0.13414844870567322,
|
86 |
+
"logps/chosen": -511.35809326171875,
|
87 |
+
"logps/rejected": -534.3345947265625,
|
88 |
+
"loss": 0.21,
|
89 |
+
"rewards/accuracies": 0.4000000059604645,
|
90 |
+
"rewards/chosen": -0.0013925038510933518,
|
91 |
+
"rewards/margins": 0.0007838442688807845,
|
92 |
+
"rewards/rejected": -0.00217634835280478,
|
93 |
+
"step": 50
|
94 |
+
},
|
95 |
+
{
|
96 |
+
"epoch": 0.06,
|
97 |
+
"learning_rate": 3.191489361702128e-06,
|
98 |
+
"logits/chosen": -0.22053027153015137,
|
99 |
+
"logits/rejected": -0.08493414521217346,
|
100 |
+
"logps/chosen": -547.6785278320312,
|
101 |
+
"logps/rejected": -560.6593627929688,
|
102 |
+
"loss": 0.2124,
|
103 |
+
"rewards/accuracies": 0.4749999940395355,
|
104 |
+
"rewards/chosen": -0.002587032737210393,
|
105 |
+
"rewards/margins": 0.0012574782595038414,
|
106 |
+
"rewards/rejected": -0.003844511229544878,
|
107 |
+
"step": 60
|
108 |
+
},
|
109 |
+
{
|
110 |
+
"epoch": 0.07,
|
111 |
+
"learning_rate": 3.723404255319149e-06,
|
112 |
+
"logits/chosen": -0.11390117555856705,
|
113 |
+
"logits/rejected": -0.09883277863264084,
|
114 |
+
"logps/chosen": -533.27978515625,
|
115 |
+
"logps/rejected": -538.4425048828125,
|
116 |
+
"loss": 0.2137,
|
117 |
+
"rewards/accuracies": 0.4312500059604645,
|
118 |
+
"rewards/chosen": -0.005767723545432091,
|
119 |
+
"rewards/margins": 0.0026725970674306154,
|
120 |
+
"rewards/rejected": -0.008440319448709488,
|
121 |
+
"step": 70
|
122 |
+
},
|
123 |
+
{
|
124 |
+
"epoch": 0.09,
|
125 |
+
"learning_rate": 4.255319148936171e-06,
|
126 |
+
"logits/chosen": -0.1157851368188858,
|
127 |
+
"logits/rejected": -0.17272169888019562,
|
128 |
+
"logps/chosen": -507.7423400878906,
|
129 |
+
"logps/rejected": -523.6263427734375,
|
130 |
+
"loss": 0.2106,
|
131 |
+
"rewards/accuracies": 0.46875,
|
132 |
+
"rewards/chosen": -0.00815755594521761,
|
133 |
+
"rewards/margins": 0.0021632923744618893,
|
134 |
+
"rewards/rejected": -0.010320847854018211,
|
135 |
+
"step": 80
|
136 |
+
},
|
137 |
+
{
|
138 |
+
"epoch": 0.1,
|
139 |
+
"learning_rate": 4.787234042553192e-06,
|
140 |
+
"logits/chosen": -0.21070297062397003,
|
141 |
+
"logits/rejected": -0.20428261160850525,
|
142 |
+
"logps/chosen": -488.6622009277344,
|
143 |
+
"logps/rejected": -501.21856689453125,
|
144 |
+
"loss": 0.2103,
|
145 |
+
"rewards/accuracies": 0.4749999940395355,
|
146 |
+
"rewards/chosen": -0.011966566555202007,
|
147 |
+
"rewards/margins": 0.002781093353405595,
|
148 |
+
"rewards/rejected": -0.014747659675776958,
|
149 |
+
"step": 90
|
150 |
+
},
|
151 |
+
{
|
152 |
+
"epoch": 0.11,
|
153 |
+
"learning_rate": 4.999375059004058e-06,
|
154 |
+
"logits/chosen": -0.12732869386672974,
|
155 |
+
"logits/rejected": -0.2185029536485672,
|
156 |
+
"logps/chosen": -523.1364135742188,
|
157 |
+
"logps/rejected": -521.9183349609375,
|
158 |
+
"loss": 0.2006,
|
159 |
+
"rewards/accuracies": 0.41874998807907104,
|
160 |
+
"rewards/chosen": -0.01658087968826294,
|
161 |
+
"rewards/margins": 0.0014435439370572567,
|
162 |
+
"rewards/rejected": -0.018024425953626633,
|
163 |
+
"step": 100
|
164 |
+
},
|
165 |
+
{
|
166 |
+
"epoch": 0.12,
|
167 |
+
"learning_rate": 4.9955571065548795e-06,
|
168 |
+
"logits/chosen": -0.23100972175598145,
|
169 |
+
"logits/rejected": -0.26281121373176575,
|
170 |
+
"logps/chosen": -547.6220092773438,
|
171 |
+
"logps/rejected": -568.766845703125,
|
172 |
+
"loss": 0.202,
|
173 |
+
"rewards/accuracies": 0.44999998807907104,
|
174 |
+
"rewards/chosen": -0.022633420303463936,
|
175 |
+
"rewards/margins": 0.007052992470562458,
|
176 |
+
"rewards/rejected": -0.02968641184270382,
|
177 |
+
"step": 110
|
178 |
+
},
|
179 |
+
{
|
180 |
+
"epoch": 0.13,
|
181 |
+
"learning_rate": 4.9882736864879e-06,
|
182 |
+
"logits/chosen": -0.26259124279022217,
|
183 |
+
"logits/rejected": -0.22674131393432617,
|
184 |
+
"logps/chosen": -564.6571044921875,
|
185 |
+
"logps/rejected": -574.2418212890625,
|
186 |
+
"loss": 0.2018,
|
187 |
+
"rewards/accuracies": 0.45625001192092896,
|
188 |
+
"rewards/chosen": -0.03153581544756889,
|
189 |
+
"rewards/margins": 0.011116450652480125,
|
190 |
+
"rewards/rejected": -0.04265226796269417,
|
191 |
+
"step": 120
|
192 |
+
},
|
193 |
+
{
|
194 |
+
"epoch": 0.14,
|
195 |
+
"learning_rate": 4.977534912960124e-06,
|
196 |
+
"logits/chosen": -0.19227923452854156,
|
197 |
+
"logits/rejected": -0.21840712428092957,
|
198 |
+
"logps/chosen": -534.7987060546875,
|
199 |
+
"logps/rejected": -557.68994140625,
|
200 |
+
"loss": 0.199,
|
201 |
+
"rewards/accuracies": 0.4437499940395355,
|
202 |
+
"rewards/chosen": -0.04380916431546211,
|
203 |
+
"rewards/margins": 0.012184342369437218,
|
204 |
+
"rewards/rejected": -0.05599350482225418,
|
205 |
+
"step": 130
|
206 |
+
},
|
207 |
+
{
|
208 |
+
"epoch": 0.15,
|
209 |
+
"learning_rate": 4.963355698422092e-06,
|
210 |
+
"logits/chosen": -0.2127476930618286,
|
211 |
+
"logits/rejected": -0.24109581112861633,
|
212 |
+
"logps/chosen": -527.8475952148438,
|
213 |
+
"logps/rejected": -546.1341552734375,
|
214 |
+
"loss": 0.1976,
|
215 |
+
"rewards/accuracies": 0.4749999940395355,
|
216 |
+
"rewards/chosen": -0.039759356528520584,
|
217 |
+
"rewards/margins": 0.020705662667751312,
|
218 |
+
"rewards/rejected": -0.060465019196271896,
|
219 |
+
"step": 140
|
220 |
+
},
|
221 |
+
{
|
222 |
+
"epoch": 0.16,
|
223 |
+
"learning_rate": 4.945755732909625e-06,
|
224 |
+
"logits/chosen": -0.29052820801734924,
|
225 |
+
"logits/rejected": -0.18534401059150696,
|
226 |
+
"logps/chosen": -568.7386474609375,
|
227 |
+
"logps/rejected": -623.6467895507812,
|
228 |
+
"loss": 0.203,
|
229 |
+
"rewards/accuracies": 0.46875,
|
230 |
+
"rewards/chosen": -0.05505699664354324,
|
231 |
+
"rewards/margins": 0.02482803724706173,
|
232 |
+
"rewards/rejected": -0.07988502830266953,
|
233 |
+
"step": 150
|
234 |
+
},
|
235 |
+
{
|
236 |
+
"epoch": 0.17,
|
237 |
+
"learning_rate": 4.924759456701167e-06,
|
238 |
+
"logits/chosen": -0.21109052002429962,
|
239 |
+
"logits/rejected": -0.23127368092536926,
|
240 |
+
"logps/chosen": -509.3150329589844,
|
241 |
+
"logps/rejected": -569.0074462890625,
|
242 |
+
"loss": 0.1995,
|
243 |
+
"rewards/accuracies": 0.45625001192092896,
|
244 |
+
"rewards/chosen": -0.04530967399477959,
|
245 |
+
"rewards/margins": 0.026868581771850586,
|
246 |
+
"rewards/rejected": -0.07217825949192047,
|
247 |
+
"step": 160
|
248 |
+
},
|
249 |
+
{
|
250 |
+
"epoch": 0.18,
|
251 |
+
"learning_rate": 4.900396026378671e-06,
|
252 |
+
"logits/chosen": -0.2945798635482788,
|
253 |
+
"logits/rejected": -0.25223153829574585,
|
254 |
+
"logps/chosen": -614.588134765625,
|
255 |
+
"logps/rejected": -705.424072265625,
|
256 |
+
"loss": 0.1997,
|
257 |
+
"rewards/accuracies": 0.48124998807907104,
|
258 |
+
"rewards/chosen": -0.07361334562301636,
|
259 |
+
"rewards/margins": 0.043305329978466034,
|
260 |
+
"rewards/rejected": -0.11691868305206299,
|
261 |
+
"step": 170
|
262 |
+
},
|
263 |
+
{
|
264 |
+
"epoch": 0.19,
|
265 |
+
"learning_rate": 4.872699274339169e-06,
|
266 |
+
"logits/chosen": -0.2314704954624176,
|
267 |
+
"logits/rejected": -0.11344774067401886,
|
268 |
+
"logps/chosen": -570.4790649414062,
|
269 |
+
"logps/rejected": -626.4537353515625,
|
270 |
+
"loss": 0.203,
|
271 |
+
"rewards/accuracies": 0.5062500238418579,
|
272 |
+
"rewards/chosen": -0.06511107087135315,
|
273 |
+
"rewards/margins": 0.03761782869696617,
|
274 |
+
"rewards/rejected": -0.10272890329360962,
|
275 |
+
"step": 180
|
276 |
+
},
|
277 |
+
{
|
278 |
+
"epoch": 0.2,
|
279 |
+
"learning_rate": 4.8417076618132434e-06,
|
280 |
+
"logits/chosen": -0.13913798332214355,
|
281 |
+
"logits/rejected": -0.23164553940296173,
|
282 |
+
"logps/chosen": -599.2550048828125,
|
283 |
+
"logps/rejected": -628.2545776367188,
|
284 |
+
"loss": 0.1984,
|
285 |
+
"rewards/accuracies": 0.5249999761581421,
|
286 |
+
"rewards/chosen": -0.06176919490098953,
|
287 |
+
"rewards/margins": 0.029368698596954346,
|
288 |
+
"rewards/rejected": -0.09113789349794388,
|
289 |
+
"step": 190
|
290 |
+
},
|
291 |
+
{
|
292 |
+
"epoch": 0.21,
|
293 |
+
"learning_rate": 4.807464225455655e-06,
|
294 |
+
"logits/chosen": -0.1732599437236786,
|
295 |
+
"logits/rejected": -0.24373570084571838,
|
296 |
+
"logps/chosen": -551.7747802734375,
|
297 |
+
"logps/rejected": -587.905517578125,
|
298 |
+
"loss": 0.1991,
|
299 |
+
"rewards/accuracies": 0.4375,
|
300 |
+
"rewards/chosen": -0.060781996697187424,
|
301 |
+
"rewards/margins": 0.024652186781167984,
|
302 |
+
"rewards/rejected": -0.08543418347835541,
|
303 |
+
"step": 200
|
304 |
+
},
|
305 |
+
{
|
306 |
+
"epoch": 0.22,
|
307 |
+
"learning_rate": 4.770016517582283e-06,
|
308 |
+
"logits/chosen": -0.19644469022750854,
|
309 |
+
"logits/rejected": -0.19420596957206726,
|
310 |
+
"logps/chosen": -596.8599853515625,
|
311 |
+
"logps/rejected": -661.6636962890625,
|
312 |
+
"loss": 0.1942,
|
313 |
+
"rewards/accuracies": 0.543749988079071,
|
314 |
+
"rewards/chosen": -0.04881254583597183,
|
315 |
+
"rewards/margins": 0.03198176622390747,
|
316 |
+
"rewards/rejected": -0.0807943120598793,
|
317 |
+
"step": 210
|
318 |
+
},
|
319 |
+
{
|
320 |
+
"epoch": 0.23,
|
321 |
+
"learning_rate": 4.7294165401363616e-06,
|
322 |
+
"logits/chosen": -0.2224278450012207,
|
323 |
+
"logits/rejected": -0.10729773342609406,
|
324 |
+
"logps/chosen": -604.28662109375,
|
325 |
+
"logps/rejected": -651.0653076171875,
|
326 |
+
"loss": 0.186,
|
327 |
+
"rewards/accuracies": 0.550000011920929,
|
328 |
+
"rewards/chosen": -0.04510096088051796,
|
329 |
+
"rewards/margins": 0.03683970868587494,
|
330 |
+
"rewards/rejected": -0.0819406658411026,
|
331 |
+
"step": 220
|
332 |
+
},
|
333 |
+
{
|
334 |
+
"epoch": 0.25,
|
335 |
+
"learning_rate": 4.68572067247573e-06,
|
336 |
+
"logits/chosen": -0.20905578136444092,
|
337 |
+
"logits/rejected": -0.2025284320116043,
|
338 |
+
"logps/chosen": -502.0655822753906,
|
339 |
+
"logps/rejected": -572.6324462890625,
|
340 |
+
"loss": 0.199,
|
341 |
+
"rewards/accuracies": 0.48750001192092896,
|
342 |
+
"rewards/chosen": -0.05115518718957901,
|
343 |
+
"rewards/margins": 0.020948706194758415,
|
344 |
+
"rewards/rejected": -0.07210389524698257,
|
345 |
+
"step": 230
|
346 |
+
},
|
347 |
+
{
|
348 |
+
"epoch": 0.26,
|
349 |
+
"learning_rate": 4.638989593081364e-06,
|
350 |
+
"logits/chosen": -0.1577962040901184,
|
351 |
+
"logits/rejected": -0.16388146579265594,
|
352 |
+
"logps/chosen": -501.21551513671875,
|
353 |
+
"logps/rejected": -542.8304443359375,
|
354 |
+
"loss": 0.1941,
|
355 |
+
"rewards/accuracies": 0.4312500059604645,
|
356 |
+
"rewards/chosen": -0.03727288171648979,
|
357 |
+
"rewards/margins": 0.03399407118558884,
|
358 |
+
"rewards/rejected": -0.07126694917678833,
|
359 |
+
"step": 240
|
360 |
+
},
|
361 |
+
{
|
362 |
+
"epoch": 0.27,
|
363 |
+
"learning_rate": 4.5892881952959015e-06,
|
364 |
+
"logits/chosen": -0.13874894380569458,
|
365 |
+
"logits/rejected": -0.10486537218093872,
|
366 |
+
"logps/chosen": -536.9546508789062,
|
367 |
+
"logps/rejected": -576.2243041992188,
|
368 |
+
"loss": 0.1903,
|
369 |
+
"rewards/accuracies": 0.5062500238418579,
|
370 |
+
"rewards/chosen": -0.03655445948243141,
|
371 |
+
"rewards/margins": 0.023150647059082985,
|
372 |
+
"rewards/rejected": -0.05970510095357895,
|
373 |
+
"step": 250
|
374 |
+
},
|
375 |
+
{
|
376 |
+
"epoch": 0.28,
|
377 |
+
"learning_rate": 4.536685497209182e-06,
|
378 |
+
"logits/chosen": -0.15160498023033142,
|
379 |
+
"logits/rejected": -0.20615887641906738,
|
380 |
+
"logps/chosen": -536.8988647460938,
|
381 |
+
"logps/rejected": -600.5662231445312,
|
382 |
+
"loss": 0.1915,
|
383 |
+
"rewards/accuracies": 0.512499988079071,
|
384 |
+
"rewards/chosen": -0.03945186734199524,
|
385 |
+
"rewards/margins": 0.040184784680604935,
|
386 |
+
"rewards/rejected": -0.07963664829730988,
|
387 |
+
"step": 260
|
388 |
+
},
|
389 |
+
{
|
390 |
+
"epoch": 0.29,
|
391 |
+
"learning_rate": 4.481254545815943e-06,
|
392 |
+
"logits/chosen": -0.16373318433761597,
|
393 |
+
"logits/rejected": -0.15171615779399872,
|
394 |
+
"logps/chosen": -630.7620239257812,
|
395 |
+
"logps/rejected": -654.4037475585938,
|
396 |
+
"loss": 0.2068,
|
397 |
+
"rewards/accuracies": 0.40625,
|
398 |
+
"rewards/chosen": -0.05563617870211601,
|
399 |
+
"rewards/margins": 0.019675999879837036,
|
400 |
+
"rewards/rejected": -0.07531218230724335,
|
401 |
+
"step": 270
|
402 |
+
},
|
403 |
+
{
|
404 |
+
"epoch": 0.3,
|
405 |
+
"learning_rate": 4.42307231557875e-06,
|
406 |
+
"logits/chosen": -0.06917186826467514,
|
407 |
+
"logits/rejected": -0.1992654800415039,
|
408 |
+
"logps/chosen": -583.5831909179688,
|
409 |
+
"logps/rejected": -621.246826171875,
|
410 |
+
"loss": 0.1911,
|
411 |
+
"rewards/accuracies": 0.4937500059604645,
|
412 |
+
"rewards/chosen": -0.03815799206495285,
|
413 |
+
"rewards/margins": 0.02716698870062828,
|
414 |
+
"rewards/rejected": -0.06532497704029083,
|
415 |
+
"step": 280
|
416 |
+
},
|
417 |
+
{
|
418 |
+
"epoch": 0.31,
|
419 |
+
"learning_rate": 4.3622196015370305e-06,
|
420 |
+
"logits/chosen": -0.09893601387739182,
|
421 |
+
"logits/rejected": -0.15837347507476807,
|
422 |
+
"logps/chosen": -517.561279296875,
|
423 |
+
"logps/rejected": -599.6585693359375,
|
424 |
+
"loss": 0.1935,
|
425 |
+
"rewards/accuracies": 0.5062500238418579,
|
426 |
+
"rewards/chosen": -0.03791649267077446,
|
427 |
+
"rewards/margins": 0.02659284509718418,
|
428 |
+
"rewards/rejected": -0.0645093321800232,
|
429 |
+
"step": 290
|
430 |
+
},
|
431 |
+
{
|
432 |
+
"epoch": 0.32,
|
433 |
+
"learning_rate": 4.298780907110648e-06,
|
434 |
+
"logits/chosen": -0.0635145753622055,
|
435 |
+
"logits/rejected": -0.10447721183300018,
|
436 |
+
"logps/chosen": -549.46875,
|
437 |
+
"logps/rejected": -607.9141235351562,
|
438 |
+
"loss": 0.1957,
|
439 |
+
"rewards/accuracies": 0.48124998807907104,
|
440 |
+
"rewards/chosen": -0.04532330110669136,
|
441 |
+
"rewards/margins": 0.0257760938256979,
|
442 |
+
"rewards/rejected": -0.07109940052032471,
|
443 |
+
"step": 300
|
444 |
+
},
|
445 |
+
{
|
446 |
+
"epoch": 0.33,
|
447 |
+
"learning_rate": 4.23284432675381e-06,
|
448 |
+
"logits/chosen": -0.1379682570695877,
|
449 |
+
"logits/rejected": -0.16269415616989136,
|
450 |
+
"logps/chosen": -533.80029296875,
|
451 |
+
"logps/rejected": -575.4193115234375,
|
452 |
+
"loss": 0.1931,
|
453 |
+
"rewards/accuracies": 0.512499988079071,
|
454 |
+
"rewards/chosen": -0.05671093985438347,
|
455 |
+
"rewards/margins": 0.0316205732524395,
|
456 |
+
"rewards/rejected": -0.08833150565624237,
|
457 |
+
"step": 310
|
458 |
+
},
|
459 |
+
{
|
460 |
+
"epoch": 0.34,
|
461 |
+
"learning_rate": 4.164501423622277e-06,
|
462 |
+
"logits/chosen": -0.19988815486431122,
|
463 |
+
"logits/rejected": -0.12660416960716248,
|
464 |
+
"logps/chosen": -556.7050170898438,
|
465 |
+
"logps/rejected": -533.2452392578125,
|
466 |
+
"loss": 0.2061,
|
467 |
+
"rewards/accuracies": 0.4000000059604645,
|
468 |
+
"rewards/chosen": -0.0512329563498497,
|
469 |
+
"rewards/margins": 0.006737883202731609,
|
470 |
+
"rewards/rejected": -0.057970840483903885,
|
471 |
+
"step": 320
|
472 |
+
},
|
473 |
+
{
|
474 |
+
"epoch": 0.35,
|
475 |
+
"learning_rate": 4.0938471024237355e-06,
|
476 |
+
"logits/chosen": -0.12714505195617676,
|
477 |
+
"logits/rejected": -0.0678124725818634,
|
478 |
+
"logps/chosen": -551.6757202148438,
|
479 |
+
"logps/rejected": -634.5169677734375,
|
480 |
+
"loss": 0.1845,
|
481 |
+
"rewards/accuracies": 0.4937500059604645,
|
482 |
+
"rewards/chosen": -0.04442024976015091,
|
483 |
+
"rewards/margins": 0.053262293338775635,
|
484 |
+
"rewards/rejected": -0.09768255054950714,
|
485 |
+
"step": 330
|
486 |
+
},
|
487 |
+
{
|
488 |
+
"epoch": 0.36,
|
489 |
+
"learning_rate": 4.020979477627907e-06,
|
490 |
+
"logits/chosen": -0.08381222188472748,
|
491 |
+
"logits/rejected": -0.1388164460659027,
|
492 |
+
"logps/chosen": -567.6676635742188,
|
493 |
+
"logps/rejected": -620.1175537109375,
|
494 |
+
"loss": 0.1928,
|
495 |
+
"rewards/accuracies": 0.5375000238418579,
|
496 |
+
"rewards/chosen": -0.04053609445691109,
|
497 |
+
"rewards/margins": 0.04673929885029793,
|
498 |
+
"rewards/rejected": -0.08727538585662842,
|
499 |
+
"step": 340
|
500 |
+
},
|
501 |
+
{
|
502 |
+
"epoch": 0.37,
|
503 |
+
"learning_rate": 3.9459997372194105e-06,
|
504 |
+
"logits/chosen": -0.09087273478507996,
|
505 |
+
"logits/rejected": -0.052247822284698486,
|
506 |
+
"logps/chosen": -529.18408203125,
|
507 |
+
"logps/rejected": -596.254638671875,
|
508 |
+
"loss": 0.1969,
|
509 |
+
"rewards/accuracies": 0.48124998807907104,
|
510 |
+
"rewards/chosen": -0.05547628924250603,
|
511 |
+
"rewards/margins": 0.04215295612812042,
|
512 |
+
"rewards/rejected": -0.09762924164533615,
|
513 |
+
"step": 350
|
514 |
+
},
|
515 |
+
{
|
516 |
+
"epoch": 0.38,
|
517 |
+
"learning_rate": 3.869012002182573e-06,
|
518 |
+
"logits/chosen": -0.2067844122648239,
|
519 |
+
"logits/rejected": -0.17433646321296692,
|
520 |
+
"logps/chosen": -554.9356689453125,
|
521 |
+
"logps/rejected": -630.5103759765625,
|
522 |
+
"loss": 0.1922,
|
523 |
+
"rewards/accuracies": 0.53125,
|
524 |
+
"rewards/chosen": -0.05108080059289932,
|
525 |
+
"rewards/margins": 0.03945896774530411,
|
526 |
+
"rewards/rejected": -0.09053976833820343,
|
527 |
+
"step": 360
|
528 |
+
},
|
529 |
+
{
|
530 |
+
"epoch": 0.39,
|
531 |
+
"learning_rate": 3.7901231819133104e-06,
|
532 |
+
"logits/chosen": -0.11504027992486954,
|
533 |
+
"logits/rejected": -0.04383224993944168,
|
534 |
+
"logps/chosen": -577.4075317382812,
|
535 |
+
"logps/rejected": -607.4254150390625,
|
536 |
+
"loss": 0.1934,
|
537 |
+
"rewards/accuracies": 0.4625000059604645,
|
538 |
+
"rewards/chosen": -0.044865719974040985,
|
539 |
+
"rewards/margins": 0.025182927027344704,
|
540 |
+
"rewards/rejected": -0.07004864513874054,
|
541 |
+
"step": 370
|
542 |
+
},
|
543 |
+
{
|
544 |
+
"epoch": 0.41,
|
545 |
+
"learning_rate": 3.709442825758875e-06,
|
546 |
+
"logits/chosen": -0.16869018971920013,
|
547 |
+
"logits/rejected": -0.1371726095676422,
|
548 |
+
"logps/chosen": -584.6583251953125,
|
549 |
+
"logps/rejected": -648.9109497070312,
|
550 |
+
"loss": 0.2022,
|
551 |
+
"rewards/accuracies": 0.518750011920929,
|
552 |
+
"rewards/chosen": -0.06359410285949707,
|
553 |
+
"rewards/margins": 0.041747674345970154,
|
554 |
+
"rewards/rejected": -0.10534176975488663,
|
555 |
+
"step": 380
|
556 |
+
},
|
557 |
+
{
|
558 |
+
"epoch": 0.42,
|
559 |
+
"learning_rate": 3.6270829708916113e-06,
|
560 |
+
"logits/chosen": -0.21822664141654968,
|
561 |
+
"logits/rejected": -0.15734872221946716,
|
562 |
+
"logps/chosen": -559.946533203125,
|
563 |
+
"logps/rejected": -605.324951171875,
|
564 |
+
"loss": 0.1974,
|
565 |
+
"rewards/accuracies": 0.5,
|
566 |
+
"rewards/chosen": -0.051652587950229645,
|
567 |
+
"rewards/margins": 0.02688014507293701,
|
568 |
+
"rewards/rejected": -0.07853273302316666,
|
569 |
+
"step": 390
|
570 |
+
},
|
571 |
+
{
|
572 |
+
"epoch": 0.43,
|
573 |
+
"learning_rate": 3.543157986727991e-06,
|
574 |
+
"logits/chosen": -0.0996035560965538,
|
575 |
+
"logits/rejected": -0.1632828712463379,
|
576 |
+
"logps/chosen": -560.0746459960938,
|
577 |
+
"logps/rejected": -612.9156494140625,
|
578 |
+
"loss": 0.201,
|
579 |
+
"rewards/accuracies": 0.4937500059604645,
|
580 |
+
"rewards/chosen": -0.051237620413303375,
|
581 |
+
"rewards/margins": 0.04104907065629959,
|
582 |
+
"rewards/rejected": -0.09228668361902237,
|
583 |
+
"step": 400
|
584 |
+
},
|
585 |
+
{
|
586 |
+
"epoch": 0.44,
|
587 |
+
"learning_rate": 3.4577844161089614e-06,
|
588 |
+
"logits/chosen": -0.01575249433517456,
|
589 |
+
"logits/rejected": -0.1881057471036911,
|
590 |
+
"logps/chosen": -559.406494140625,
|
591 |
+
"logps/rejected": -582.1732177734375,
|
592 |
+
"loss": 0.2061,
|
593 |
+
"rewards/accuracies": 0.41874998807907104,
|
594 |
+
"rewards/chosen": -0.03435258939862251,
|
595 |
+
"rewards/margins": 0.023811260238289833,
|
596 |
+
"rewards/rejected": -0.05816385895013809,
|
597 |
+
"step": 410
|
598 |
+
},
|
599 |
+
{
|
600 |
+
"epoch": 0.45,
|
601 |
+
"learning_rate": 3.3710808134621577e-06,
|
602 |
+
"logits/chosen": -0.1044248715043068,
|
603 |
+
"logits/rejected": -0.12559100985527039,
|
604 |
+
"logps/chosen": -518.0560913085938,
|
605 |
+
"logps/rejected": -559.5712890625,
|
606 |
+
"loss": 0.2056,
|
607 |
+
"rewards/accuracies": 0.48124998807907104,
|
608 |
+
"rewards/chosen": -0.029239226132631302,
|
609 |
+
"rewards/margins": 0.02849581278860569,
|
610 |
+
"rewards/rejected": -0.05773504450917244,
|
611 |
+
"step": 420
|
612 |
+
},
|
613 |
+
{
|
614 |
+
"epoch": 0.46,
|
615 |
+
"learning_rate": 3.2831675801707126e-06,
|
616 |
+
"logits/chosen": -0.14103475213050842,
|
617 |
+
"logits/rejected": -0.08099827915430069,
|
618 |
+
"logps/chosen": -546.7091064453125,
|
619 |
+
"logps/rejected": -608.6934814453125,
|
620 |
+
"loss": 0.1944,
|
621 |
+
"rewards/accuracies": 0.518750011920929,
|
622 |
+
"rewards/chosen": -0.028408538550138474,
|
623 |
+
"rewards/margins": 0.034466955810785294,
|
624 |
+
"rewards/rejected": -0.06287549436092377,
|
625 |
+
"step": 430
|
626 |
+
},
|
627 |
+
{
|
628 |
+
"epoch": 0.47,
|
629 |
+
"learning_rate": 3.194166797377289e-06,
|
630 |
+
"logits/chosen": -0.1468835175037384,
|
631 |
+
"logits/rejected": -0.10806012153625488,
|
632 |
+
"logps/chosen": -539.1943969726562,
|
633 |
+
"logps/rejected": -607.8096923828125,
|
634 |
+
"loss": 0.1959,
|
635 |
+
"rewards/accuracies": 0.53125,
|
636 |
+
"rewards/chosen": -0.016944291070103645,
|
637 |
+
"rewards/margins": 0.03214774280786514,
|
638 |
+
"rewards/rejected": -0.04909203201532364,
|
639 |
+
"step": 440
|
640 |
+
},
|
641 |
+
{
|
642 |
+
"epoch": 0.48,
|
643 |
+
"learning_rate": 3.104202056455501e-06,
|
644 |
+
"logits/chosen": -0.07705841958522797,
|
645 |
+
"logits/rejected": -0.13577719032764435,
|
646 |
+
"logps/chosen": -534.7425537109375,
|
647 |
+
"logps/rejected": -590.2757568359375,
|
648 |
+
"loss": 0.1886,
|
649 |
+
"rewards/accuracies": 0.543749988079071,
|
650 |
+
"rewards/chosen": -0.027219612151384354,
|
651 |
+
"rewards/margins": 0.03623734414577484,
|
652 |
+
"rewards/rejected": -0.0634569600224495,
|
653 |
+
"step": 450
|
654 |
+
},
|
655 |
+
{
|
656 |
+
"epoch": 0.49,
|
657 |
+
"learning_rate": 3.013398287384144e-06,
|
658 |
+
"logits/chosen": -0.05825149267911911,
|
659 |
+
"logits/rejected": -0.0818571001291275,
|
660 |
+
"logps/chosen": -504.8671875,
|
661 |
+
"logps/rejected": -530.8888549804688,
|
662 |
+
"loss": 0.1971,
|
663 |
+
"rewards/accuracies": 0.45625001192092896,
|
664 |
+
"rewards/chosen": -0.02120174467563629,
|
665 |
+
"rewards/margins": 0.018600985407829285,
|
666 |
+
"rewards/rejected": -0.039802733808755875,
|
667 |
+
"step": 460
|
668 |
+
},
|
669 |
+
{
|
670 |
+
"epoch": 0.5,
|
671 |
+
"learning_rate": 2.9218815852625717e-06,
|
672 |
+
"logits/chosen": -0.1541839838027954,
|
673 |
+
"logits/rejected": -0.06153281405568123,
|
674 |
+
"logps/chosen": -549.5711669921875,
|
675 |
+
"logps/rejected": -612.5526733398438,
|
676 |
+
"loss": 0.1952,
|
677 |
+
"rewards/accuracies": 0.5249999761581421,
|
678 |
+
"rewards/chosen": -0.017424913123250008,
|
679 |
+
"rewards/margins": 0.02850305661559105,
|
680 |
+
"rewards/rejected": -0.04592796787619591,
|
681 |
+
"step": 470
|
682 |
+
},
|
683 |
+
{
|
684 |
+
"epoch": 0.51,
|
685 |
+
"learning_rate": 2.829779035208113e-06,
|
686 |
+
"logits/chosen": -0.03961920738220215,
|
687 |
+
"logits/rejected": -0.1473100781440735,
|
688 |
+
"logps/chosen": -557.7020874023438,
|
689 |
+
"logps/rejected": -588.0806884765625,
|
690 |
+
"loss": 0.1877,
|
691 |
+
"rewards/accuracies": 0.4937500059604645,
|
692 |
+
"rewards/chosen": -0.027860945090651512,
|
693 |
+
"rewards/margins": 0.024425392970442772,
|
694 |
+
"rewards/rejected": -0.052286338061094284,
|
695 |
+
"step": 480
|
696 |
+
},
|
697 |
+
{
|
698 |
+
"epoch": 0.52,
|
699 |
+
"learning_rate": 2.737218535878705e-06,
|
700 |
+
"logits/chosen": -0.029465090483427048,
|
701 |
+
"logits/rejected": -0.08229364454746246,
|
702 |
+
"logps/chosen": -511.9815368652344,
|
703 |
+
"logps/rejected": -621.8477172851562,
|
704 |
+
"loss": 0.1862,
|
705 |
+
"rewards/accuracies": 0.5562499761581421,
|
706 |
+
"rewards/chosen": -0.014062300324440002,
|
707 |
+
"rewards/margins": 0.042990971356630325,
|
708 |
+
"rewards/rejected": -0.05705326795578003,
|
709 |
+
"step": 490
|
710 |
+
},
|
711 |
+
{
|
712 |
+
"epoch": 0.53,
|
713 |
+
"learning_rate": 2.64432862186579e-06,
|
714 |
+
"logits/chosen": -0.11332341283559799,
|
715 |
+
"logits/rejected": -0.022313905879855156,
|
716 |
+
"logps/chosen": -538.530517578125,
|
717 |
+
"logps/rejected": -565.3148193359375,
|
718 |
+
"loss": 0.1903,
|
719 |
+
"rewards/accuracies": 0.574999988079071,
|
720 |
+
"rewards/chosen": -0.0199245847761631,
|
721 |
+
"rewards/margins": 0.040730707347393036,
|
722 |
+
"rewards/rejected": -0.06065528467297554,
|
723 |
+
"step": 500
|
724 |
+
},
|
725 |
+
{
|
726 |
+
"epoch": 0.54,
|
727 |
+
"learning_rate": 2.551238285204126e-06,
|
728 |
+
"logits/chosen": -0.17237767577171326,
|
729 |
+
"logits/rejected": 0.029280701652169228,
|
730 |
+
"logps/chosen": -522.3345947265625,
|
731 |
+
"logps/rejected": -557.0701293945312,
|
732 |
+
"loss": 0.185,
|
733 |
+
"rewards/accuracies": 0.518750011920929,
|
734 |
+
"rewards/chosen": -0.02585068717598915,
|
735 |
+
"rewards/margins": 0.03859715908765793,
|
736 |
+
"rewards/rejected": -0.06444784253835678,
|
737 |
+
"step": 510
|
738 |
+
},
|
739 |
+
{
|
740 |
+
"epoch": 0.55,
|
741 |
+
"learning_rate": 2.4580767962463688e-06,
|
742 |
+
"logits/chosen": -0.1446155458688736,
|
743 |
+
"logits/rejected": -0.15588733553886414,
|
744 |
+
"logps/chosen": -495.2469177246094,
|
745 |
+
"logps/rejected": -525.423095703125,
|
746 |
+
"loss": 0.2006,
|
747 |
+
"rewards/accuracies": 0.4937500059604645,
|
748 |
+
"rewards/chosen": -0.03724539279937744,
|
749 |
+
"rewards/margins": 0.0233670175075531,
|
750 |
+
"rewards/rejected": -0.060612402856349945,
|
751 |
+
"step": 520
|
752 |
+
},
|
753 |
+
{
|
754 |
+
"epoch": 0.57,
|
755 |
+
"learning_rate": 2.3649735241511546e-06,
|
756 |
+
"logits/chosen": -0.10832977294921875,
|
757 |
+
"logits/rejected": -0.17813685536384583,
|
758 |
+
"logps/chosen": -487.40252685546875,
|
759 |
+
"logps/rejected": -573.7322998046875,
|
760 |
+
"loss": 0.1847,
|
761 |
+
"rewards/accuracies": 0.5249999761581421,
|
762 |
+
"rewards/chosen": -0.022226981818675995,
|
763 |
+
"rewards/margins": 0.03627447038888931,
|
764 |
+
"rewards/rejected": -0.05850145220756531,
|
765 |
+
"step": 530
|
766 |
+
},
|
767 |
+
{
|
768 |
+
"epoch": 0.58,
|
769 |
+
"learning_rate": 2.2720577572339914e-06,
|
770 |
+
"logits/chosen": -0.12618735432624817,
|
771 |
+
"logits/rejected": -0.17089055478572845,
|
772 |
+
"logps/chosen": -491.43927001953125,
|
773 |
+
"logps/rejected": -632.1901245117188,
|
774 |
+
"loss": 0.1815,
|
775 |
+
"rewards/accuracies": 0.543749988079071,
|
776 |
+
"rewards/chosen": -0.0382881760597229,
|
777 |
+
"rewards/margins": 0.06080668047070503,
|
778 |
+
"rewards/rejected": -0.09909485280513763,
|
779 |
+
"step": 540
|
780 |
+
},
|
781 |
+
{
|
782 |
+
"epoch": 0.59,
|
783 |
+
"learning_rate": 2.1794585234303995e-06,
|
784 |
+
"logits/chosen": -0.16708405315876007,
|
785 |
+
"logits/rejected": -0.11356528848409653,
|
786 |
+
"logps/chosen": -579.8703002929688,
|
787 |
+
"logps/rejected": -630.3308715820312,
|
788 |
+
"loss": 0.2008,
|
789 |
+
"rewards/accuracies": 0.5249999761581421,
|
790 |
+
"rewards/chosen": -0.04018039256334305,
|
791 |
+
"rewards/margins": 0.03480806201696396,
|
792 |
+
"rewards/rejected": -0.074988454580307,
|
793 |
+
"step": 550
|
794 |
+
},
|
795 |
+
{
|
796 |
+
"epoch": 0.6,
|
797 |
+
"learning_rate": 2.0873044111206407e-06,
|
798 |
+
"logits/chosen": -0.030877122655510902,
|
799 |
+
"logits/rejected": -0.18640507757663727,
|
800 |
+
"logps/chosen": -507.278564453125,
|
801 |
+
"logps/rejected": -627.2957153320312,
|
802 |
+
"loss": 0.1897,
|
803 |
+
"rewards/accuracies": 0.4625000059604645,
|
804 |
+
"rewards/chosen": -0.040919769555330276,
|
805 |
+
"rewards/margins": 0.057074230164289474,
|
806 |
+
"rewards/rejected": -0.09799399971961975,
|
807 |
+
"step": 560
|
808 |
+
},
|
809 |
+
{
|
810 |
+
"epoch": 0.61,
|
811 |
+
"learning_rate": 1.9957233905648293e-06,
|
812 |
+
"logits/chosen": -0.17021183669567108,
|
813 |
+
"logits/rejected": -0.23673537373542786,
|
814 |
+
"logps/chosen": -544.7388916015625,
|
815 |
+
"logps/rejected": -587.5650024414062,
|
816 |
+
"loss": 0.1941,
|
817 |
+
"rewards/accuracies": 0.46875,
|
818 |
+
"rewards/chosen": -0.04041619971394539,
|
819 |
+
"rewards/margins": 0.024558167904615402,
|
820 |
+
"rewards/rejected": -0.06497436761856079,
|
821 |
+
"step": 570
|
822 |
+
},
|
823 |
+
{
|
824 |
+
"epoch": 0.62,
|
825 |
+
"learning_rate": 1.904842636196402e-06,
|
826 |
+
"logits/chosen": -0.13388361036777496,
|
827 |
+
"logits/rejected": -0.0614379458129406,
|
828 |
+
"logps/chosen": -553.6912841796875,
|
829 |
+
"logps/rejected": -629.6470336914062,
|
830 |
+
"loss": 0.1903,
|
831 |
+
"rewards/accuracies": 0.543749988079071,
|
832 |
+
"rewards/chosen": -0.025933533906936646,
|
833 |
+
"rewards/margins": 0.0378827229142189,
|
834 |
+
"rewards/rejected": -0.06381625682115555,
|
835 |
+
"step": 580
|
836 |
+
},
|
837 |
+
{
|
838 |
+
"epoch": 0.63,
|
839 |
+
"learning_rate": 1.814788350020726e-06,
|
840 |
+
"logits/chosen": -0.06043483689427376,
|
841 |
+
"logits/rejected": -0.10723511874675751,
|
842 |
+
"logps/chosen": -521.6217041015625,
|
843 |
+
"logps/rejected": -643.6219482421875,
|
844 |
+
"loss": 0.197,
|
845 |
+
"rewards/accuracies": 0.5625,
|
846 |
+
"rewards/chosen": -0.024261657148599625,
|
847 |
+
"rewards/margins": 0.03998683765530586,
|
848 |
+
"rewards/rejected": -0.06424849480390549,
|
849 |
+
"step": 590
|
850 |
+
},
|
851 |
+
{
|
852 |
+
"epoch": 0.64,
|
853 |
+
"learning_rate": 1.725685586364051e-06,
|
854 |
+
"logits/chosen": -0.1038782149553299,
|
855 |
+
"logits/rejected": -0.047629911452531815,
|
856 |
+
"logps/chosen": -535.3336181640625,
|
857 |
+
"logps/rejected": -564.0821533203125,
|
858 |
+
"loss": 0.1948,
|
859 |
+
"rewards/accuracies": 0.5249999761581421,
|
860 |
+
"rewards/chosen": -0.024827763438224792,
|
861 |
+
"rewards/margins": 0.024268008768558502,
|
862 |
+
"rewards/rejected": -0.049095772206783295,
|
863 |
+
"step": 600
|
864 |
+
},
|
865 |
+
{
|
866 |
+
"epoch": 0.65,
|
867 |
+
"learning_rate": 1.6376580782162172e-06,
|
868 |
+
"logits/chosen": -0.12502829730510712,
|
869 |
+
"logits/rejected": -0.13951411843299866,
|
870 |
+
"logps/chosen": -565.2417602539062,
|
871 |
+
"logps/rejected": -653.1436157226562,
|
872 |
+
"loss": 0.1921,
|
873 |
+
"rewards/accuracies": 0.4937500059604645,
|
874 |
+
"rewards/chosen": -0.027302825823426247,
|
875 |
+
"rewards/margins": 0.03322170302271843,
|
876 |
+
"rewards/rejected": -0.060524530708789825,
|
877 |
+
"step": 610
|
878 |
+
},
|
879 |
+
{
|
880 |
+
"epoch": 0.66,
|
881 |
+
"learning_rate": 1.550828065408227e-06,
|
882 |
+
"logits/chosen": -0.15310420095920563,
|
883 |
+
"logits/rejected": -0.035526055842638016,
|
884 |
+
"logps/chosen": -510.84759521484375,
|
885 |
+
"logps/rejected": -573.4019775390625,
|
886 |
+
"loss": 0.1942,
|
887 |
+
"rewards/accuracies": 0.550000011920929,
|
888 |
+
"rewards/chosen": -0.02462349459528923,
|
889 |
+
"rewards/margins": 0.02571946755051613,
|
890 |
+
"rewards/rejected": -0.05034296587109566,
|
891 |
+
"step": 620
|
892 |
+
},
|
893 |
+
{
|
894 |
+
"epoch": 0.67,
|
895 |
+
"learning_rate": 1.4653161248633053e-06,
|
896 |
+
"logits/chosen": -0.12701451778411865,
|
897 |
+
"logits/rejected": -0.03017173707485199,
|
898 |
+
"logps/chosen": -507.4302673339844,
|
899 |
+
"logps/rejected": -554.05029296875,
|
900 |
+
"loss": 0.1878,
|
901 |
+
"rewards/accuracies": 0.48124998807907104,
|
902 |
+
"rewards/chosen": -0.028983917087316513,
|
903 |
+
"rewards/margins": 0.03194578364491463,
|
904 |
+
"rewards/rejected": -0.06092969700694084,
|
905 |
+
"step": 630
|
906 |
+
},
|
907 |
+
{
|
908 |
+
"epoch": 0.68,
|
909 |
+
"learning_rate": 1.381241003157162e-06,
|
910 |
+
"logits/chosen": -0.1072702407836914,
|
911 |
+
"logits/rejected": -0.12664569914340973,
|
912 |
+
"logps/chosen": -511.8506774902344,
|
913 |
+
"logps/rejected": -582.0699462890625,
|
914 |
+
"loss": 0.199,
|
915 |
+
"rewards/accuracies": 0.512499988079071,
|
916 |
+
"rewards/chosen": -0.034758515655994415,
|
917 |
+
"rewards/margins": 0.027269739657640457,
|
918 |
+
"rewards/rejected": -0.062028251588344574,
|
919 |
+
"step": 640
|
920 |
+
},
|
921 |
+
{
|
922 |
+
"epoch": 0.69,
|
923 |
+
"learning_rate": 1.298719451619979e-06,
|
924 |
+
"logits/chosen": -0.05098239332437515,
|
925 |
+
"logits/rejected": -0.1476130485534668,
|
926 |
+
"logps/chosen": -521.5150146484375,
|
927 |
+
"logps/rejected": -601.8914184570312,
|
928 |
+
"loss": 0.1877,
|
929 |
+
"rewards/accuracies": 0.512499988079071,
|
930 |
+
"rewards/chosen": -0.020487060770392418,
|
931 |
+
"rewards/margins": 0.033522095531225204,
|
932 |
+
"rewards/rejected": -0.054009150713682175,
|
933 |
+
"step": 650
|
934 |
+
},
|
935 |
+
{
|
936 |
+
"epoch": 0.7,
|
937 |
+
"learning_rate": 1.2178660642091036e-06,
|
938 |
+
"logits/chosen": -0.10045331716537476,
|
939 |
+
"logits/rejected": -0.10961699485778809,
|
940 |
+
"logps/chosen": -547.9536743164062,
|
941 |
+
"logps/rejected": -613.6334228515625,
|
942 |
+
"loss": 0.1825,
|
943 |
+
"rewards/accuracies": 0.59375,
|
944 |
+
"rewards/chosen": -0.02681289240717888,
|
945 |
+
"rewards/margins": 0.05183552950620651,
|
946 |
+
"rewards/rejected": -0.07864842563867569,
|
947 |
+
"step": 660
|
948 |
+
},
|
949 |
+
{
|
950 |
+
"epoch": 0.71,
|
951 |
+
"learning_rate": 1.1387931183775821e-06,
|
952 |
+
"logits/chosen": -0.08650709688663483,
|
953 |
+
"logits/rejected": -0.17840668559074402,
|
954 |
+
"logps/chosen": -531.9118041992188,
|
955 |
+
"logps/rejected": -626.980224609375,
|
956 |
+
"loss": 0.1847,
|
957 |
+
"rewards/accuracies": 0.574999988079071,
|
958 |
+
"rewards/chosen": -0.019130777567625046,
|
959 |
+
"rewards/margins": 0.05559687688946724,
|
960 |
+
"rewards/rejected": -0.07472765445709229,
|
961 |
+
"step": 670
|
962 |
+
},
|
963 |
+
{
|
964 |
+
"epoch": 0.73,
|
965 |
+
"learning_rate": 1.061610419159532e-06,
|
966 |
+
"logits/chosen": -0.04611722379922867,
|
967 |
+
"logits/rejected": -0.09121225029230118,
|
968 |
+
"logps/chosen": -550.1459350585938,
|
969 |
+
"logps/rejected": -607.9168090820312,
|
970 |
+
"loss": 0.1874,
|
971 |
+
"rewards/accuracies": 0.550000011920929,
|
972 |
+
"rewards/chosen": -0.029284963384270668,
|
973 |
+
"rewards/margins": 0.05334455519914627,
|
974 |
+
"rewards/rejected": -0.08262950927019119,
|
975 |
+
"step": 680
|
976 |
+
},
|
977 |
+
{
|
978 |
+
"epoch": 0.74,
|
979 |
+
"learning_rate": 9.864251466888364e-07,
|
980 |
+
"logits/chosen": -0.15967205166816711,
|
981 |
+
"logits/rejected": -0.06603795289993286,
|
982 |
+
"logps/chosen": -557.4961547851562,
|
983 |
+
"logps/rejected": -607.64306640625,
|
984 |
+
"loss": 0.1812,
|
985 |
+
"rewards/accuracies": 0.518750011920929,
|
986 |
+
"rewards/chosen": -0.034804992377758026,
|
987 |
+
"rewards/margins": 0.03645947948098183,
|
988 |
+
"rewards/rejected": -0.07126447558403015,
|
989 |
+
"step": 690
|
990 |
+
},
|
991 |
+
{
|
992 |
+
"epoch": 0.75,
|
993 |
+
"learning_rate": 9.133417073629288e-07,
|
994 |
+
"logits/chosen": -0.15150725841522217,
|
995 |
+
"logits/rejected": -0.11879472434520721,
|
996 |
+
"logps/chosen": -504.88665771484375,
|
997 |
+
"logps/rejected": -582.1925659179688,
|
998 |
+
"loss": 0.192,
|
999 |
+
"rewards/accuracies": 0.5,
|
1000 |
+
"rewards/chosen": -0.03086964413523674,
|
1001 |
+
"rewards/margins": 0.04862387850880623,
|
1002 |
+
"rewards/rejected": -0.07949351519346237,
|
1003 |
+
"step": 700
|
1004 |
+
},
|
1005 |
+
{
|
1006 |
+
"epoch": 0.76,
|
1007 |
+
"learning_rate": 8.424615888583332e-07,
|
1008 |
+
"logits/chosen": -0.13878367841243744,
|
1009 |
+
"logits/rejected": -0.08197399228811264,
|
1010 |
+
"logps/chosen": -521.1176147460938,
|
1011 |
+
"logps/rejected": -626.109375,
|
1012 |
+
"loss": 0.1877,
|
1013 |
+
"rewards/accuracies": 0.574999988079071,
|
1014 |
+
"rewards/chosen": -0.028367381542921066,
|
1015 |
+
"rewards/margins": 0.05848144367337227,
|
1016 |
+
"rewards/rejected": -0.08684882521629333,
|
1017 |
+
"step": 710
|
1018 |
+
},
|
1019 |
+
{
|
1020 |
+
"epoch": 0.77,
|
1021 |
+
"learning_rate": 7.738832191993092e-07,
|
1022 |
+
"logits/chosen": -0.11670303344726562,
|
1023 |
+
"logits/rejected": -0.1110248789191246,
|
1024 |
+
"logps/chosen": -546.02685546875,
|
1025 |
+
"logps/rejected": -633.3377075195312,
|
1026 |
+
"loss": 0.183,
|
1027 |
+
"rewards/accuracies": 0.518750011920929,
|
1028 |
+
"rewards/chosen": -0.030107159167528152,
|
1029 |
+
"rewards/margins": 0.04191134124994278,
|
1030 |
+
"rewards/rejected": -0.07201849669218063,
|
1031 |
+
"step": 720
|
1032 |
+
},
|
1033 |
+
{
|
1034 |
+
"epoch": 0.78,
|
1035 |
+
"learning_rate": 7.077018300752917e-07,
|
1036 |
+
"logits/chosen": -0.20101483166217804,
|
1037 |
+
"logits/rejected": -0.09887303411960602,
|
1038 |
+
"logps/chosen": -515.279296875,
|
1039 |
+
"logps/rejected": -566.751220703125,
|
1040 |
+
"loss": 0.1959,
|
1041 |
+
"rewards/accuracies": 0.48750001192092896,
|
1042 |
+
"rewards/chosen": -0.02799845300614834,
|
1043 |
+
"rewards/margins": 0.039141736924648285,
|
1044 |
+
"rewards/rejected": -0.06714018434286118,
|
1045 |
+
"step": 730
|
1046 |
+
},
|
1047 |
+
{
|
1048 |
+
"epoch": 0.79,
|
1049 |
+
"learning_rate": 6.440093245969342e-07,
|
1050 |
+
"logits/chosen": -0.21685993671417236,
|
1051 |
+
"logits/rejected": -0.014155983924865723,
|
1052 |
+
"logps/chosen": -536.5148315429688,
|
1053 |
+
"logps/rejected": -555.0255126953125,
|
1054 |
+
"loss": 0.1909,
|
1055 |
+
"rewards/accuracies": 0.4937500059604645,
|
1056 |
+
"rewards/chosen": -0.042235199362039566,
|
1057 |
+
"rewards/margins": 0.033806830644607544,
|
1058 |
+
"rewards/rejected": -0.07604203373193741,
|
1059 |
+
"step": 740
|
1060 |
+
},
|
1061 |
+
{
|
1062 |
+
"epoch": 0.8,
|
1063 |
+
"learning_rate": 5.828941496744075e-07,
|
1064 |
+
"logits/chosen": -0.1709945797920227,
|
1065 |
+
"logits/rejected": -0.09519165009260178,
|
1066 |
+
"logps/chosen": -487.0716247558594,
|
1067 |
+
"logps/rejected": -570.6549072265625,
|
1068 |
+
"loss": 0.1891,
|
1069 |
+
"rewards/accuracies": 0.550000011920929,
|
1070 |
+
"rewards/chosen": -0.03168477118015289,
|
1071 |
+
"rewards/margins": 0.05396095663309097,
|
1072 |
+
"rewards/rejected": -0.08564572036266327,
|
1073 |
+
"step": 750
|
1074 |
+
},
|
1075 |
+
{
|
1076 |
+
"epoch": 0.81,
|
1077 |
+
"learning_rate": 5.244411731951671e-07,
|
1078 |
+
"logits/chosen": -0.17727656662464142,
|
1079 |
+
"logits/rejected": -0.11151037365198135,
|
1080 |
+
"logps/chosen": -520.3614501953125,
|
1081 |
+
"logps/rejected": -573.2659301757812,
|
1082 |
+
"loss": 0.1921,
|
1083 |
+
"rewards/accuracies": 0.4937500059604645,
|
1084 |
+
"rewards/chosen": -0.024441594257950783,
|
1085 |
+
"rewards/margins": 0.03574614226818085,
|
1086 |
+
"rewards/rejected": -0.06018773466348648,
|
1087 |
+
"step": 760
|
1088 |
+
},
|
1089 |
+
{
|
1090 |
+
"epoch": 0.82,
|
1091 |
+
"learning_rate": 4.6873156617173594e-07,
|
1092 |
+
"logits/chosen": -0.16788466274738312,
|
1093 |
+
"logits/rejected": -0.04893454164266586,
|
1094 |
+
"logps/chosen": -508.6934509277344,
|
1095 |
+
"logps/rejected": -575.983154296875,
|
1096 |
+
"loss": 0.1867,
|
1097 |
+
"rewards/accuracies": 0.4437499940395355,
|
1098 |
+
"rewards/chosen": -0.021922405809164047,
|
1099 |
+
"rewards/margins": 0.034182045608758926,
|
1100 |
+
"rewards/rejected": -0.05610444396734238,
|
1101 |
+
"step": 770
|
1102 |
+
},
|
1103 |
+
{
|
1104 |
+
"epoch": 0.83,
|
1105 |
+
"learning_rate": 4.1584269002318653e-07,
|
1106 |
+
"logits/chosen": -0.07019881904125214,
|
1107 |
+
"logits/rejected": -0.18743988871574402,
|
1108 |
+
"logps/chosen": -485.97833251953125,
|
1109 |
+
"logps/rejected": -594.727783203125,
|
1110 |
+
"loss": 0.1866,
|
1111 |
+
"rewards/accuracies": 0.518750011920929,
|
1112 |
+
"rewards/chosen": -0.03881939500570297,
|
1113 |
+
"rewards/margins": 0.05370662733912468,
|
1114 |
+
"rewards/rejected": -0.09252602607011795,
|
1115 |
+
"step": 780
|
1116 |
+
},
|
1117 |
+
{
|
1118 |
+
"epoch": 0.84,
|
1119 |
+
"learning_rate": 3.658479891468258e-07,
|
1120 |
+
"logits/chosen": -0.17362000048160553,
|
1121 |
+
"logits/rejected": -0.11806299537420273,
|
1122 |
+
"logps/chosen": -479.9105529785156,
|
1123 |
+
"logps/rejected": -610.3316040039062,
|
1124 |
+
"loss": 0.1924,
|
1125 |
+
"rewards/accuracies": 0.518750011920929,
|
1126 |
+
"rewards/chosen": -0.023625101894140244,
|
1127 |
+
"rewards/margins": 0.07324331998825073,
|
1128 |
+
"rewards/rejected": -0.09686841815710068,
|
1129 |
+
"step": 790
|
1130 |
+
},
|
1131 |
+
{
|
1132 |
+
"epoch": 0.85,
|
1133 |
+
"learning_rate": 3.18816888929272e-07,
|
1134 |
+
"logits/chosen": -0.13044703006744385,
|
1135 |
+
"logits/rejected": -0.13462410867214203,
|
1136 |
+
"logps/chosen": -511.01275634765625,
|
1137 |
+
"logps/rejected": -576.4395141601562,
|
1138 |
+
"loss": 0.1845,
|
1139 |
+
"rewards/accuracies": 0.4625000059604645,
|
1140 |
+
"rewards/chosen": -0.024302612990140915,
|
1141 |
+
"rewards/margins": 0.04658300057053566,
|
1142 |
+
"rewards/rejected": -0.07088561356067657,
|
1143 |
+
"step": 800
|
1144 |
+
},
|
1145 |
+
{
|
1146 |
+
"epoch": 0.86,
|
1147 |
+
"learning_rate": 2.748146993385484e-07,
|
1148 |
+
"logits/chosen": -0.09310281276702881,
|
1149 |
+
"logits/rejected": -0.14107745885849,
|
1150 |
+
"logps/chosen": -542.4686279296875,
|
1151 |
+
"logps/rejected": -590.4827270507812,
|
1152 |
+
"loss": 0.1911,
|
1153 |
+
"rewards/accuracies": 0.4749999940395355,
|
1154 |
+
"rewards/chosen": -0.03655749931931496,
|
1155 |
+
"rewards/margins": 0.04121191054582596,
|
1156 |
+
"rewards/rejected": -0.07776941359043121,
|
1157 |
+
"step": 810
|
1158 |
+
},
|
1159 |
+
{
|
1160 |
+
"epoch": 0.87,
|
1161 |
+
"learning_rate": 2.3390252423108077e-07,
|
1162 |
+
"logits/chosen": -0.10764355957508087,
|
1163 |
+
"logits/rejected": -0.12145823240280151,
|
1164 |
+
"logps/chosen": -505.4794006347656,
|
1165 |
+
"logps/rejected": -602.5386352539062,
|
1166 |
+
"loss": 0.1825,
|
1167 |
+
"rewards/accuracies": 0.5562499761581421,
|
1168 |
+
"rewards/chosen": -0.023508111014962196,
|
1169 |
+
"rewards/margins": 0.05395573377609253,
|
1170 |
+
"rewards/rejected": -0.07746385037899017,
|
1171 |
+
"step": 820
|
1172 |
+
},
|
1173 |
+
{
|
1174 |
+
"epoch": 0.89,
|
1175 |
+
"learning_rate": 1.961371764995243e-07,
|
1176 |
+
"logits/chosen": -0.10990991443395615,
|
1177 |
+
"logits/rejected": -0.13984012603759766,
|
1178 |
+
"logps/chosen": -488.1162109375,
|
1179 |
+
"logps/rejected": -533.455810546875,
|
1180 |
+
"loss": 0.1824,
|
1181 |
+
"rewards/accuracies": 0.4312500059604645,
|
1182 |
+
"rewards/chosen": -0.02799038216471672,
|
1183 |
+
"rewards/margins": 0.03639952093362808,
|
1184 |
+
"rewards/rejected": -0.0643898993730545,
|
1185 |
+
"step": 830
|
1186 |
+
},
|
1187 |
+
{
|
1188 |
+
"epoch": 0.9,
|
1189 |
+
"learning_rate": 1.61571099179261e-07,
|
1190 |
+
"logits/chosen": -0.1293899267911911,
|
1191 |
+
"logits/rejected": -0.13913944363594055,
|
1192 |
+
"logps/chosen": -565.6857299804688,
|
1193 |
+
"logps/rejected": -594.4491577148438,
|
1194 |
+
"loss": 0.1899,
|
1195 |
+
"rewards/accuracies": 0.5,
|
1196 |
+
"rewards/chosen": -0.038701124489307404,
|
1197 |
+
"rewards/margins": 0.03532193973660469,
|
1198 |
+
"rewards/rejected": -0.07402306795120239,
|
1199 |
+
"step": 840
|
1200 |
+
},
|
1201 |
+
{
|
1202 |
+
"epoch": 0.91,
|
1203 |
+
"learning_rate": 1.3025229262312367e-07,
|
1204 |
+
"logits/chosen": -0.11940865218639374,
|
1205 |
+
"logits/rejected": -0.14275464415550232,
|
1206 |
+
"logps/chosen": -517.4210815429688,
|
1207 |
+
"logps/rejected": -559.6951904296875,
|
1208 |
+
"loss": 0.1943,
|
1209 |
+
"rewards/accuracies": 0.4937500059604645,
|
1210 |
+
"rewards/chosen": -0.029175525531172752,
|
1211 |
+
"rewards/margins": 0.038891829550266266,
|
1212 |
+
"rewards/rejected": -0.06806735694408417,
|
1213 |
+
"step": 850
|
1214 |
+
},
|
1215 |
+
{
|
1216 |
+
"epoch": 0.92,
|
1217 |
+
"learning_rate": 1.0222424784546853e-07,
|
1218 |
+
"logits/chosen": -0.14586542546749115,
|
1219 |
+
"logits/rejected": -0.08942988514900208,
|
1220 |
+
"logps/chosen": -556.1111450195312,
|
1221 |
+
"logps/rejected": -648.3941650390625,
|
1222 |
+
"loss": 0.1964,
|
1223 |
+
"rewards/accuracies": 0.550000011920929,
|
1224 |
+
"rewards/chosen": -0.03367298096418381,
|
1225 |
+
"rewards/margins": 0.05108407139778137,
|
1226 |
+
"rewards/rejected": -0.08475704491138458,
|
1227 |
+
"step": 860
|
1228 |
+
},
|
1229 |
+
{
|
1230 |
+
"epoch": 0.93,
|
1231 |
+
"learning_rate": 7.752588612816553e-08,
|
1232 |
+
"logits/chosen": -0.1201920285820961,
|
1233 |
+
"logits/rejected": -0.1033986434340477,
|
1234 |
+
"logps/chosen": -535.7073364257812,
|
1235 |
+
"logps/rejected": -629.0509033203125,
|
1236 |
+
"loss": 0.1942,
|
1237 |
+
"rewards/accuracies": 0.53125,
|
1238 |
+
"rewards/chosen": -0.03432587534189224,
|
1239 |
+
"rewards/margins": 0.05863683670759201,
|
1240 |
+
"rewards/rejected": -0.09296270459890366,
|
1241 |
+
"step": 870
|
1242 |
+
},
|
1243 |
+
{
|
1244 |
+
"epoch": 0.94,
|
1245 |
+
"learning_rate": 5.619150497236991e-08,
|
1246 |
+
"logits/chosen": -0.06793894618749619,
|
1247 |
+
"logits/rejected": -0.14243273437023163,
|
1248 |
+
"logps/chosen": -543.0166625976562,
|
1249 |
+
"logps/rejected": -573.83349609375,
|
1250 |
+
"loss": 0.1839,
|
1251 |
+
"rewards/accuracies": 0.48750001192092896,
|
1252 |
+
"rewards/chosen": -0.02393261529505253,
|
1253 |
+
"rewards/margins": 0.04055067151784897,
|
1254 |
+
"rewards/rejected": -0.06448328495025635,
|
1255 |
+
"step": 880
|
1256 |
+
},
|
1257 |
+
{
|
1258 |
+
"epoch": 0.95,
|
1259 |
+
"learning_rate": 3.825073047112743e-08,
|
1260 |
+
"logits/chosen": -0.09055639803409576,
|
1261 |
+
"logits/rejected": -0.09645718336105347,
|
1262 |
+
"logps/chosen": -589.1398315429688,
|
1263 |
+
"logps/rejected": -638.0357666015625,
|
1264 |
+
"loss": 0.2051,
|
1265 |
+
"rewards/accuracies": 0.5249999761581421,
|
1266 |
+
"rewards/chosen": -0.0366409607231617,
|
1267 |
+
"rewards/margins": 0.046200644224882126,
|
1268 |
+
"rewards/rejected": -0.08284159749746323,
|
1269 |
+
"step": 890
|
1270 |
+
},
|
1271 |
+
{
|
1272 |
+
"epoch": 0.96,
|
1273 |
+
"learning_rate": 2.372847616895685e-08,
|
1274 |
+
"logits/chosen": -0.24545852839946747,
|
1275 |
+
"logits/rejected": -0.13447916507720947,
|
1276 |
+
"logps/chosen": -500.57177734375,
|
1277 |
+
"logps/rejected": -562.6080322265625,
|
1278 |
+
"loss": 0.1894,
|
1279 |
+
"rewards/accuracies": 0.4375,
|
1280 |
+
"rewards/chosen": -0.04158453270792961,
|
1281 |
+
"rewards/margins": 0.03680362552404404,
|
1282 |
+
"rewards/rejected": -0.07838816195726395,
|
1283 |
+
"step": 900
|
1284 |
+
},
|
1285 |
+
{
|
1286 |
+
"epoch": 0.97,
|
1287 |
+
"learning_rate": 1.264490846553279e-08,
|
1288 |
+
"logits/chosen": -0.15378254652023315,
|
1289 |
+
"logits/rejected": -0.09510638564825058,
|
1290 |
+
"logps/chosen": -532.951416015625,
|
1291 |
+
"logps/rejected": -582.2918090820312,
|
1292 |
+
"loss": 0.1947,
|
1293 |
+
"rewards/accuracies": 0.5062500238418579,
|
1294 |
+
"rewards/chosen": -0.033251263201236725,
|
1295 |
+
"rewards/margins": 0.04195471107959747,
|
1296 |
+
"rewards/rejected": -0.0752059668302536,
|
1297 |
+
"step": 910
|
1298 |
+
},
|
1299 |
+
{
|
1300 |
+
"epoch": 0.98,
|
1301 |
+
"learning_rate": 5.015418611516165e-09,
|
1302 |
+
"logits/chosen": -0.06104772537946701,
|
1303 |
+
"logits/rejected": -0.19414536654949188,
|
1304 |
+
"logps/chosen": -515.5169067382812,
|
1305 |
+
"logps/rejected": -601.0435791015625,
|
1306 |
+
"loss": 0.1897,
|
1307 |
+
"rewards/accuracies": 0.512499988079071,
|
1308 |
+
"rewards/chosen": -0.03219000995159149,
|
1309 |
+
"rewards/margins": 0.048967912793159485,
|
1310 |
+
"rewards/rejected": -0.08115791529417038,
|
1311 |
+
"step": 920
|
1312 |
+
},
|
1313 |
+
{
|
1314 |
+
"epoch": 0.99,
|
1315 |
+
"learning_rate": 8.506013354186993e-10,
|
1316 |
+
"logits/chosen": -0.08989392220973969,
|
1317 |
+
"logits/rejected": -0.16842840611934662,
|
1318 |
+
"logps/chosen": -527.9075927734375,
|
1319 |
+
"logps/rejected": -573.1229858398438,
|
1320 |
+
"loss": 0.1872,
|
1321 |
+
"rewards/accuracies": 0.512499988079071,
|
1322 |
+
"rewards/chosen": -0.03477403521537781,
|
1323 |
+
"rewards/margins": 0.039979539811611176,
|
1324 |
+
"rewards/rejected": -0.07475356757640839,
|
1325 |
+
"step": 930
|
1326 |
+
},
|
1327 |
+
{
|
1328 |
+
"epoch": 1.0,
|
1329 |
+
"step": 937,
|
1330 |
+
"total_flos": 0.0,
|
1331 |
+
"train_loss": 0.08862862108611247,
|
1332 |
+
"train_runtime": 3263.0882,
|
1333 |
+
"train_samples_per_second": 9.194,
|
1334 |
+
"train_steps_per_second": 0.287
|
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 |
+
}
|