Model save
Browse files- README.md +60 -0
- all_results.json +9 -0
- generation_config.json +12 -0
- train_results.json +9 -0
- trainer_state.json +1932 -0
README.md
ADDED
|
@@ -0,0 +1,60 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: llama3
|
| 3 |
+
base_model: Jackie999/llama3-sudo-5epochs-tofu_full_sft
|
| 4 |
+
tags:
|
| 5 |
+
- trl
|
| 6 |
+
- dpo
|
| 7 |
+
- generated_from_trainer
|
| 8 |
+
model-index:
|
| 9 |
+
- name: llama3-sudo-dpo-10epochs-forget10mixall-1sft-2fullpara-1e-5
|
| 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 |
+
# llama3-sudo-dpo-10epochs-forget10mixall-1sft-2fullpara-1e-5
|
| 17 |
+
|
| 18 |
+
This model is a fine-tuned version of [Jackie999/llama3-sudo-5epochs-tofu_full_sft](https://huggingface.co/Jackie999/llama3-sudo-5epochs-tofu_full_sft) on an unknown 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: 1e-05
|
| 38 |
+
- train_batch_size: 4
|
| 39 |
+
- eval_batch_size: 8
|
| 40 |
+
- seed: 42
|
| 41 |
+
- distributed_type: multi-GPU
|
| 42 |
+
- num_devices: 4
|
| 43 |
+
- gradient_accumulation_steps: 4
|
| 44 |
+
- total_train_batch_size: 64
|
| 45 |
+
- total_eval_batch_size: 32
|
| 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: 10
|
| 50 |
+
|
| 51 |
+
### Training results
|
| 52 |
+
|
| 53 |
+
|
| 54 |
+
|
| 55 |
+
### Framework versions
|
| 56 |
+
|
| 57 |
+
- Transformers 4.44.0
|
| 58 |
+
- Pytorch 2.1.2
|
| 59 |
+
- Datasets 3.0.0
|
| 60 |
+
- Tokenizers 0.19.1
|
all_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"total_flos": 0.0,
|
| 4 |
+
"train_loss": 0.015015173686749768,
|
| 5 |
+
"train_runtime": 115157.0828,
|
| 6 |
+
"train_samples": 8000,
|
| 7 |
+
"train_samples_per_second": 0.695,
|
| 8 |
+
"train_steps_per_second": 0.011
|
| 9 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token_id": 128000,
|
| 3 |
+
"do_sample": true,
|
| 4 |
+
"eos_token_id": [
|
| 5 |
+
128001,
|
| 6 |
+
128009
|
| 7 |
+
],
|
| 8 |
+
"max_length": 4096,
|
| 9 |
+
"temperature": 0.6,
|
| 10 |
+
"top_p": 0.9,
|
| 11 |
+
"transformers_version": "4.44.0"
|
| 12 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"total_flos": 0.0,
|
| 4 |
+
"train_loss": 0.015015173686749768,
|
| 5 |
+
"train_runtime": 115157.0828,
|
| 6 |
+
"train_samples": 8000,
|
| 7 |
+
"train_samples_per_second": 0.695,
|
| 8 |
+
"train_steps_per_second": 0.011
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1932 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 10.0,
|
| 5 |
+
"eval_steps": 1000,
|
| 6 |
+
"global_step": 1250,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.008,
|
| 13 |
+
"grad_norm": 7.598788468990178,
|
| 14 |
+
"learning_rate": 8e-08,
|
| 15 |
+
"logits/chosen": -1.7671998739242554,
|
| 16 |
+
"logits/rejected": -2.2639822959899902,
|
| 17 |
+
"logps/chosen": -46.430763244628906,
|
| 18 |
+
"logps/rejected": -102.85381317138672,
|
| 19 |
+
"loss": 0.6931,
|
| 20 |
+
"rewards/accuracies": 0.0,
|
| 21 |
+
"rewards/chosen": 0.0,
|
| 22 |
+
"rewards/margins": 0.0,
|
| 23 |
+
"rewards/rejected": 0.0,
|
| 24 |
+
"step": 1
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.08,
|
| 28 |
+
"grad_norm": 8.744481268270988,
|
| 29 |
+
"learning_rate": 8.000000000000001e-07,
|
| 30 |
+
"logits/chosen": -1.6799863576889038,
|
| 31 |
+
"logits/rejected": -2.1152544021606445,
|
| 32 |
+
"logps/chosen": -50.29410934448242,
|
| 33 |
+
"logps/rejected": -101.58956909179688,
|
| 34 |
+
"loss": 0.6791,
|
| 35 |
+
"rewards/accuracies": 0.7291666865348816,
|
| 36 |
+
"rewards/chosen": 0.0007337681599892676,
|
| 37 |
+
"rewards/margins": 0.02895486354827881,
|
| 38 |
+
"rewards/rejected": -0.028221096843481064,
|
| 39 |
+
"step": 10
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 0.16,
|
| 43 |
+
"grad_norm": 3.572341321811657,
|
| 44 |
+
"learning_rate": 1.6000000000000001e-06,
|
| 45 |
+
"logits/chosen": -1.558701992034912,
|
| 46 |
+
"logits/rejected": -2.037860870361328,
|
| 47 |
+
"logps/chosen": -41.180152893066406,
|
| 48 |
+
"logps/rejected": -187.47267150878906,
|
| 49 |
+
"loss": 0.4012,
|
| 50 |
+
"rewards/accuracies": 0.9937499761581421,
|
| 51 |
+
"rewards/chosen": 0.05867023020982742,
|
| 52 |
+
"rewards/margins": 0.8922162055969238,
|
| 53 |
+
"rewards/rejected": -0.833545982837677,
|
| 54 |
+
"step": 20
|
| 55 |
+
},
|
| 56 |
+
{
|
| 57 |
+
"epoch": 0.24,
|
| 58 |
+
"grad_norm": 4.668066827355425,
|
| 59 |
+
"learning_rate": 2.4000000000000003e-06,
|
| 60 |
+
"logits/chosen": -1.24845290184021,
|
| 61 |
+
"logits/rejected": -1.647834062576294,
|
| 62 |
+
"logps/chosen": -57.645042419433594,
|
| 63 |
+
"logps/rejected": -346.35516357421875,
|
| 64 |
+
"loss": 0.2058,
|
| 65 |
+
"rewards/accuracies": 0.925000011920929,
|
| 66 |
+
"rewards/chosen": -0.06952311843633652,
|
| 67 |
+
"rewards/margins": 2.4007816314697266,
|
| 68 |
+
"rewards/rejected": -2.470304489135742,
|
| 69 |
+
"step": 30
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 0.32,
|
| 73 |
+
"grad_norm": 3.5896956533073925,
|
| 74 |
+
"learning_rate": 3.2000000000000003e-06,
|
| 75 |
+
"logits/chosen": -0.7071009874343872,
|
| 76 |
+
"logits/rejected": -1.1559783220291138,
|
| 77 |
+
"logps/chosen": -38.179378509521484,
|
| 78 |
+
"logps/rejected": -475.6770935058594,
|
| 79 |
+
"loss": 0.158,
|
| 80 |
+
"rewards/accuracies": 0.9750000238418579,
|
| 81 |
+
"rewards/chosen": 0.1055324524641037,
|
| 82 |
+
"rewards/margins": 3.844930648803711,
|
| 83 |
+
"rewards/rejected": -3.739398241043091,
|
| 84 |
+
"step": 40
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 0.4,
|
| 88 |
+
"grad_norm": 6.069124428046055,
|
| 89 |
+
"learning_rate": 4.000000000000001e-06,
|
| 90 |
+
"logits/chosen": -0.44346046447753906,
|
| 91 |
+
"logits/rejected": -0.9727121591567993,
|
| 92 |
+
"logps/chosen": -35.12229919433594,
|
| 93 |
+
"logps/rejected": -532.2540893554688,
|
| 94 |
+
"loss": 0.1132,
|
| 95 |
+
"rewards/accuracies": 0.9750000238418579,
|
| 96 |
+
"rewards/chosen": 0.13596299290657043,
|
| 97 |
+
"rewards/margins": 4.437729358673096,
|
| 98 |
+
"rewards/rejected": -4.301766395568848,
|
| 99 |
+
"step": 50
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 0.48,
|
| 103 |
+
"grad_norm": 2.7055226938318175,
|
| 104 |
+
"learning_rate": 4.800000000000001e-06,
|
| 105 |
+
"logits/chosen": -0.20063920319080353,
|
| 106 |
+
"logits/rejected": -0.6503936648368835,
|
| 107 |
+
"logps/chosen": -47.3806037902832,
|
| 108 |
+
"logps/rejected": -589.556396484375,
|
| 109 |
+
"loss": 0.0932,
|
| 110 |
+
"rewards/accuracies": 0.9375,
|
| 111 |
+
"rewards/chosen": 0.03769879788160324,
|
| 112 |
+
"rewards/margins": 4.943137168884277,
|
| 113 |
+
"rewards/rejected": -4.9054388999938965,
|
| 114 |
+
"step": 60
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.56,
|
| 118 |
+
"grad_norm": 0.25261425361308987,
|
| 119 |
+
"learning_rate": 5.600000000000001e-06,
|
| 120 |
+
"logits/chosen": -0.23743292689323425,
|
| 121 |
+
"logits/rejected": -0.7351670861244202,
|
| 122 |
+
"logps/chosen": -63.814796447753906,
|
| 123 |
+
"logps/rejected": -670.7484741210938,
|
| 124 |
+
"loss": 0.0621,
|
| 125 |
+
"rewards/accuracies": 0.987500011920929,
|
| 126 |
+
"rewards/chosen": -0.13264700770378113,
|
| 127 |
+
"rewards/margins": 5.5608930587768555,
|
| 128 |
+
"rewards/rejected": -5.693540573120117,
|
| 129 |
+
"step": 70
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.64,
|
| 133 |
+
"grad_norm": 6.481510280384696,
|
| 134 |
+
"learning_rate": 6.4000000000000006e-06,
|
| 135 |
+
"logits/chosen": -0.03323043882846832,
|
| 136 |
+
"logits/rejected": -0.603085994720459,
|
| 137 |
+
"logps/chosen": -121.7635498046875,
|
| 138 |
+
"logps/rejected": -793.174560546875,
|
| 139 |
+
"loss": 0.0371,
|
| 140 |
+
"rewards/accuracies": 0.9937499761581421,
|
| 141 |
+
"rewards/chosen": -0.7093967199325562,
|
| 142 |
+
"rewards/margins": 6.21297550201416,
|
| 143 |
+
"rewards/rejected": -6.922372341156006,
|
| 144 |
+
"step": 80
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 0.72,
|
| 148 |
+
"grad_norm": 0.9177658561115962,
|
| 149 |
+
"learning_rate": 7.2000000000000005e-06,
|
| 150 |
+
"logits/chosen": 0.006849220488220453,
|
| 151 |
+
"logits/rejected": -0.6091193556785583,
|
| 152 |
+
"logps/chosen": -155.50479125976562,
|
| 153 |
+
"logps/rejected": -916.3829345703125,
|
| 154 |
+
"loss": 0.0084,
|
| 155 |
+
"rewards/accuracies": 1.0,
|
| 156 |
+
"rewards/chosen": -1.0437133312225342,
|
| 157 |
+
"rewards/margins": 7.108320713043213,
|
| 158 |
+
"rewards/rejected": -8.152033805847168,
|
| 159 |
+
"step": 90
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 0.8,
|
| 163 |
+
"grad_norm": 0.06845804795916455,
|
| 164 |
+
"learning_rate": 8.000000000000001e-06,
|
| 165 |
+
"logits/chosen": 0.13967576622962952,
|
| 166 |
+
"logits/rejected": -0.4425061345100403,
|
| 167 |
+
"logps/chosen": -133.66134643554688,
|
| 168 |
+
"logps/rejected": -994.0302734375,
|
| 169 |
+
"loss": 0.0126,
|
| 170 |
+
"rewards/accuracies": 0.9937499761581421,
|
| 171 |
+
"rewards/chosen": -0.8357334136962891,
|
| 172 |
+
"rewards/margins": 8.085713386535645,
|
| 173 |
+
"rewards/rejected": -8.921446800231934,
|
| 174 |
+
"step": 100
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 0.88,
|
| 178 |
+
"grad_norm": 5.6107817864989356,
|
| 179 |
+
"learning_rate": 8.8e-06,
|
| 180 |
+
"logits/chosen": 0.22975310683250427,
|
| 181 |
+
"logits/rejected": -0.414044052362442,
|
| 182 |
+
"logps/chosen": -106.53846740722656,
|
| 183 |
+
"logps/rejected": -1022.2738037109375,
|
| 184 |
+
"loss": 0.0208,
|
| 185 |
+
"rewards/accuracies": 0.9937499761581421,
|
| 186 |
+
"rewards/chosen": -0.5662684440612793,
|
| 187 |
+
"rewards/margins": 8.641936302185059,
|
| 188 |
+
"rewards/rejected": -9.20820426940918,
|
| 189 |
+
"step": 110
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 0.96,
|
| 193 |
+
"grad_norm": 0.01682091967533478,
|
| 194 |
+
"learning_rate": 9.600000000000001e-06,
|
| 195 |
+
"logits/chosen": 0.3314499855041504,
|
| 196 |
+
"logits/rejected": -0.5161079168319702,
|
| 197 |
+
"logps/chosen": -112.84576416015625,
|
| 198 |
+
"logps/rejected": -1047.9407958984375,
|
| 199 |
+
"loss": 0.0003,
|
| 200 |
+
"rewards/accuracies": 1.0,
|
| 201 |
+
"rewards/chosen": -0.6609307527542114,
|
| 202 |
+
"rewards/margins": 8.777119636535645,
|
| 203 |
+
"rewards/rejected": -9.438051223754883,
|
| 204 |
+
"step": 120
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 1.04,
|
| 208 |
+
"grad_norm": 2.1463004224597944,
|
| 209 |
+
"learning_rate": 9.999512620046523e-06,
|
| 210 |
+
"logits/chosen": 0.020063649863004684,
|
| 211 |
+
"logits/rejected": -0.6272026300430298,
|
| 212 |
+
"logps/chosen": -129.10415649414062,
|
| 213 |
+
"logps/rejected": -1125.7564697265625,
|
| 214 |
+
"loss": 0.0126,
|
| 215 |
+
"rewards/accuracies": 1.0,
|
| 216 |
+
"rewards/chosen": -0.7790005207061768,
|
| 217 |
+
"rewards/margins": 9.478322982788086,
|
| 218 |
+
"rewards/rejected": -10.257322311401367,
|
| 219 |
+
"step": 130
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 1.12,
|
| 223 |
+
"grad_norm": 0.12425603579424757,
|
| 224 |
+
"learning_rate": 9.995614150494293e-06,
|
| 225 |
+
"logits/chosen": -0.6389492154121399,
|
| 226 |
+
"logits/rejected": -1.3565341234207153,
|
| 227 |
+
"logps/chosen": -101.65057373046875,
|
| 228 |
+
"logps/rejected": -955.8663330078125,
|
| 229 |
+
"loss": 0.0057,
|
| 230 |
+
"rewards/accuracies": 1.0,
|
| 231 |
+
"rewards/chosen": -0.516974925994873,
|
| 232 |
+
"rewards/margins": 8.029803276062012,
|
| 233 |
+
"rewards/rejected": -8.546777725219727,
|
| 234 |
+
"step": 140
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 1.2,
|
| 238 |
+
"grad_norm": 0.013089818267371892,
|
| 239 |
+
"learning_rate": 9.987820251299121e-06,
|
| 240 |
+
"logits/chosen": -0.40948599576950073,
|
| 241 |
+
"logits/rejected": -1.0153663158416748,
|
| 242 |
+
"logps/chosen": -87.0394287109375,
|
| 243 |
+
"logps/rejected": -1100.132568359375,
|
| 244 |
+
"loss": 0.0004,
|
| 245 |
+
"rewards/accuracies": 1.0,
|
| 246 |
+
"rewards/chosen": -0.3547939658164978,
|
| 247 |
+
"rewards/margins": 9.640316009521484,
|
| 248 |
+
"rewards/rejected": -9.995109558105469,
|
| 249 |
+
"step": 150
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 1.28,
|
| 253 |
+
"grad_norm": 0.01621300125862196,
|
| 254 |
+
"learning_rate": 9.976136999909156e-06,
|
| 255 |
+
"logits/chosen": -0.7102096080780029,
|
| 256 |
+
"logits/rejected": -1.363110065460205,
|
| 257 |
+
"logps/chosen": -88.75753021240234,
|
| 258 |
+
"logps/rejected": -1060.689208984375,
|
| 259 |
+
"loss": 0.0052,
|
| 260 |
+
"rewards/accuracies": 0.9937499761581421,
|
| 261 |
+
"rewards/chosen": -0.3866530954837799,
|
| 262 |
+
"rewards/margins": 9.206262588500977,
|
| 263 |
+
"rewards/rejected": -9.592916488647461,
|
| 264 |
+
"step": 160
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 1.3599999999999999,
|
| 268 |
+
"grad_norm": 0.0026423921093196084,
|
| 269 |
+
"learning_rate": 9.960573506572391e-06,
|
| 270 |
+
"logits/chosen": -0.8623284101486206,
|
| 271 |
+
"logits/rejected": -1.6030025482177734,
|
| 272 |
+
"logps/chosen": -77.21472930908203,
|
| 273 |
+
"logps/rejected": -1054.2359619140625,
|
| 274 |
+
"loss": 0.0006,
|
| 275 |
+
"rewards/accuracies": 1.0,
|
| 276 |
+
"rewards/chosen": -0.29167506098747253,
|
| 277 |
+
"rewards/margins": 9.230081558227539,
|
| 278 |
+
"rewards/rejected": -9.521757125854492,
|
| 279 |
+
"step": 170
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 1.44,
|
| 283 |
+
"grad_norm": 0.003569937734506862,
|
| 284 |
+
"learning_rate": 9.941141907232766e-06,
|
| 285 |
+
"logits/chosen": -0.7774030566215515,
|
| 286 |
+
"logits/rejected": -1.4183218479156494,
|
| 287 |
+
"logps/chosen": -95.35404968261719,
|
| 288 |
+
"logps/rejected": -1243.467529296875,
|
| 289 |
+
"loss": 0.0003,
|
| 290 |
+
"rewards/accuracies": 1.0,
|
| 291 |
+
"rewards/chosen": -0.45167961716651917,
|
| 292 |
+
"rewards/margins": 10.976358413696289,
|
| 293 |
+
"rewards/rejected": -11.428037643432617,
|
| 294 |
+
"step": 180
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 1.52,
|
| 298 |
+
"grad_norm": 0.004345604176761015,
|
| 299 |
+
"learning_rate": 9.91785735406693e-06,
|
| 300 |
+
"logits/chosen": -0.6248332262039185,
|
| 301 |
+
"logits/rejected": -1.2730491161346436,
|
| 302 |
+
"logps/chosen": -98.45368957519531,
|
| 303 |
+
"logps/rejected": -1197.553466796875,
|
| 304 |
+
"loss": 0.007,
|
| 305 |
+
"rewards/accuracies": 1.0,
|
| 306 |
+
"rewards/chosen": -0.48718851804733276,
|
| 307 |
+
"rewards/margins": 10.482004165649414,
|
| 308 |
+
"rewards/rejected": -10.969191551208496,
|
| 309 |
+
"step": 190
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 1.6,
|
| 313 |
+
"grad_norm": 0.6594773158021805,
|
| 314 |
+
"learning_rate": 9.890738003669029e-06,
|
| 315 |
+
"logits/chosen": -0.5561812520027161,
|
| 316 |
+
"logits/rejected": -1.166264533996582,
|
| 317 |
+
"logps/chosen": -94.01605224609375,
|
| 318 |
+
"logps/rejected": -1177.841064453125,
|
| 319 |
+
"loss": 0.0013,
|
| 320 |
+
"rewards/accuracies": 1.0,
|
| 321 |
+
"rewards/chosen": -0.43924322724342346,
|
| 322 |
+
"rewards/margins": 10.321309089660645,
|
| 323 |
+
"rewards/rejected": -10.760551452636719,
|
| 324 |
+
"step": 200
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 1.6800000000000002,
|
| 328 |
+
"grad_norm": 0.014550855783893434,
|
| 329 |
+
"learning_rate": 9.859805002892733e-06,
|
| 330 |
+
"logits/chosen": -0.30123084783554077,
|
| 331 |
+
"logits/rejected": -0.9458983540534973,
|
| 332 |
+
"logps/chosen": -84.95938110351562,
|
| 333 |
+
"logps/rejected": -1145.023681640625,
|
| 334 |
+
"loss": 0.001,
|
| 335 |
+
"rewards/accuracies": 1.0,
|
| 336 |
+
"rewards/chosen": -0.35356029868125916,
|
| 337 |
+
"rewards/margins": 10.082305908203125,
|
| 338 |
+
"rewards/rejected": -10.435864448547363,
|
| 339 |
+
"step": 210
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 1.76,
|
| 343 |
+
"grad_norm": 0.004377641691260569,
|
| 344 |
+
"learning_rate": 9.825082472361558e-06,
|
| 345 |
+
"logits/chosen": -0.36692723631858826,
|
| 346 |
+
"logits/rejected": -1.0984880924224854,
|
| 347 |
+
"logps/chosen": -58.7005615234375,
|
| 348 |
+
"logps/rejected": -1094.305419921875,
|
| 349 |
+
"loss": 0.0071,
|
| 350 |
+
"rewards/accuracies": 1.0,
|
| 351 |
+
"rewards/chosen": -0.10480108112096786,
|
| 352 |
+
"rewards/margins": 9.808042526245117,
|
| 353 |
+
"rewards/rejected": -9.91284465789795,
|
| 354 |
+
"step": 220
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 1.8399999999999999,
|
| 358 |
+
"grad_norm": 0.018821455289326113,
|
| 359 |
+
"learning_rate": 9.786597487660336e-06,
|
| 360 |
+
"logits/chosen": -0.23969046771526337,
|
| 361 |
+
"logits/rejected": -0.9250698089599609,
|
| 362 |
+
"logps/chosen": -67.17304992675781,
|
| 363 |
+
"logps/rejected": -1114.4503173828125,
|
| 364 |
+
"loss": 0.0004,
|
| 365 |
+
"rewards/accuracies": 1.0,
|
| 366 |
+
"rewards/chosen": -0.18792255222797394,
|
| 367 |
+
"rewards/margins": 9.926887512207031,
|
| 368 |
+
"rewards/rejected": -10.114809036254883,
|
| 369 |
+
"step": 230
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 1.92,
|
| 373 |
+
"grad_norm": 0.017614530283405035,
|
| 374 |
+
"learning_rate": 9.744380058222483e-06,
|
| 375 |
+
"logits/chosen": -0.45373648405075073,
|
| 376 |
+
"logits/rejected": -1.0718930959701538,
|
| 377 |
+
"logps/chosen": -128.1130828857422,
|
| 378 |
+
"logps/rejected": -1161.447509765625,
|
| 379 |
+
"loss": 0.0039,
|
| 380 |
+
"rewards/accuracies": 1.0,
|
| 381 |
+
"rewards/chosen": -0.774398684501648,
|
| 382 |
+
"rewards/margins": 9.828145980834961,
|
| 383 |
+
"rewards/rejected": -10.602544784545898,
|
| 384 |
+
"step": 240
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 2.0,
|
| 388 |
+
"grad_norm": 0.020987613586006735,
|
| 389 |
+
"learning_rate": 9.698463103929542e-06,
|
| 390 |
+
"logits/chosen": -0.34758588671684265,
|
| 391 |
+
"logits/rejected": -0.9034555554389954,
|
| 392 |
+
"logps/chosen": -164.2602996826172,
|
| 393 |
+
"logps/rejected": -1219.2265625,
|
| 394 |
+
"loss": 0.0004,
|
| 395 |
+
"rewards/accuracies": 1.0,
|
| 396 |
+
"rewards/chosen": -1.1342051029205322,
|
| 397 |
+
"rewards/margins": 10.054884910583496,
|
| 398 |
+
"rewards/rejected": -11.189090728759766,
|
| 399 |
+
"step": 250
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 2.08,
|
| 403 |
+
"grad_norm": 0.004509813266090499,
|
| 404 |
+
"learning_rate": 9.648882429441258e-06,
|
| 405 |
+
"logits/chosen": -0.07843346893787384,
|
| 406 |
+
"logits/rejected": -0.7051417827606201,
|
| 407 |
+
"logps/chosen": -154.61831665039062,
|
| 408 |
+
"logps/rejected": -1228.2486572265625,
|
| 409 |
+
"loss": 0.0014,
|
| 410 |
+
"rewards/accuracies": 1.0,
|
| 411 |
+
"rewards/chosen": -1.0459339618682861,
|
| 412 |
+
"rewards/margins": 10.211647987365723,
|
| 413 |
+
"rewards/rejected": -11.25758171081543,
|
| 414 |
+
"step": 260
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 2.16,
|
| 418 |
+
"grad_norm": 0.10540035188708549,
|
| 419 |
+
"learning_rate": 9.595676696276173e-06,
|
| 420 |
+
"logits/chosen": -0.16252107918262482,
|
| 421 |
+
"logits/rejected": -0.6854008436203003,
|
| 422 |
+
"logps/chosen": -215.68759155273438,
|
| 423 |
+
"logps/rejected": -1336.6497802734375,
|
| 424 |
+
"loss": 0.0007,
|
| 425 |
+
"rewards/accuracies": 1.0,
|
| 426 |
+
"rewards/chosen": -1.645402193069458,
|
| 427 |
+
"rewards/margins": 10.722552299499512,
|
| 428 |
+
"rewards/rejected": -12.367955207824707,
|
| 429 |
+
"step": 270
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 2.24,
|
| 433 |
+
"grad_norm": 0.0014480877122579167,
|
| 434 |
+
"learning_rate": 9.538887392664544e-06,
|
| 435 |
+
"logits/chosen": -0.04962470754981041,
|
| 436 |
+
"logits/rejected": -0.597977340221405,
|
| 437 |
+
"logps/chosen": -182.0589599609375,
|
| 438 |
+
"logps/rejected": -1372.7177734375,
|
| 439 |
+
"loss": 0.0003,
|
| 440 |
+
"rewards/accuracies": 1.0,
|
| 441 |
+
"rewards/chosen": -1.3132940530776978,
|
| 442 |
+
"rewards/margins": 11.408597946166992,
|
| 443 |
+
"rewards/rejected": -12.721891403198242,
|
| 444 |
+
"step": 280
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 2.32,
|
| 448 |
+
"grad_norm": 0.0169674710944411,
|
| 449 |
+
"learning_rate": 9.478558801197065e-06,
|
| 450 |
+
"logits/chosen": -0.06320101767778397,
|
| 451 |
+
"logits/rejected": -0.6581576466560364,
|
| 452 |
+
"logps/chosen": -164.2813720703125,
|
| 453 |
+
"logps/rejected": -1302.8653564453125,
|
| 454 |
+
"loss": 0.0071,
|
| 455 |
+
"rewards/accuracies": 0.9937499761581421,
|
| 456 |
+
"rewards/chosen": -1.146195888519287,
|
| 457 |
+
"rewards/margins": 10.875839233398438,
|
| 458 |
+
"rewards/rejected": -12.0220365524292,
|
| 459 |
+
"step": 290
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 2.4,
|
| 463 |
+
"grad_norm": 0.0007002624784587229,
|
| 464 |
+
"learning_rate": 9.414737964294636e-06,
|
| 465 |
+
"logits/chosen": -0.18863362073898315,
|
| 466 |
+
"logits/rejected": -0.7296714782714844,
|
| 467 |
+
"logps/chosen": -155.30770874023438,
|
| 468 |
+
"logps/rejected": -1376.25830078125,
|
| 469 |
+
"loss": 0.0002,
|
| 470 |
+
"rewards/accuracies": 1.0,
|
| 471 |
+
"rewards/chosen": -1.0464993715286255,
|
| 472 |
+
"rewards/margins": 11.702775001525879,
|
| 473 |
+
"rewards/rejected": -12.749274253845215,
|
| 474 |
+
"step": 300
|
| 475 |
+
},
|
| 476 |
+
{
|
| 477 |
+
"epoch": 2.48,
|
| 478 |
+
"grad_norm": 0.0005702077735596562,
|
| 479 |
+
"learning_rate": 9.347474647526095e-06,
|
| 480 |
+
"logits/chosen": -0.2102803885936737,
|
| 481 |
+
"logits/rejected": -0.6938263773918152,
|
| 482 |
+
"logps/chosen": -163.68673706054688,
|
| 483 |
+
"logps/rejected": -1451.975830078125,
|
| 484 |
+
"loss": 0.0001,
|
| 485 |
+
"rewards/accuracies": 1.0,
|
| 486 |
+
"rewards/chosen": -1.128525972366333,
|
| 487 |
+
"rewards/margins": 12.382759094238281,
|
| 488 |
+
"rewards/rejected": -13.511285781860352,
|
| 489 |
+
"step": 310
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 2.56,
|
| 493 |
+
"grad_norm": 0.0011316887816189948,
|
| 494 |
+
"learning_rate": 9.276821300802535e-06,
|
| 495 |
+
"logits/chosen": -0.20408546924591064,
|
| 496 |
+
"logits/rejected": -0.7608290910720825,
|
| 497 |
+
"logps/chosen": -148.53030395507812,
|
| 498 |
+
"logps/rejected": -1382.659423828125,
|
| 499 |
+
"loss": 0.0002,
|
| 500 |
+
"rewards/accuracies": 1.0,
|
| 501 |
+
"rewards/chosen": -0.995039165019989,
|
| 502 |
+
"rewards/margins": 11.816390991210938,
|
| 503 |
+
"rewards/rejected": -12.811429977416992,
|
| 504 |
+
"step": 320
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 2.64,
|
| 508 |
+
"grad_norm": 0.0028257198320044193,
|
| 509 |
+
"learning_rate": 9.202833017478421e-06,
|
| 510 |
+
"logits/chosen": -0.23039910197257996,
|
| 511 |
+
"logits/rejected": -0.8503023386001587,
|
| 512 |
+
"logps/chosen": -127.6865234375,
|
| 513 |
+
"logps/rejected": -1280.506591796875,
|
| 514 |
+
"loss": 0.0,
|
| 515 |
+
"rewards/accuracies": 1.0,
|
| 516 |
+
"rewards/chosen": -0.78924959897995,
|
| 517 |
+
"rewards/margins": 10.988354682922363,
|
| 518 |
+
"rewards/rejected": -11.777605056762695,
|
| 519 |
+
"step": 330
|
| 520 |
+
},
|
| 521 |
+
{
|
| 522 |
+
"epoch": 2.7199999999999998,
|
| 523 |
+
"grad_norm": 0.03275958999147167,
|
| 524 |
+
"learning_rate": 9.125567491391476e-06,
|
| 525 |
+
"logits/chosen": -0.16481304168701172,
|
| 526 |
+
"logits/rejected": -0.6871527433395386,
|
| 527 |
+
"logps/chosen": -165.09945678710938,
|
| 528 |
+
"logps/rejected": -1415.8587646484375,
|
| 529 |
+
"loss": 0.0019,
|
| 530 |
+
"rewards/accuracies": 1.0,
|
| 531 |
+
"rewards/chosen": -1.147678017616272,
|
| 532 |
+
"rewards/margins": 12.000423431396484,
|
| 533 |
+
"rewards/rejected": -13.148101806640625,
|
| 534 |
+
"step": 340
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 2.8,
|
| 538 |
+
"grad_norm": 0.040681404406324485,
|
| 539 |
+
"learning_rate": 9.045084971874738e-06,
|
| 540 |
+
"logits/chosen": -0.2651657462120056,
|
| 541 |
+
"logits/rejected": -0.8547052145004272,
|
| 542 |
+
"logps/chosen": -169.6096954345703,
|
| 543 |
+
"logps/rejected": -1442.6204833984375,
|
| 544 |
+
"loss": 0.0005,
|
| 545 |
+
"rewards/accuracies": 1.0,
|
| 546 |
+
"rewards/chosen": -1.185150384902954,
|
| 547 |
+
"rewards/margins": 12.228256225585938,
|
| 548 |
+
"rewards/rejected": -13.413406372070312,
|
| 549 |
+
"step": 350
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 2.88,
|
| 553 |
+
"grad_norm": 0.0005914776735028229,
|
| 554 |
+
"learning_rate": 8.961448216775955e-06,
|
| 555 |
+
"logits/chosen": -0.3672965168952942,
|
| 556 |
+
"logits/rejected": -0.9414142370223999,
|
| 557 |
+
"logps/chosen": -155.79690551757812,
|
| 558 |
+
"logps/rejected": -1344.467529296875,
|
| 559 |
+
"loss": 0.0003,
|
| 560 |
+
"rewards/accuracies": 1.0,
|
| 561 |
+
"rewards/chosen": -1.0553600788116455,
|
| 562 |
+
"rewards/margins": 11.368631362915039,
|
| 563 |
+
"rewards/rejected": -12.423991203308105,
|
| 564 |
+
"step": 360
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"epoch": 2.96,
|
| 568 |
+
"grad_norm": 0.0005670082430647187,
|
| 569 |
+
"learning_rate": 8.874722443520898e-06,
|
| 570 |
+
"logits/chosen": -0.3921353816986084,
|
| 571 |
+
"logits/rejected": -0.8953983187675476,
|
| 572 |
+
"logps/chosen": -158.56431579589844,
|
| 573 |
+
"logps/rejected": -1356.921875,
|
| 574 |
+
"loss": 0.0007,
|
| 575 |
+
"rewards/accuracies": 1.0,
|
| 576 |
+
"rewards/chosen": -1.090203046798706,
|
| 577 |
+
"rewards/margins": 11.472868919372559,
|
| 578 |
+
"rewards/rejected": -12.563072204589844,
|
| 579 |
+
"step": 370
|
| 580 |
+
},
|
| 581 |
+
{
|
| 582 |
+
"epoch": 3.04,
|
| 583 |
+
"grad_norm": 0.005160483245130726,
|
| 584 |
+
"learning_rate": 8.784975278258783e-06,
|
| 585 |
+
"logits/chosen": -0.5358741879463196,
|
| 586 |
+
"logits/rejected": -1.061417579650879,
|
| 587 |
+
"logps/chosen": -175.1329803466797,
|
| 588 |
+
"logps/rejected": -1406.556396484375,
|
| 589 |
+
"loss": 0.0001,
|
| 590 |
+
"rewards/accuracies": 1.0,
|
| 591 |
+
"rewards/chosen": -1.2551603317260742,
|
| 592 |
+
"rewards/margins": 11.804767608642578,
|
| 593 |
+
"rewards/rejected": -13.059926986694336,
|
| 594 |
+
"step": 380
|
| 595 |
+
},
|
| 596 |
+
{
|
| 597 |
+
"epoch": 3.12,
|
| 598 |
+
"grad_norm": 0.006657927449314328,
|
| 599 |
+
"learning_rate": 8.692276703129421e-06,
|
| 600 |
+
"logits/chosen": -0.6002996563911438,
|
| 601 |
+
"logits/rejected": -1.1299346685409546,
|
| 602 |
+
"logps/chosen": -152.69854736328125,
|
| 603 |
+
"logps/rejected": -1418.118896484375,
|
| 604 |
+
"loss": 0.0002,
|
| 605 |
+
"rewards/accuracies": 1.0,
|
| 606 |
+
"rewards/chosen": -1.0260169506072998,
|
| 607 |
+
"rewards/margins": 12.134782791137695,
|
| 608 |
+
"rewards/rejected": -13.160799980163574,
|
| 609 |
+
"step": 390
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 3.2,
|
| 613 |
+
"grad_norm": 0.00039712494546681335,
|
| 614 |
+
"learning_rate": 8.596699001693257e-06,
|
| 615 |
+
"logits/chosen": -0.6303237676620483,
|
| 616 |
+
"logits/rejected": -1.101627230644226,
|
| 617 |
+
"logps/chosen": -196.59848022460938,
|
| 618 |
+
"logps/rejected": -1504.538818359375,
|
| 619 |
+
"loss": 0.0001,
|
| 620 |
+
"rewards/accuracies": 1.0,
|
| 621 |
+
"rewards/chosen": -1.4475935697555542,
|
| 622 |
+
"rewards/margins": 12.599725723266602,
|
| 623 |
+
"rewards/rejected": -14.047317504882812,
|
| 624 |
+
"step": 400
|
| 625 |
+
},
|
| 626 |
+
{
|
| 627 |
+
"epoch": 3.2800000000000002,
|
| 628 |
+
"grad_norm": 0.0040631132506810195,
|
| 629 |
+
"learning_rate": 8.498316702566828e-06,
|
| 630 |
+
"logits/chosen": -0.5605727434158325,
|
| 631 |
+
"logits/rejected": -1.073054313659668,
|
| 632 |
+
"logps/chosen": -163.1142120361328,
|
| 633 |
+
"logps/rejected": -1435.8837890625,
|
| 634 |
+
"loss": 0.0,
|
| 635 |
+
"rewards/accuracies": 1.0,
|
| 636 |
+
"rewards/chosen": -1.1389962434768677,
|
| 637 |
+
"rewards/margins": 12.203179359436035,
|
| 638 |
+
"rewards/rejected": -13.342175483703613,
|
| 639 |
+
"step": 410
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 3.36,
|
| 643 |
+
"grad_norm": 0.041567571634384244,
|
| 644 |
+
"learning_rate": 8.397206521307584e-06,
|
| 645 |
+
"logits/chosen": -0.6039146780967712,
|
| 646 |
+
"logits/rejected": -1.0801668167114258,
|
| 647 |
+
"logps/chosen": -178.77394104003906,
|
| 648 |
+
"logps/rejected": -1475.822998046875,
|
| 649 |
+
"loss": 0.0001,
|
| 650 |
+
"rewards/accuracies": 1.0,
|
| 651 |
+
"rewards/chosen": -1.293518304824829,
|
| 652 |
+
"rewards/margins": 12.463233947753906,
|
| 653 |
+
"rewards/rejected": -13.756753921508789,
|
| 654 |
+
"step": 420
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 3.44,
|
| 658 |
+
"grad_norm": 0.0029527470935980476,
|
| 659 |
+
"learning_rate": 8.293447300593402e-06,
|
| 660 |
+
"logits/chosen": -0.6083961129188538,
|
| 661 |
+
"logits/rejected": -1.0437417030334473,
|
| 662 |
+
"logps/chosen": -192.32000732421875,
|
| 663 |
+
"logps/rejected": -1542.182373046875,
|
| 664 |
+
"loss": 0.0001,
|
| 665 |
+
"rewards/accuracies": 1.0,
|
| 666 |
+
"rewards/chosen": -1.4102303981781006,
|
| 667 |
+
"rewards/margins": 13.007156372070312,
|
| 668 |
+
"rewards/rejected": -14.417387008666992,
|
| 669 |
+
"step": 430
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 3.52,
|
| 673 |
+
"grad_norm": 0.0011462076542202355,
|
| 674 |
+
"learning_rate": 8.18711994874345e-06,
|
| 675 |
+
"logits/chosen": -0.5748112201690674,
|
| 676 |
+
"logits/rejected": -0.9151349067687988,
|
| 677 |
+
"logps/chosen": -222.9849395751953,
|
| 678 |
+
"logps/rejected": -1637.6048583984375,
|
| 679 |
+
"loss": 0.0001,
|
| 680 |
+
"rewards/accuracies": 1.0,
|
| 681 |
+
"rewards/chosen": -1.6979871988296509,
|
| 682 |
+
"rewards/margins": 13.692995071411133,
|
| 683 |
+
"rewards/rejected": -15.390981674194336,
|
| 684 |
+
"step": 440
|
| 685 |
+
},
|
| 686 |
+
{
|
| 687 |
+
"epoch": 3.6,
|
| 688 |
+
"grad_norm": 0.0010367050759902319,
|
| 689 |
+
"learning_rate": 8.078307376628292e-06,
|
| 690 |
+
"logits/chosen": -0.5503496527671814,
|
| 691 |
+
"logits/rejected": -1.0760968923568726,
|
| 692 |
+
"logps/chosen": -169.05398559570312,
|
| 693 |
+
"logps/rejected": -1424.4925537109375,
|
| 694 |
+
"loss": 0.0,
|
| 695 |
+
"rewards/accuracies": 1.0,
|
| 696 |
+
"rewards/chosen": -1.197346568107605,
|
| 697 |
+
"rewards/margins": 12.040412902832031,
|
| 698 |
+
"rewards/rejected": -13.237760543823242,
|
| 699 |
+
"step": 450
|
| 700 |
+
},
|
| 701 |
+
{
|
| 702 |
+
"epoch": 3.68,
|
| 703 |
+
"grad_norm": 0.00030163796426365894,
|
| 704 |
+
"learning_rate": 7.967094433018508e-06,
|
| 705 |
+
"logits/chosen": -0.522539496421814,
|
| 706 |
+
"logits/rejected": -1.063500165939331,
|
| 707 |
+
"logps/chosen": -175.1356658935547,
|
| 708 |
+
"logps/rejected": -1461.3223876953125,
|
| 709 |
+
"loss": 0.0013,
|
| 710 |
+
"rewards/accuracies": 0.9937499761581421,
|
| 711 |
+
"rewards/chosen": -1.2569339275360107,
|
| 712 |
+
"rewards/margins": 12.344822883605957,
|
| 713 |
+
"rewards/rejected": -13.601758003234863,
|
| 714 |
+
"step": 460
|
| 715 |
+
},
|
| 716 |
+
{
|
| 717 |
+
"epoch": 3.76,
|
| 718 |
+
"grad_norm": 0.017735619925505812,
|
| 719 |
+
"learning_rate": 7.85356783842216e-06,
|
| 720 |
+
"logits/chosen": -0.5384105443954468,
|
| 721 |
+
"logits/rejected": -1.0146501064300537,
|
| 722 |
+
"logps/chosen": -158.76834106445312,
|
| 723 |
+
"logps/rejected": -1430.92041015625,
|
| 724 |
+
"loss": 0.004,
|
| 725 |
+
"rewards/accuracies": 1.0,
|
| 726 |
+
"rewards/chosen": -1.0810316801071167,
|
| 727 |
+
"rewards/margins": 12.214686393737793,
|
| 728 |
+
"rewards/rejected": -13.2957181930542,
|
| 729 |
+
"step": 470
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 3.84,
|
| 733 |
+
"grad_norm": 0.016912085649151143,
|
| 734 |
+
"learning_rate": 7.737816117462752e-06,
|
| 735 |
+
"logits/chosen": -0.6946359872817993,
|
| 736 |
+
"logits/rejected": -1.0958452224731445,
|
| 737 |
+
"logps/chosen": -171.2017364501953,
|
| 738 |
+
"logps/rejected": -1454.8028564453125,
|
| 739 |
+
"loss": 0.0002,
|
| 740 |
+
"rewards/accuracies": 1.0,
|
| 741 |
+
"rewards/chosen": -1.2047702074050903,
|
| 742 |
+
"rewards/margins": 12.34769058227539,
|
| 743 |
+
"rewards/rejected": -13.552459716796875,
|
| 744 |
+
"step": 480
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 3.92,
|
| 748 |
+
"grad_norm": 0.014315964975255087,
|
| 749 |
+
"learning_rate": 7.619929529850397e-06,
|
| 750 |
+
"logits/chosen": -0.6201928853988647,
|
| 751 |
+
"logits/rejected": -1.0985499620437622,
|
| 752 |
+
"logps/chosen": -137.38720703125,
|
| 753 |
+
"logps/rejected": -1331.304443359375,
|
| 754 |
+
"loss": 0.0151,
|
| 755 |
+
"rewards/accuracies": 1.0,
|
| 756 |
+
"rewards/chosen": -0.8758776783943176,
|
| 757 |
+
"rewards/margins": 11.414178848266602,
|
| 758 |
+
"rewards/rejected": -12.290056228637695,
|
| 759 |
+
"step": 490
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 4.0,
|
| 763 |
+
"grad_norm": 0.0008323648197206945,
|
| 764 |
+
"learning_rate": 7.500000000000001e-06,
|
| 765 |
+
"logits/chosen": -0.6446269750595093,
|
| 766 |
+
"logits/rejected": -1.0706923007965088,
|
| 767 |
+
"logps/chosen": -151.62704467773438,
|
| 768 |
+
"logps/rejected": -1406.1658935546875,
|
| 769 |
+
"loss": 0.0001,
|
| 770 |
+
"rewards/accuracies": 1.0,
|
| 771 |
+
"rewards/chosen": -1.0133395195007324,
|
| 772 |
+
"rewards/margins": 12.042734146118164,
|
| 773 |
+
"rewards/rejected": -13.056074142456055,
|
| 774 |
+
"step": 500
|
| 775 |
+
},
|
| 776 |
+
{
|
| 777 |
+
"epoch": 4.08,
|
| 778 |
+
"grad_norm": 0.005439240404948527,
|
| 779 |
+
"learning_rate": 7.378121045351378e-06,
|
| 780 |
+
"logits/chosen": -0.6006187200546265,
|
| 781 |
+
"logits/rejected": -0.9754472970962524,
|
| 782 |
+
"logps/chosen": -175.7959747314453,
|
| 783 |
+
"logps/rejected": -1473.20947265625,
|
| 784 |
+
"loss": 0.0003,
|
| 785 |
+
"rewards/accuracies": 1.0,
|
| 786 |
+
"rewards/chosen": -1.240095853805542,
|
| 787 |
+
"rewards/margins": 12.511373519897461,
|
| 788 |
+
"rewards/rejected": -13.75146770477295,
|
| 789 |
+
"step": 510
|
| 790 |
+
},
|
| 791 |
+
{
|
| 792 |
+
"epoch": 4.16,
|
| 793 |
+
"grad_norm": 0.005118727114675882,
|
| 794 |
+
"learning_rate": 7.254387703447154e-06,
|
| 795 |
+
"logits/chosen": -0.6073562502861023,
|
| 796 |
+
"logits/rejected": -1.0599517822265625,
|
| 797 |
+
"logps/chosen": -158.0718994140625,
|
| 798 |
+
"logps/rejected": -1458.5751953125,
|
| 799 |
+
"loss": 0.0001,
|
| 800 |
+
"rewards/accuracies": 1.0,
|
| 801 |
+
"rewards/chosen": -1.0801844596862793,
|
| 802 |
+
"rewards/margins": 12.502667427062988,
|
| 803 |
+
"rewards/rejected": -13.582852363586426,
|
| 804 |
+
"step": 520
|
| 805 |
+
},
|
| 806 |
+
{
|
| 807 |
+
"epoch": 4.24,
|
| 808 |
+
"grad_norm": 0.0006207044531316051,
|
| 809 |
+
"learning_rate": 7.128896457825364e-06,
|
| 810 |
+
"logits/chosen": -0.6554626226425171,
|
| 811 |
+
"logits/rejected": -1.1788212060928345,
|
| 812 |
+
"logps/chosen": -147.23751831054688,
|
| 813 |
+
"logps/rejected": -1414.2178955078125,
|
| 814 |
+
"loss": 0.0,
|
| 815 |
+
"rewards/accuracies": 1.0,
|
| 816 |
+
"rewards/chosen": -0.9758721590042114,
|
| 817 |
+
"rewards/margins": 12.157670974731445,
|
| 818 |
+
"rewards/rejected": -13.133543014526367,
|
| 819 |
+
"step": 530
|
| 820 |
+
},
|
| 821 |
+
{
|
| 822 |
+
"epoch": 4.32,
|
| 823 |
+
"grad_norm": 0.017948964858749528,
|
| 824 |
+
"learning_rate": 7.0017451627844765e-06,
|
| 825 |
+
"logits/chosen": -0.6513161063194275,
|
| 826 |
+
"logits/rejected": -1.2309751510620117,
|
| 827 |
+
"logps/chosen": -156.2422637939453,
|
| 828 |
+
"logps/rejected": -1466.906494140625,
|
| 829 |
+
"loss": 0.0003,
|
| 830 |
+
"rewards/accuracies": 1.0,
|
| 831 |
+
"rewards/chosen": -1.0636872053146362,
|
| 832 |
+
"rewards/margins": 12.582140922546387,
|
| 833 |
+
"rewards/rejected": -13.645828247070312,
|
| 834 |
+
"step": 540
|
| 835 |
+
},
|
| 836 |
+
{
|
| 837 |
+
"epoch": 4.4,
|
| 838 |
+
"grad_norm": 0.003856347784286379,
|
| 839 |
+
"learning_rate": 6.873032967079562e-06,
|
| 840 |
+
"logits/chosen": -0.6816533803939819,
|
| 841 |
+
"logits/rejected": -1.114074468612671,
|
| 842 |
+
"logps/chosen": -242.92758178710938,
|
| 843 |
+
"logps/rejected": -1675.7265625,
|
| 844 |
+
"loss": 0.0001,
|
| 845 |
+
"rewards/accuracies": 1.0,
|
| 846 |
+
"rewards/chosen": -1.8909059762954712,
|
| 847 |
+
"rewards/margins": 13.899398803710938,
|
| 848 |
+
"rewards/rejected": -15.790304183959961,
|
| 849 |
+
"step": 550
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 4.48,
|
| 853 |
+
"grad_norm": 0.0019510230690865061,
|
| 854 |
+
"learning_rate": 6.7428602366090764e-06,
|
| 855 |
+
"logits/chosen": -0.4731020927429199,
|
| 856 |
+
"logits/rejected": -0.9437645673751831,
|
| 857 |
+
"logps/chosen": -153.3728790283203,
|
| 858 |
+
"logps/rejected": -1450.95458984375,
|
| 859 |
+
"loss": 0.0002,
|
| 860 |
+
"rewards/accuracies": 1.0,
|
| 861 |
+
"rewards/chosen": -1.0217416286468506,
|
| 862 |
+
"rewards/margins": 12.475053787231445,
|
| 863 |
+
"rewards/rejected": -13.496793746948242,
|
| 864 |
+
"step": 560
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 4.5600000000000005,
|
| 868 |
+
"grad_norm": 0.0015858533428631317,
|
| 869 |
+
"learning_rate": 6.611328476152557e-06,
|
| 870 |
+
"logits/chosen": -0.45364946126937866,
|
| 871 |
+
"logits/rejected": -0.953697681427002,
|
| 872 |
+
"logps/chosen": -169.60000610351562,
|
| 873 |
+
"logps/rejected": -1481.096435546875,
|
| 874 |
+
"loss": 0.0001,
|
| 875 |
+
"rewards/accuracies": 1.0,
|
| 876 |
+
"rewards/chosen": -1.1921452283859253,
|
| 877 |
+
"rewards/margins": 12.605159759521484,
|
| 878 |
+
"rewards/rejected": -13.797304153442383,
|
| 879 |
+
"step": 570
|
| 880 |
+
},
|
| 881 |
+
{
|
| 882 |
+
"epoch": 4.64,
|
| 883 |
+
"grad_norm": 0.004170987996850058,
|
| 884 |
+
"learning_rate": 6.4785402502202345e-06,
|
| 885 |
+
"logits/chosen": -0.3475385904312134,
|
| 886 |
+
"logits/rejected": -0.9073439836502075,
|
| 887 |
+
"logps/chosen": -140.90237426757812,
|
| 888 |
+
"logps/rejected": -1381.1484375,
|
| 889 |
+
"loss": 0.0,
|
| 890 |
+
"rewards/accuracies": 1.0,
|
| 891 |
+
"rewards/chosen": -0.921575665473938,
|
| 892 |
+
"rewards/margins": 11.858769416809082,
|
| 893 |
+
"rewards/rejected": -12.78034496307373,
|
| 894 |
+
"step": 580
|
| 895 |
+
},
|
| 896 |
+
{
|
| 897 |
+
"epoch": 4.72,
|
| 898 |
+
"grad_norm": 0.0013538463974769534,
|
| 899 |
+
"learning_rate": 6.344599103076329e-06,
|
| 900 |
+
"logits/chosen": -0.3682401776313782,
|
| 901 |
+
"logits/rejected": -0.9007229804992676,
|
| 902 |
+
"logps/chosen": -147.44747924804688,
|
| 903 |
+
"logps/rejected": -1403.5577392578125,
|
| 904 |
+
"loss": 0.0,
|
| 905 |
+
"rewards/accuracies": 1.0,
|
| 906 |
+
"rewards/chosen": -0.9861438870429993,
|
| 907 |
+
"rewards/margins": 12.035198211669922,
|
| 908 |
+
"rewards/rejected": -13.021342277526855,
|
| 909 |
+
"step": 590
|
| 910 |
+
},
|
| 911 |
+
{
|
| 912 |
+
"epoch": 4.8,
|
| 913 |
+
"grad_norm": 0.022561866804519767,
|
| 914 |
+
"learning_rate": 6.209609477998339e-06,
|
| 915 |
+
"logits/chosen": -0.4450947642326355,
|
| 916 |
+
"logits/rejected": -0.9426549077033997,
|
| 917 |
+
"logps/chosen": -155.5927276611328,
|
| 918 |
+
"logps/rejected": -1427.051513671875,
|
| 919 |
+
"loss": 0.0002,
|
| 920 |
+
"rewards/accuracies": 1.0,
|
| 921 |
+
"rewards/chosen": -1.061085820198059,
|
| 922 |
+
"rewards/margins": 12.196368217468262,
|
| 923 |
+
"rewards/rejected": -13.257455825805664,
|
| 924 |
+
"step": 600
|
| 925 |
+
},
|
| 926 |
+
{
|
| 927 |
+
"epoch": 4.88,
|
| 928 |
+
"grad_norm": 0.012428458617134987,
|
| 929 |
+
"learning_rate": 6.073676635835317e-06,
|
| 930 |
+
"logits/chosen": -0.47192803025245667,
|
| 931 |
+
"logits/rejected": -0.9936426281929016,
|
| 932 |
+
"logps/chosen": -163.2184295654297,
|
| 933 |
+
"logps/rejected": -1448.4224853515625,
|
| 934 |
+
"loss": 0.0001,
|
| 935 |
+
"rewards/accuracies": 1.0,
|
| 936 |
+
"rewards/chosen": -1.1418524980545044,
|
| 937 |
+
"rewards/margins": 12.326138496398926,
|
| 938 |
+
"rewards/rejected": -13.467989921569824,
|
| 939 |
+
"step": 610
|
| 940 |
+
},
|
| 941 |
+
{
|
| 942 |
+
"epoch": 4.96,
|
| 943 |
+
"grad_norm": 0.00027433530717749753,
|
| 944 |
+
"learning_rate": 5.936906572928625e-06,
|
| 945 |
+
"logits/chosen": -0.4440648555755615,
|
| 946 |
+
"logits/rejected": -0.9577704668045044,
|
| 947 |
+
"logps/chosen": -123.54743957519531,
|
| 948 |
+
"logps/rejected": -1388.123779296875,
|
| 949 |
+
"loss": 0.0001,
|
| 950 |
+
"rewards/accuracies": 1.0,
|
| 951 |
+
"rewards/chosen": -0.7459378838539124,
|
| 952 |
+
"rewards/margins": 12.11369514465332,
|
| 953 |
+
"rewards/rejected": -12.85963249206543,
|
| 954 |
+
"step": 620
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 5.04,
|
| 958 |
+
"grad_norm": 0.0003376506922564907,
|
| 959 |
+
"learning_rate": 5.799405938459175e-06,
|
| 960 |
+
"logits/chosen": -0.449303537607193,
|
| 961 |
+
"logits/rejected": -0.9946249723434448,
|
| 962 |
+
"logps/chosen": -126.98262023925781,
|
| 963 |
+
"logps/rejected": -1411.4127197265625,
|
| 964 |
+
"loss": 0.0,
|
| 965 |
+
"rewards/accuracies": 1.0,
|
| 966 |
+
"rewards/chosen": -0.7872297763824463,
|
| 967 |
+
"rewards/margins": 12.30899715423584,
|
| 968 |
+
"rewards/rejected": -13.096226692199707,
|
| 969 |
+
"step": 630
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 5.12,
|
| 973 |
+
"grad_norm": 0.005188794996734725,
|
| 974 |
+
"learning_rate": 5.661281951285613e-06,
|
| 975 |
+
"logits/chosen": -0.39260971546173096,
|
| 976 |
+
"logits/rejected": -0.8862244486808777,
|
| 977 |
+
"logps/chosen": -139.25743103027344,
|
| 978 |
+
"logps/rejected": -1483.836669921875,
|
| 979 |
+
"loss": 0.0,
|
| 980 |
+
"rewards/accuracies": 1.0,
|
| 981 |
+
"rewards/chosen": -0.8930659294128418,
|
| 982 |
+
"rewards/margins": 12.942354202270508,
|
| 983 |
+
"rewards/rejected": -13.835420608520508,
|
| 984 |
+
"step": 640
|
| 985 |
+
},
|
| 986 |
+
{
|
| 987 |
+
"epoch": 5.2,
|
| 988 |
+
"grad_norm": 0.0005439385645287033,
|
| 989 |
+
"learning_rate": 5.522642316338268e-06,
|
| 990 |
+
"logits/chosen": -0.42691946029663086,
|
| 991 |
+
"logits/rejected": -0.9130016565322876,
|
| 992 |
+
"logps/chosen": -146.08859252929688,
|
| 993 |
+
"logps/rejected": -1495.5863037109375,
|
| 994 |
+
"loss": 0.0,
|
| 995 |
+
"rewards/accuracies": 1.0,
|
| 996 |
+
"rewards/chosen": -0.9477936625480652,
|
| 997 |
+
"rewards/margins": 13.005441665649414,
|
| 998 |
+
"rewards/rejected": -13.953234672546387,
|
| 999 |
+
"step": 650
|
| 1000 |
+
},
|
| 1001 |
+
{
|
| 1002 |
+
"epoch": 5.28,
|
| 1003 |
+
"grad_norm": 0.000583833541599897,
|
| 1004 |
+
"learning_rate": 5.383595140634093e-06,
|
| 1005 |
+
"logits/chosen": -0.34127289056777954,
|
| 1006 |
+
"logits/rejected": -0.829853892326355,
|
| 1007 |
+
"logps/chosen": -133.29141235351562,
|
| 1008 |
+
"logps/rejected": -1512.446044921875,
|
| 1009 |
+
"loss": 0.0,
|
| 1010 |
+
"rewards/accuracies": 1.0,
|
| 1011 |
+
"rewards/chosen": -0.8321868777275085,
|
| 1012 |
+
"rewards/margins": 13.277082443237305,
|
| 1013 |
+
"rewards/rejected": -14.109270095825195,
|
| 1014 |
+
"step": 660
|
| 1015 |
+
},
|
| 1016 |
+
{
|
| 1017 |
+
"epoch": 5.36,
|
| 1018 |
+
"grad_norm": 0.005030134807238761,
|
| 1019 |
+
"learning_rate": 5.244248848978067e-06,
|
| 1020 |
+
"logits/chosen": -0.3165265917778015,
|
| 1021 |
+
"logits/rejected": -0.9533147811889648,
|
| 1022 |
+
"logps/chosen": -104.08113098144531,
|
| 1023 |
+
"logps/rejected": -1304.6605224609375,
|
| 1024 |
+
"loss": 0.0001,
|
| 1025 |
+
"rewards/accuracies": 1.0,
|
| 1026 |
+
"rewards/chosen": -0.5602812170982361,
|
| 1027 |
+
"rewards/margins": 11.446382522583008,
|
| 1028 |
+
"rewards/rejected": -12.00666332244873,
|
| 1029 |
+
"step": 670
|
| 1030 |
+
},
|
| 1031 |
+
{
|
| 1032 |
+
"epoch": 5.44,
|
| 1033 |
+
"grad_norm": 0.003070807279465456,
|
| 1034 |
+
"learning_rate": 5.1047120994167855e-06,
|
| 1035 |
+
"logits/chosen": -0.36942166090011597,
|
| 1036 |
+
"logits/rejected": -0.9378865957260132,
|
| 1037 |
+
"logps/chosen": -122.25132751464844,
|
| 1038 |
+
"logps/rejected": -1403.7523193359375,
|
| 1039 |
+
"loss": 0.0,
|
| 1040 |
+
"rewards/accuracies": 1.0,
|
| 1041 |
+
"rewards/chosen": -0.7320325970649719,
|
| 1042 |
+
"rewards/margins": 12.275636672973633,
|
| 1043 |
+
"rewards/rejected": -13.007669448852539,
|
| 1044 |
+
"step": 680
|
| 1045 |
+
},
|
| 1046 |
+
{
|
| 1047 |
+
"epoch": 5.52,
|
| 1048 |
+
"grad_norm": 0.0008123370720346338,
|
| 1049 |
+
"learning_rate": 4.965093698510192e-06,
|
| 1050 |
+
"logits/chosen": -0.3969612419605255,
|
| 1051 |
+
"logits/rejected": -0.9130090475082397,
|
| 1052 |
+
"logps/chosen": -164.13131713867188,
|
| 1053 |
+
"logps/rejected": -1582.1212158203125,
|
| 1054 |
+
"loss": 0.0,
|
| 1055 |
+
"rewards/accuracies": 1.0,
|
| 1056 |
+
"rewards/chosen": -1.1271440982818604,
|
| 1057 |
+
"rewards/margins": 13.687349319458008,
|
| 1058 |
+
"rewards/rejected": -14.814491271972656,
|
| 1059 |
+
"step": 690
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 5.6,
|
| 1063 |
+
"grad_norm": 0.0008344763947590128,
|
| 1064 |
+
"learning_rate": 4.825502516487497e-06,
|
| 1065 |
+
"logits/chosen": -0.3772839903831482,
|
| 1066 |
+
"logits/rejected": -0.8978067636489868,
|
| 1067 |
+
"logps/chosen": -145.3348846435547,
|
| 1068 |
+
"logps/rejected": -1506.9521484375,
|
| 1069 |
+
"loss": 0.0001,
|
| 1070 |
+
"rewards/accuracies": 1.0,
|
| 1071 |
+
"rewards/chosen": -0.9458595514297485,
|
| 1072 |
+
"rewards/margins": 13.112927436828613,
|
| 1073 |
+
"rewards/rejected": -14.05878734588623,
|
| 1074 |
+
"step": 700
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 5.68,
|
| 1078 |
+
"grad_norm": 0.0007531302584682961,
|
| 1079 |
+
"learning_rate": 4.686047402353433e-06,
|
| 1080 |
+
"logits/chosen": -0.45077744126319885,
|
| 1081 |
+
"logits/rejected": -1.0165283679962158,
|
| 1082 |
+
"logps/chosen": -132.34994506835938,
|
| 1083 |
+
"logps/rejected": -1427.802734375,
|
| 1084 |
+
"loss": 0.0,
|
| 1085 |
+
"rewards/accuracies": 1.0,
|
| 1086 |
+
"rewards/chosen": -0.8286944627761841,
|
| 1087 |
+
"rewards/margins": 12.429733276367188,
|
| 1088 |
+
"rewards/rejected": -13.258428573608398,
|
| 1089 |
+
"step": 710
|
| 1090 |
+
},
|
| 1091 |
+
{
|
| 1092 |
+
"epoch": 5.76,
|
| 1093 |
+
"grad_norm": 0.0003263470597252238,
|
| 1094 |
+
"learning_rate": 4.546837099011101e-06,
|
| 1095 |
+
"logits/chosen": -0.43557509779930115,
|
| 1096 |
+
"logits/rejected": -0.9456827044487,
|
| 1097 |
+
"logps/chosen": -133.05575561523438,
|
| 1098 |
+
"logps/rejected": -1460.742431640625,
|
| 1099 |
+
"loss": 0.0,
|
| 1100 |
+
"rewards/accuracies": 1.0,
|
| 1101 |
+
"rewards/chosen": -0.832891583442688,
|
| 1102 |
+
"rewards/margins": 12.750478744506836,
|
| 1103 |
+
"rewards/rejected": -13.58337116241455,
|
| 1104 |
+
"step": 720
|
| 1105 |
+
},
|
| 1106 |
+
{
|
| 1107 |
+
"epoch": 5.84,
|
| 1108 |
+
"grad_norm": 0.0003894890886611588,
|
| 1109 |
+
"learning_rate": 4.4079801584674955e-06,
|
| 1110 |
+
"logits/chosen": -0.39473429322242737,
|
| 1111 |
+
"logits/rejected": -0.9540184140205383,
|
| 1112 |
+
"logps/chosen": -120.5350341796875,
|
| 1113 |
+
"logps/rejected": -1388.319580078125,
|
| 1114 |
+
"loss": 0.0,
|
| 1115 |
+
"rewards/accuracies": 1.0,
|
| 1116 |
+
"rewards/chosen": -0.7102386951446533,
|
| 1117 |
+
"rewards/margins": 12.142650604248047,
|
| 1118 |
+
"rewards/rejected": -12.852888107299805,
|
| 1119 |
+
"step": 730
|
| 1120 |
+
},
|
| 1121 |
+
{
|
| 1122 |
+
"epoch": 5.92,
|
| 1123 |
+
"grad_norm": 0.00021687658056609183,
|
| 1124 |
+
"learning_rate": 4.269584857187942e-06,
|
| 1125 |
+
"logits/chosen": -0.41277170181274414,
|
| 1126 |
+
"logits/rejected": -0.8866747617721558,
|
| 1127 |
+
"logps/chosen": -169.7379150390625,
|
| 1128 |
+
"logps/rejected": -1618.2318115234375,
|
| 1129 |
+
"loss": 0.0,
|
| 1130 |
+
"rewards/accuracies": 1.0,
|
| 1131 |
+
"rewards/chosen": -1.1842690706253052,
|
| 1132 |
+
"rewards/margins": 13.998774528503418,
|
| 1133 |
+
"rewards/rejected": -15.183042526245117,
|
| 1134 |
+
"step": 740
|
| 1135 |
+
},
|
| 1136 |
+
{
|
| 1137 |
+
"epoch": 6.0,
|
| 1138 |
+
"grad_norm": 0.00017363950347948243,
|
| 1139 |
+
"learning_rate": 4.131759111665349e-06,
|
| 1140 |
+
"logits/chosen": -0.380937397480011,
|
| 1141 |
+
"logits/rejected": -0.941672146320343,
|
| 1142 |
+
"logps/chosen": -138.58726501464844,
|
| 1143 |
+
"logps/rejected": -1467.87109375,
|
| 1144 |
+
"loss": 0.0001,
|
| 1145 |
+
"rewards/accuracies": 1.0,
|
| 1146 |
+
"rewards/chosen": -0.8956032991409302,
|
| 1147 |
+
"rewards/margins": 12.76499080657959,
|
| 1148 |
+
"rewards/rejected": -13.66059398651123,
|
| 1149 |
+
"step": 750
|
| 1150 |
+
},
|
| 1151 |
+
{
|
| 1152 |
+
"epoch": 6.08,
|
| 1153 |
+
"grad_norm": 0.0002869387785848348,
|
| 1154 |
+
"learning_rate": 3.994610394270178e-06,
|
| 1155 |
+
"logits/chosen": -0.3407558500766754,
|
| 1156 |
+
"logits/rejected": -0.9460574984550476,
|
| 1157 |
+
"logps/chosen": -120.59251403808594,
|
| 1158 |
+
"logps/rejected": -1365.903564453125,
|
| 1159 |
+
"loss": 0.0,
|
| 1160 |
+
"rewards/accuracies": 1.0,
|
| 1161 |
+
"rewards/chosen": -0.7153140306472778,
|
| 1162 |
+
"rewards/margins": 11.916279792785645,
|
| 1163 |
+
"rewards/rejected": -12.631593704223633,
|
| 1164 |
+
"step": 760
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 6.16,
|
| 1168 |
+
"grad_norm": 0.00022830076793507427,
|
| 1169 |
+
"learning_rate": 3.8582456494467214e-06,
|
| 1170 |
+
"logits/chosen": -0.3290683329105377,
|
| 1171 |
+
"logits/rejected": -0.8766037225723267,
|
| 1172 |
+
"logps/chosen": -137.58761596679688,
|
| 1173 |
+
"logps/rejected": -1491.4849853515625,
|
| 1174 |
+
"loss": 0.0,
|
| 1175 |
+
"rewards/accuracies": 1.0,
|
| 1176 |
+
"rewards/chosen": -0.8715416789054871,
|
| 1177 |
+
"rewards/margins": 13.01722240447998,
|
| 1178 |
+
"rewards/rejected": -13.888763427734375,
|
| 1179 |
+
"step": 770
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 6.24,
|
| 1183 |
+
"grad_norm": 0.00018905031588614676,
|
| 1184 |
+
"learning_rate": 3.7227712103210485e-06,
|
| 1185 |
+
"logits/chosen": -0.4009069502353668,
|
| 1186 |
+
"logits/rejected": -0.8301585912704468,
|
| 1187 |
+
"logps/chosen": -168.92832946777344,
|
| 1188 |
+
"logps/rejected": -1632.318115234375,
|
| 1189 |
+
"loss": 0.0001,
|
| 1190 |
+
"rewards/accuracies": 1.0,
|
| 1191 |
+
"rewards/chosen": -1.167081594467163,
|
| 1192 |
+
"rewards/margins": 14.170913696289062,
|
| 1193 |
+
"rewards/rejected": -15.337995529174805,
|
| 1194 |
+
"step": 780
|
| 1195 |
+
},
|
| 1196 |
+
{
|
| 1197 |
+
"epoch": 6.32,
|
| 1198 |
+
"grad_norm": 0.00044615726295310295,
|
| 1199 |
+
"learning_rate": 3.5882927157856175e-06,
|
| 1200 |
+
"logits/chosen": -0.3509620130062103,
|
| 1201 |
+
"logits/rejected": -0.8593319654464722,
|
| 1202 |
+
"logps/chosen": -129.48745727539062,
|
| 1203 |
+
"logps/rejected": -1498.2364501953125,
|
| 1204 |
+
"loss": 0.0,
|
| 1205 |
+
"rewards/accuracies": 1.0,
|
| 1206 |
+
"rewards/chosen": -0.7902848720550537,
|
| 1207 |
+
"rewards/margins": 13.172172546386719,
|
| 1208 |
+
"rewards/rejected": -13.962457656860352,
|
| 1209 |
+
"step": 790
|
| 1210 |
+
},
|
| 1211 |
+
{
|
| 1212 |
+
"epoch": 6.4,
|
| 1213 |
+
"grad_norm": 0.00017684322493448067,
|
| 1214 |
+
"learning_rate": 3.4549150281252635e-06,
|
| 1215 |
+
"logits/chosen": -0.4214898645877838,
|
| 1216 |
+
"logits/rejected": -0.8738347887992859,
|
| 1217 |
+
"logps/chosen": -171.40628051757812,
|
| 1218 |
+
"logps/rejected": -1577.4674072265625,
|
| 1219 |
+
"loss": 0.0,
|
| 1220 |
+
"rewards/accuracies": 1.0,
|
| 1221 |
+
"rewards/chosen": -1.1943633556365967,
|
| 1222 |
+
"rewards/margins": 13.582929611206055,
|
| 1223 |
+
"rewards/rejected": -14.77729320526123,
|
| 1224 |
+
"step": 800
|
| 1225 |
+
},
|
| 1226 |
+
{
|
| 1227 |
+
"epoch": 6.48,
|
| 1228 |
+
"grad_norm": 0.00035092578034022254,
|
| 1229 |
+
"learning_rate": 3.322742151248726e-06,
|
| 1230 |
+
"logits/chosen": -0.33817845582962036,
|
| 1231 |
+
"logits/rejected": -0.8328613042831421,
|
| 1232 |
+
"logps/chosen": -131.61691284179688,
|
| 1233 |
+
"logps/rejected": -1441.378662109375,
|
| 1234 |
+
"loss": 0.0,
|
| 1235 |
+
"rewards/accuracies": 1.0,
|
| 1236 |
+
"rewards/chosen": -0.8236126899719238,
|
| 1237 |
+
"rewards/margins": 12.581751823425293,
|
| 1238 |
+
"rewards/rejected": -13.405363082885742,
|
| 1239 |
+
"step": 810
|
| 1240 |
+
},
|
| 1241 |
+
{
|
| 1242 |
+
"epoch": 6.5600000000000005,
|
| 1243 |
+
"grad_norm": 0.0016578838421130056,
|
| 1244 |
+
"learning_rate": 3.1918771495895395e-06,
|
| 1245 |
+
"logits/chosen": -0.3588394224643707,
|
| 1246 |
+
"logits/rejected": -0.890457034111023,
|
| 1247 |
+
"logps/chosen": -147.90191650390625,
|
| 1248 |
+
"logps/rejected": -1499.8687744140625,
|
| 1249 |
+
"loss": 0.0,
|
| 1250 |
+
"rewards/accuracies": 1.0,
|
| 1251 |
+
"rewards/chosen": -0.9654636383056641,
|
| 1252 |
+
"rewards/margins": 13.020975112915039,
|
| 1253 |
+
"rewards/rejected": -13.986437797546387,
|
| 1254 |
+
"step": 820
|
| 1255 |
+
},
|
| 1256 |
+
{
|
| 1257 |
+
"epoch": 6.64,
|
| 1258 |
+
"grad_norm": 0.0001813290316556118,
|
| 1259 |
+
"learning_rate": 3.0624220677394854e-06,
|
| 1260 |
+
"logits/chosen": -0.4224972724914551,
|
| 1261 |
+
"logits/rejected": -0.9041132926940918,
|
| 1262 |
+
"logps/chosen": -146.0318145751953,
|
| 1263 |
+
"logps/rejected": -1597.461669921875,
|
| 1264 |
+
"loss": 0.0,
|
| 1265 |
+
"rewards/accuracies": 1.0,
|
| 1266 |
+
"rewards/chosen": -0.9522792100906372,
|
| 1267 |
+
"rewards/margins": 14.007960319519043,
|
| 1268 |
+
"rewards/rejected": -14.960238456726074,
|
| 1269 |
+
"step": 830
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 6.72,
|
| 1273 |
+
"grad_norm": 0.007917608250023045,
|
| 1274 |
+
"learning_rate": 2.934477850877292e-06,
|
| 1275 |
+
"logits/chosen": -0.36725106835365295,
|
| 1276 |
+
"logits/rejected": -0.8673862218856812,
|
| 1277 |
+
"logps/chosen": -137.09661865234375,
|
| 1278 |
+
"logps/rejected": -1493.104248046875,
|
| 1279 |
+
"loss": 0.0,
|
| 1280 |
+
"rewards/accuracies": 1.0,
|
| 1281 |
+
"rewards/chosen": -0.8727433085441589,
|
| 1282 |
+
"rewards/margins": 13.050328254699707,
|
| 1283 |
+
"rewards/rejected": -13.923070907592773,
|
| 1284 |
+
"step": 840
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 6.8,
|
| 1288 |
+
"grad_norm": 0.00018574879436063482,
|
| 1289 |
+
"learning_rate": 2.8081442660546126e-06,
|
| 1290 |
+
"logits/chosen": -0.3642883896827698,
|
| 1291 |
+
"logits/rejected": -0.9518113136291504,
|
| 1292 |
+
"logps/chosen": -126.36039733886719,
|
| 1293 |
+
"logps/rejected": -1447.3575439453125,
|
| 1294 |
+
"loss": 0.0,
|
| 1295 |
+
"rewards/accuracies": 1.0,
|
| 1296 |
+
"rewards/chosen": -0.7840255498886108,
|
| 1297 |
+
"rewards/margins": 12.662456512451172,
|
| 1298 |
+
"rewards/rejected": -13.446481704711914,
|
| 1299 |
+
"step": 850
|
| 1300 |
+
},
|
| 1301 |
+
{
|
| 1302 |
+
"epoch": 6.88,
|
| 1303 |
+
"grad_norm": 0.00015909977255206564,
|
| 1304 |
+
"learning_rate": 2.683519824400693e-06,
|
| 1305 |
+
"logits/chosen": -0.292142391204834,
|
| 1306 |
+
"logits/rejected": -0.8652153015136719,
|
| 1307 |
+
"logps/chosen": -124.55448150634766,
|
| 1308 |
+
"logps/rejected": -1406.6923828125,
|
| 1309 |
+
"loss": 0.0,
|
| 1310 |
+
"rewards/accuracies": 1.0,
|
| 1311 |
+
"rewards/chosen": -0.7584154605865479,
|
| 1312 |
+
"rewards/margins": 12.294742584228516,
|
| 1313 |
+
"rewards/rejected": -13.0531587600708,
|
| 1314 |
+
"step": 860
|
| 1315 |
+
},
|
| 1316 |
+
{
|
| 1317 |
+
"epoch": 6.96,
|
| 1318 |
+
"grad_norm": 0.0001625250801281625,
|
| 1319 |
+
"learning_rate": 2.560701704306336e-06,
|
| 1320 |
+
"logits/chosen": -0.34566569328308105,
|
| 1321 |
+
"logits/rejected": -0.8776302337646484,
|
| 1322 |
+
"logps/chosen": -149.78610229492188,
|
| 1323 |
+
"logps/rejected": -1521.619384765625,
|
| 1324 |
+
"loss": 0.0,
|
| 1325 |
+
"rewards/accuracies": 1.0,
|
| 1326 |
+
"rewards/chosen": -0.9827316999435425,
|
| 1327 |
+
"rewards/margins": 13.225728988647461,
|
| 1328 |
+
"rewards/rejected": -14.208460807800293,
|
| 1329 |
+
"step": 870
|
| 1330 |
+
},
|
| 1331 |
+
{
|
| 1332 |
+
"epoch": 7.04,
|
| 1333 |
+
"grad_norm": 0.0001369508246975124,
|
| 1334 |
+
"learning_rate": 2.4397856756471435e-06,
|
| 1335 |
+
"logits/chosen": -0.32312315702438354,
|
| 1336 |
+
"logits/rejected": -0.8410239219665527,
|
| 1337 |
+
"logps/chosen": -129.86459350585938,
|
| 1338 |
+
"logps/rejected": -1484.922607421875,
|
| 1339 |
+
"loss": 0.0,
|
| 1340 |
+
"rewards/accuracies": 1.0,
|
| 1341 |
+
"rewards/chosen": -0.8025870323181152,
|
| 1342 |
+
"rewards/margins": 13.027560234069824,
|
| 1343 |
+
"rewards/rejected": -13.830146789550781,
|
| 1344 |
+
"step": 880
|
| 1345 |
+
},
|
| 1346 |
+
{
|
| 1347 |
+
"epoch": 7.12,
|
| 1348 |
+
"grad_norm": 0.00018364985576119344,
|
| 1349 |
+
"learning_rate": 2.320866025105016e-06,
|
| 1350 |
+
"logits/chosen": -0.3720766007900238,
|
| 1351 |
+
"logits/rejected": -0.8578876256942749,
|
| 1352 |
+
"logps/chosen": -173.00662231445312,
|
| 1353 |
+
"logps/rejected": -1543.7581787109375,
|
| 1354 |
+
"loss": 0.0,
|
| 1355 |
+
"rewards/accuracies": 1.0,
|
| 1356 |
+
"rewards/chosen": -1.2114688158035278,
|
| 1357 |
+
"rewards/margins": 13.233538627624512,
|
| 1358 |
+
"rewards/rejected": -14.445009231567383,
|
| 1359 |
+
"step": 890
|
| 1360 |
+
},
|
| 1361 |
+
{
|
| 1362 |
+
"epoch": 7.2,
|
| 1363 |
+
"grad_norm": 0.0008614496535749721,
|
| 1364 |
+
"learning_rate": 2.204035482646267e-06,
|
| 1365 |
+
"logits/chosen": -0.34498947858810425,
|
| 1366 |
+
"logits/rejected": -0.8200048208236694,
|
| 1367 |
+
"logps/chosen": -159.82220458984375,
|
| 1368 |
+
"logps/rejected": -1630.5267333984375,
|
| 1369 |
+
"loss": 0.0,
|
| 1370 |
+
"rewards/accuracies": 1.0,
|
| 1371 |
+
"rewards/chosen": -1.0811651945114136,
|
| 1372 |
+
"rewards/margins": 14.221280097961426,
|
| 1373 |
+
"rewards/rejected": -15.302446365356445,
|
| 1374 |
+
"step": 900
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 7.28,
|
| 1378 |
+
"grad_norm": 0.00012796605758885017,
|
| 1379 |
+
"learning_rate": 2.0893851492135536e-06,
|
| 1380 |
+
"logits/chosen": -0.4566912055015564,
|
| 1381 |
+
"logits/rejected": -0.9449372291564941,
|
| 1382 |
+
"logps/chosen": -158.0172119140625,
|
| 1383 |
+
"logps/rejected": -1570.6591796875,
|
| 1384 |
+
"loss": 0.0,
|
| 1385 |
+
"rewards/accuracies": 1.0,
|
| 1386 |
+
"rewards/chosen": -1.0653969049453735,
|
| 1387 |
+
"rewards/margins": 13.639493942260742,
|
| 1388 |
+
"rewards/rejected": -14.7048921585083,
|
| 1389 |
+
"step": 910
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 7.36,
|
| 1393 |
+
"grad_norm": 0.002657232072751586,
|
| 1394 |
+
"learning_rate": 1.977004425688126e-06,
|
| 1395 |
+
"logits/chosen": -0.4158572554588318,
|
| 1396 |
+
"logits/rejected": -0.9982155561447144,
|
| 1397 |
+
"logps/chosen": -130.85671997070312,
|
| 1398 |
+
"logps/rejected": -1492.7935791015625,
|
| 1399 |
+
"loss": 0.0,
|
| 1400 |
+
"rewards/accuracies": 1.0,
|
| 1401 |
+
"rewards/chosen": -0.8182401657104492,
|
| 1402 |
+
"rewards/margins": 13.083892822265625,
|
| 1403 |
+
"rewards/rejected": -13.902132987976074,
|
| 1404 |
+
"step": 920
|
| 1405 |
+
},
|
| 1406 |
+
{
|
| 1407 |
+
"epoch": 7.44,
|
| 1408 |
+
"grad_norm": 0.00015579327036541104,
|
| 1409 |
+
"learning_rate": 1.8669809431776991e-06,
|
| 1410 |
+
"logits/chosen": -0.40051618218421936,
|
| 1411 |
+
"logits/rejected": -0.930172324180603,
|
| 1412 |
+
"logps/chosen": -153.02651977539062,
|
| 1413 |
+
"logps/rejected": -1536.75732421875,
|
| 1414 |
+
"loss": 0.0,
|
| 1415 |
+
"rewards/accuracies": 1.0,
|
| 1416 |
+
"rewards/chosen": -1.0270575284957886,
|
| 1417 |
+
"rewards/margins": 13.345794677734375,
|
| 1418 |
+
"rewards/rejected": -14.372851371765137,
|
| 1419 |
+
"step": 930
|
| 1420 |
+
},
|
| 1421 |
+
{
|
| 1422 |
+
"epoch": 7.52,
|
| 1423 |
+
"grad_norm": 0.006509186534878427,
|
| 1424 |
+
"learning_rate": 1.7594004946843458e-06,
|
| 1425 |
+
"logits/chosen": -0.3410387635231018,
|
| 1426 |
+
"logits/rejected": -0.8968143463134766,
|
| 1427 |
+
"logps/chosen": -131.83535766601562,
|
| 1428 |
+
"logps/rejected": -1495.5213623046875,
|
| 1429 |
+
"loss": 0.0,
|
| 1430 |
+
"rewards/accuracies": 1.0,
|
| 1431 |
+
"rewards/chosen": -0.8293743133544922,
|
| 1432 |
+
"rewards/margins": 13.11143970489502,
|
| 1433 |
+
"rewards/rejected": -13.940814018249512,
|
| 1434 |
+
"step": 940
|
| 1435 |
+
},
|
| 1436 |
+
{
|
| 1437 |
+
"epoch": 7.6,
|
| 1438 |
+
"grad_norm": 0.0001596335948222842,
|
| 1439 |
+
"learning_rate": 1.6543469682057105e-06,
|
| 1440 |
+
"logits/chosen": -0.3258531987667084,
|
| 1441 |
+
"logits/rejected": -0.8879348635673523,
|
| 1442 |
+
"logps/chosen": -116.85892486572266,
|
| 1443 |
+
"logps/rejected": -1435.84912109375,
|
| 1444 |
+
"loss": 0.0,
|
| 1445 |
+
"rewards/accuracies": 1.0,
|
| 1446 |
+
"rewards/chosen": -0.6833716034889221,
|
| 1447 |
+
"rewards/margins": 12.654440879821777,
|
| 1448 |
+
"rewards/rejected": -13.337811470031738,
|
| 1449 |
+
"step": 950
|
| 1450 |
+
},
|
| 1451 |
+
{
|
| 1452 |
+
"epoch": 7.68,
|
| 1453 |
+
"grad_norm": 0.00016669337554250593,
|
| 1454 |
+
"learning_rate": 1.551902281321651e-06,
|
| 1455 |
+
"logits/chosen": -0.3701092600822449,
|
| 1456 |
+
"logits/rejected": -0.9417698979377747,
|
| 1457 |
+
"logps/chosen": -125.5888671875,
|
| 1458 |
+
"logps/rejected": -1435.673095703125,
|
| 1459 |
+
"loss": 0.0,
|
| 1460 |
+
"rewards/accuracies": 1.0,
|
| 1461 |
+
"rewards/chosen": -0.7691789865493774,
|
| 1462 |
+
"rewards/margins": 12.566195487976074,
|
| 1463 |
+
"rewards/rejected": -13.33537483215332,
|
| 1464 |
+
"step": 960
|
| 1465 |
+
},
|
| 1466 |
+
{
|
| 1467 |
+
"epoch": 7.76,
|
| 1468 |
+
"grad_norm": 0.00013859610902316376,
|
| 1469 |
+
"learning_rate": 1.4521463173173966e-06,
|
| 1470 |
+
"logits/chosen": -0.3638702929019928,
|
| 1471 |
+
"logits/rejected": -0.8297930955886841,
|
| 1472 |
+
"logps/chosen": -154.0557403564453,
|
| 1473 |
+
"logps/rejected": -1617.8785400390625,
|
| 1474 |
+
"loss": 0.0,
|
| 1475 |
+
"rewards/accuracies": 1.0,
|
| 1476 |
+
"rewards/chosen": -1.033390998840332,
|
| 1477 |
+
"rewards/margins": 14.146600723266602,
|
| 1478 |
+
"rewards/rejected": -15.179990768432617,
|
| 1479 |
+
"step": 970
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 7.84,
|
| 1483 |
+
"grad_norm": 0.00012978293834875766,
|
| 1484 |
+
"learning_rate": 1.3551568628929434e-06,
|
| 1485 |
+
"logits/chosen": -0.3917458951473236,
|
| 1486 |
+
"logits/rejected": -0.9406350255012512,
|
| 1487 |
+
"logps/chosen": -130.6827392578125,
|
| 1488 |
+
"logps/rejected": -1494.780029296875,
|
| 1489 |
+
"loss": 0.0,
|
| 1490 |
+
"rewards/accuracies": 1.0,
|
| 1491 |
+
"rewards/chosen": -0.8124713897705078,
|
| 1492 |
+
"rewards/margins": 13.109392166137695,
|
| 1493 |
+
"rewards/rejected": -13.921862602233887,
|
| 1494 |
+
"step": 980
|
| 1495 |
+
},
|
| 1496 |
+
{
|
| 1497 |
+
"epoch": 7.92,
|
| 1498 |
+
"grad_norm": 0.00016484979693604264,
|
| 1499 |
+
"learning_rate": 1.2610095475073415e-06,
|
| 1500 |
+
"logits/chosen": -0.39324721693992615,
|
| 1501 |
+
"logits/rejected": -0.9108613729476929,
|
| 1502 |
+
"logps/chosen": -125.08223724365234,
|
| 1503 |
+
"logps/rejected": -1461.036865234375,
|
| 1504 |
+
"loss": 0.0,
|
| 1505 |
+
"rewards/accuracies": 1.0,
|
| 1506 |
+
"rewards/chosen": -0.75953209400177,
|
| 1507 |
+
"rewards/margins": 12.83247184753418,
|
| 1508 |
+
"rewards/rejected": -13.592002868652344,
|
| 1509 |
+
"step": 990
|
| 1510 |
+
},
|
| 1511 |
+
{
|
| 1512 |
+
"epoch": 8.0,
|
| 1513 |
+
"grad_norm": 0.00016513337681309213,
|
| 1514 |
+
"learning_rate": 1.1697777844051105e-06,
|
| 1515 |
+
"logits/chosen": -0.47117894887924194,
|
| 1516 |
+
"logits/rejected": -0.973685085773468,
|
| 1517 |
+
"logps/chosen": -137.49221801757812,
|
| 1518 |
+
"logps/rejected": -1485.5107421875,
|
| 1519 |
+
"loss": 0.0,
|
| 1520 |
+
"rewards/accuracies": 1.0,
|
| 1521 |
+
"rewards/chosen": -0.8799915313720703,
|
| 1522 |
+
"rewards/margins": 12.966386795043945,
|
| 1523 |
+
"rewards/rejected": -13.8463773727417,
|
| 1524 |
+
"step": 1000
|
| 1525 |
+
},
|
| 1526 |
+
{
|
| 1527 |
+
"epoch": 8.08,
|
| 1528 |
+
"grad_norm": 0.00014151799507930023,
|
| 1529 |
+
"learning_rate": 1.0815327133708015e-06,
|
| 1530 |
+
"logits/chosen": -0.32878726720809937,
|
| 1531 |
+
"logits/rejected": -0.8332622647285461,
|
| 1532 |
+
"logps/chosen": -141.76785278320312,
|
| 1533 |
+
"logps/rejected": -1504.282958984375,
|
| 1534 |
+
"loss": 0.0,
|
| 1535 |
+
"rewards/accuracies": 1.0,
|
| 1536 |
+
"rewards/chosen": -0.9092656373977661,
|
| 1537 |
+
"rewards/margins": 13.11542797088623,
|
| 1538 |
+
"rewards/rejected": -14.024694442749023,
|
| 1539 |
+
"step": 1010
|
| 1540 |
+
},
|
| 1541 |
+
{
|
| 1542 |
+
"epoch": 8.16,
|
| 1543 |
+
"grad_norm": 0.00022245344369295513,
|
| 1544 |
+
"learning_rate": 9.963431452563331e-07,
|
| 1545 |
+
"logits/chosen": -0.39487534761428833,
|
| 1546 |
+
"logits/rejected": -0.8839765787124634,
|
| 1547 |
+
"logps/chosen": -171.04611206054688,
|
| 1548 |
+
"logps/rejected": -1563.74365234375,
|
| 1549 |
+
"loss": 0.0,
|
| 1550 |
+
"rewards/accuracies": 1.0,
|
| 1551 |
+
"rewards/chosen": -1.1952791213989258,
|
| 1552 |
+
"rewards/margins": 13.446779251098633,
|
| 1553 |
+
"rewards/rejected": -14.642059326171875,
|
| 1554 |
+
"step": 1020
|
| 1555 |
+
},
|
| 1556 |
+
{
|
| 1557 |
+
"epoch": 8.24,
|
| 1558 |
+
"grad_norm": 0.0001694321641851975,
|
| 1559 |
+
"learning_rate": 9.142755083243577e-07,
|
| 1560 |
+
"logits/chosen": -0.3724760413169861,
|
| 1561 |
+
"logits/rejected": -0.9137827754020691,
|
| 1562 |
+
"logps/chosen": -128.69775390625,
|
| 1563 |
+
"logps/rejected": -1472.91357421875,
|
| 1564 |
+
"loss": 0.0,
|
| 1565 |
+
"rewards/accuracies": 1.0,
|
| 1566 |
+
"rewards/chosen": -0.7906900644302368,
|
| 1567 |
+
"rewards/margins": 12.917367935180664,
|
| 1568 |
+
"rewards/rejected": -13.70805835723877,
|
| 1569 |
+
"step": 1030
|
| 1570 |
+
},
|
| 1571 |
+
{
|
| 1572 |
+
"epoch": 8.32,
|
| 1573 |
+
"grad_norm": 0.0004678365231010692,
|
| 1574 |
+
"learning_rate": 8.353937964495029e-07,
|
| 1575 |
+
"logits/chosen": -0.3941499590873718,
|
| 1576 |
+
"logits/rejected": -0.8401497006416321,
|
| 1577 |
+
"logps/chosen": -165.19224548339844,
|
| 1578 |
+
"logps/rejected": -1618.8040771484375,
|
| 1579 |
+
"loss": 0.0,
|
| 1580 |
+
"rewards/accuracies": 1.0,
|
| 1581 |
+
"rewards/chosen": -1.135845422744751,
|
| 1582 |
+
"rewards/margins": 14.052388191223145,
|
| 1583 |
+
"rewards/rejected": -15.188232421875,
|
| 1584 |
+
"step": 1040
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 8.4,
|
| 1588 |
+
"grad_norm": 0.00019961493455925282,
|
| 1589 |
+
"learning_rate": 7.597595192178702e-07,
|
| 1590 |
+
"logits/chosen": -0.3931668996810913,
|
| 1591 |
+
"logits/rejected": -0.8902953863143921,
|
| 1592 |
+
"logps/chosen": -154.61544799804688,
|
| 1593 |
+
"logps/rejected": -1578.909423828125,
|
| 1594 |
+
"loss": 0.0,
|
| 1595 |
+
"rewards/accuracies": 1.0,
|
| 1596 |
+
"rewards/chosen": -1.042776346206665,
|
| 1597 |
+
"rewards/margins": 13.743990898132324,
|
| 1598 |
+
"rewards/rejected": -14.786767959594727,
|
| 1599 |
+
"step": 1050
|
| 1600 |
+
},
|
| 1601 |
+
{
|
| 1602 |
+
"epoch": 8.48,
|
| 1603 |
+
"grad_norm": 0.00015962162148062636,
|
| 1604 |
+
"learning_rate": 6.874316539637127e-07,
|
| 1605 |
+
"logits/chosen": -0.4204545021057129,
|
| 1606 |
+
"logits/rejected": -0.9560686349868774,
|
| 1607 |
+
"logps/chosen": -128.07919311523438,
|
| 1608 |
+
"logps/rejected": -1500.8055419921875,
|
| 1609 |
+
"loss": 0.0,
|
| 1610 |
+
"rewards/accuracies": 1.0,
|
| 1611 |
+
"rewards/chosen": -0.7846135497093201,
|
| 1612 |
+
"rewards/margins": 13.211763381958008,
|
| 1613 |
+
"rewards/rejected": -13.996376037597656,
|
| 1614 |
+
"step": 1060
|
| 1615 |
+
},
|
| 1616 |
+
{
|
| 1617 |
+
"epoch": 8.56,
|
| 1618 |
+
"grad_norm": 0.001197256519458604,
|
| 1619 |
+
"learning_rate": 6.184665997806832e-07,
|
| 1620 |
+
"logits/chosen": -0.42808738350868225,
|
| 1621 |
+
"logits/rejected": -0.8699554204940796,
|
| 1622 |
+
"logps/chosen": -164.53189086914062,
|
| 1623 |
+
"logps/rejected": -1619.1337890625,
|
| 1624 |
+
"loss": 0.0,
|
| 1625 |
+
"rewards/accuracies": 1.0,
|
| 1626 |
+
"rewards/chosen": -1.1256366968154907,
|
| 1627 |
+
"rewards/margins": 14.069290161132812,
|
| 1628 |
+
"rewards/rejected": -15.194926261901855,
|
| 1629 |
+
"step": 1070
|
| 1630 |
+
},
|
| 1631 |
+
{
|
| 1632 |
+
"epoch": 8.64,
|
| 1633 |
+
"grad_norm": 0.0014134395460349601,
|
| 1634 |
+
"learning_rate": 5.529181335435124e-07,
|
| 1635 |
+
"logits/chosen": -0.39539769291877747,
|
| 1636 |
+
"logits/rejected": -0.9618415832519531,
|
| 1637 |
+
"logps/chosen": -124.12815856933594,
|
| 1638 |
+
"logps/rejected": -1469.0845947265625,
|
| 1639 |
+
"loss": 0.0,
|
| 1640 |
+
"rewards/accuracies": 1.0,
|
| 1641 |
+
"rewards/chosen": -0.7464931607246399,
|
| 1642 |
+
"rewards/margins": 12.919212341308594,
|
| 1643 |
+
"rewards/rejected": -13.665704727172852,
|
| 1644 |
+
"step": 1080
|
| 1645 |
+
},
|
| 1646 |
+
{
|
| 1647 |
+
"epoch": 8.72,
|
| 1648 |
+
"grad_norm": 0.00013913379217651929,
|
| 1649 |
+
"learning_rate": 4.908373679744316e-07,
|
| 1650 |
+
"logits/chosen": -0.4164504408836365,
|
| 1651 |
+
"logits/rejected": -1.0009126663208008,
|
| 1652 |
+
"logps/chosen": -139.15652465820312,
|
| 1653 |
+
"logps/rejected": -1507.514404296875,
|
| 1654 |
+
"loss": 0.0,
|
| 1655 |
+
"rewards/accuracies": 1.0,
|
| 1656 |
+
"rewards/chosen": -0.8938719034194946,
|
| 1657 |
+
"rewards/margins": 13.158498764038086,
|
| 1658 |
+
"rewards/rejected": -14.05237102508545,
|
| 1659 |
+
"step": 1090
|
| 1660 |
+
},
|
| 1661 |
+
{
|
| 1662 |
+
"epoch": 8.8,
|
| 1663 |
+
"grad_norm": 0.001742282508212273,
|
| 1664 |
+
"learning_rate": 4.322727117869951e-07,
|
| 1665 |
+
"logits/chosen": -0.45068103075027466,
|
| 1666 |
+
"logits/rejected": -0.9950752258300781,
|
| 1667 |
+
"logps/chosen": -129.97946166992188,
|
| 1668 |
+
"logps/rejected": -1477.4703369140625,
|
| 1669 |
+
"loss": 0.0,
|
| 1670 |
+
"rewards/accuracies": 1.0,
|
| 1671 |
+
"rewards/chosen": -0.8063904643058777,
|
| 1672 |
+
"rewards/margins": 12.941418647766113,
|
| 1673 |
+
"rewards/rejected": -13.747808456420898,
|
| 1674 |
+
"step": 1100
|
| 1675 |
+
},
|
| 1676 |
+
{
|
| 1677 |
+
"epoch": 8.88,
|
| 1678 |
+
"grad_norm": 0.00014054830993016855,
|
| 1679 |
+
"learning_rate": 3.772698319384349e-07,
|
| 1680 |
+
"logits/chosen": -0.3605000376701355,
|
| 1681 |
+
"logits/rejected": -0.9571415185928345,
|
| 1682 |
+
"logps/chosen": -102.00578308105469,
|
| 1683 |
+
"logps/rejected": -1384.301513671875,
|
| 1684 |
+
"loss": 0.0,
|
| 1685 |
+
"rewards/accuracies": 1.0,
|
| 1686 |
+
"rewards/chosen": -0.5351399183273315,
|
| 1687 |
+
"rewards/margins": 12.273322105407715,
|
| 1688 |
+
"rewards/rejected": -12.80846118927002,
|
| 1689 |
+
"step": 1110
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 8.96,
|
| 1693 |
+
"grad_norm": 0.00029001415653845576,
|
| 1694 |
+
"learning_rate": 3.258716180199278e-07,
|
| 1695 |
+
"logits/chosen": -0.35858815908432007,
|
| 1696 |
+
"logits/rejected": -0.9055237770080566,
|
| 1697 |
+
"logps/chosen": -139.0391387939453,
|
| 1698 |
+
"logps/rejected": -1482.545654296875,
|
| 1699 |
+
"loss": 0.0,
|
| 1700 |
+
"rewards/accuracies": 1.0,
|
| 1701 |
+
"rewards/chosen": -0.8927180171012878,
|
| 1702 |
+
"rewards/margins": 12.917207717895508,
|
| 1703 |
+
"rewards/rejected": -13.80992603302002,
|
| 1704 |
+
"step": 1120
|
| 1705 |
+
},
|
| 1706 |
+
{
|
| 1707 |
+
"epoch": 9.04,
|
| 1708 |
+
"grad_norm": 0.0005895794083238161,
|
| 1709 |
+
"learning_rate": 2.7811814881259503e-07,
|
| 1710 |
+
"logits/chosen": -0.344729483127594,
|
| 1711 |
+
"logits/rejected": -0.8606122136116028,
|
| 1712 |
+
"logps/chosen": -132.51242065429688,
|
| 1713 |
+
"logps/rejected": -1505.0494384765625,
|
| 1714 |
+
"loss": 0.0,
|
| 1715 |
+
"rewards/accuracies": 1.0,
|
| 1716 |
+
"rewards/chosen": -0.8304397463798523,
|
| 1717 |
+
"rewards/margins": 13.201517105102539,
|
| 1718 |
+
"rewards/rejected": -14.031954765319824,
|
| 1719 |
+
"step": 1130
|
| 1720 |
+
},
|
| 1721 |
+
{
|
| 1722 |
+
"epoch": 9.12,
|
| 1723 |
+
"grad_norm": 0.00013310639980010605,
|
| 1724 |
+
"learning_rate": 2.3404666103526542e-07,
|
| 1725 |
+
"logits/chosen": -0.3512091338634491,
|
| 1726 |
+
"logits/rejected": -0.8659976124763489,
|
| 1727 |
+
"logps/chosen": -145.8941192626953,
|
| 1728 |
+
"logps/rejected": -1529.08447265625,
|
| 1729 |
+
"loss": 0.0,
|
| 1730 |
+
"rewards/accuracies": 1.0,
|
| 1731 |
+
"rewards/chosen": -0.9467619061470032,
|
| 1732 |
+
"rewards/margins": 13.3263578414917,
|
| 1733 |
+
"rewards/rejected": -14.27312183380127,
|
| 1734 |
+
"step": 1140
|
| 1735 |
+
},
|
| 1736 |
+
{
|
| 1737 |
+
"epoch": 9.2,
|
| 1738 |
+
"grad_norm": 0.00012364944710080487,
|
| 1739 |
+
"learning_rate": 1.9369152030840553e-07,
|
| 1740 |
+
"logits/chosen": -0.35208699107170105,
|
| 1741 |
+
"logits/rejected": -0.899590790271759,
|
| 1742 |
+
"logps/chosen": -143.52272033691406,
|
| 1743 |
+
"logps/rejected": -1492.267578125,
|
| 1744 |
+
"loss": 0.0,
|
| 1745 |
+
"rewards/accuracies": 1.0,
|
| 1746 |
+
"rewards/chosen": -0.9294937252998352,
|
| 1747 |
+
"rewards/margins": 12.974908828735352,
|
| 1748 |
+
"rewards/rejected": -13.904401779174805,
|
| 1749 |
+
"step": 1150
|
| 1750 |
+
},
|
| 1751 |
+
{
|
| 1752 |
+
"epoch": 9.28,
|
| 1753 |
+
"grad_norm": 0.00013560146353849627,
|
| 1754 |
+
"learning_rate": 1.5708419435684463e-07,
|
| 1755 |
+
"logits/chosen": -0.380057156085968,
|
| 1756 |
+
"logits/rejected": -0.9084013104438782,
|
| 1757 |
+
"logps/chosen": -157.8520050048828,
|
| 1758 |
+
"logps/rejected": -1592.5137939453125,
|
| 1759 |
+
"loss": 0.0,
|
| 1760 |
+
"rewards/accuracies": 1.0,
|
| 1761 |
+
"rewards/chosen": -1.075063943862915,
|
| 1762 |
+
"rewards/margins": 13.840289115905762,
|
| 1763 |
+
"rewards/rejected": -14.915351867675781,
|
| 1764 |
+
"step": 1160
|
| 1765 |
+
},
|
| 1766 |
+
{
|
| 1767 |
+
"epoch": 9.36,
|
| 1768 |
+
"grad_norm": 0.00013265989294205703,
|
| 1769 |
+
"learning_rate": 1.2425322847218368e-07,
|
| 1770 |
+
"logits/chosen": -0.33621734380722046,
|
| 1771 |
+
"logits/rejected": -0.800961971282959,
|
| 1772 |
+
"logps/chosen": -154.0393524169922,
|
| 1773 |
+
"logps/rejected": -1592.49462890625,
|
| 1774 |
+
"loss": 0.0,
|
| 1775 |
+
"rewards/accuracies": 1.0,
|
| 1776 |
+
"rewards/chosen": -1.023474931716919,
|
| 1777 |
+
"rewards/margins": 13.9025297164917,
|
| 1778 |
+
"rewards/rejected": -14.926007270812988,
|
| 1779 |
+
"step": 1170
|
| 1780 |
+
},
|
| 1781 |
+
{
|
| 1782 |
+
"epoch": 9.44,
|
| 1783 |
+
"grad_norm": 0.003418968557861204,
|
| 1784 |
+
"learning_rate": 9.522422325404234e-08,
|
| 1785 |
+
"logits/chosen": -0.42315536737442017,
|
| 1786 |
+
"logits/rejected": -0.896791934967041,
|
| 1787 |
+
"logps/chosen": -162.86721801757812,
|
| 1788 |
+
"logps/rejected": -1653.62890625,
|
| 1789 |
+
"loss": 0.0,
|
| 1790 |
+
"rewards/accuracies": 1.0,
|
| 1791 |
+
"rewards/chosen": -1.1122061014175415,
|
| 1792 |
+
"rewards/margins": 14.423039436340332,
|
| 1793 |
+
"rewards/rejected": -15.535245895385742,
|
| 1794 |
+
"step": 1180
|
| 1795 |
+
},
|
| 1796 |
+
{
|
| 1797 |
+
"epoch": 9.52,
|
| 1798 |
+
"grad_norm": 0.0004120904531551459,
|
| 1799 |
+
"learning_rate": 7.001981464747565e-08,
|
| 1800 |
+
"logits/chosen": -0.4035482406616211,
|
| 1801 |
+
"logits/rejected": -0.9819211959838867,
|
| 1802 |
+
"logps/chosen": -137.95887756347656,
|
| 1803 |
+
"logps/rejected": -1535.572021484375,
|
| 1804 |
+
"loss": 0.0,
|
| 1805 |
+
"rewards/accuracies": 1.0,
|
| 1806 |
+
"rewards/chosen": -0.8891565203666687,
|
| 1807 |
+
"rewards/margins": 13.448022842407227,
|
| 1808 |
+
"rewards/rejected": -14.337178230285645,
|
| 1809 |
+
"step": 1190
|
| 1810 |
+
},
|
| 1811 |
+
{
|
| 1812 |
+
"epoch": 9.6,
|
| 1813 |
+
"grad_norm": 0.00017994417316440225,
|
| 1814 |
+
"learning_rate": 4.865965629214819e-08,
|
| 1815 |
+
"logits/chosen": -0.35764050483703613,
|
| 1816 |
+
"logits/rejected": -0.8284955024719238,
|
| 1817 |
+
"logps/chosen": -176.431396484375,
|
| 1818 |
+
"logps/rejected": -1656.4488525390625,
|
| 1819 |
+
"loss": 0.0,
|
| 1820 |
+
"rewards/accuracies": 1.0,
|
| 1821 |
+
"rewards/chosen": -1.2497293949127197,
|
| 1822 |
+
"rewards/margins": 14.32325267791748,
|
| 1823 |
+
"rewards/rejected": -15.572982788085938,
|
| 1824 |
+
"step": 1200
|
| 1825 |
+
},
|
| 1826 |
+
{
|
| 1827 |
+
"epoch": 9.68,
|
| 1828 |
+
"grad_norm": 0.00015125200057797107,
|
| 1829 |
+
"learning_rate": 3.1160404197018155e-08,
|
| 1830 |
+
"logits/chosen": -0.3409956097602844,
|
| 1831 |
+
"logits/rejected": -0.8691213726997375,
|
| 1832 |
+
"logps/chosen": -136.07290649414062,
|
| 1833 |
+
"logps/rejected": -1491.394775390625,
|
| 1834 |
+
"loss": 0.0,
|
| 1835 |
+
"rewards/accuracies": 1.0,
|
| 1836 |
+
"rewards/chosen": -0.8631143569946289,
|
| 1837 |
+
"rewards/margins": 13.04126262664795,
|
| 1838 |
+
"rewards/rejected": -13.904377937316895,
|
| 1839 |
+
"step": 1210
|
| 1840 |
+
},
|
| 1841 |
+
{
|
| 1842 |
+
"epoch": 9.76,
|
| 1843 |
+
"grad_norm": 0.006288398835461867,
|
| 1844 |
+
"learning_rate": 1.753570375247815e-08,
|
| 1845 |
+
"logits/chosen": -0.4766923785209656,
|
| 1846 |
+
"logits/rejected": -0.9798077344894409,
|
| 1847 |
+
"logps/chosen": -154.1293487548828,
|
| 1848 |
+
"logps/rejected": -1594.1053466796875,
|
| 1849 |
+
"loss": 0.0,
|
| 1850 |
+
"rewards/accuracies": 1.0,
|
| 1851 |
+
"rewards/chosen": -1.0420167446136475,
|
| 1852 |
+
"rewards/margins": 13.897071838378906,
|
| 1853 |
+
"rewards/rejected": -14.939088821411133,
|
| 1854 |
+
"step": 1220
|
| 1855 |
+
},
|
| 1856 |
+
{
|
| 1857 |
+
"epoch": 9.84,
|
| 1858 |
+
"grad_norm": 0.00015912389084664545,
|
| 1859 |
+
"learning_rate": 7.796179090094891e-09,
|
| 1860 |
+
"logits/chosen": -0.3149864971637726,
|
| 1861 |
+
"logits/rejected": -0.8940708041191101,
|
| 1862 |
+
"logps/chosen": -117.2905044555664,
|
| 1863 |
+
"logps/rejected": -1411.2535400390625,
|
| 1864 |
+
"loss": 0.0,
|
| 1865 |
+
"rewards/accuracies": 1.0,
|
| 1866 |
+
"rewards/chosen": -0.6836379766464233,
|
| 1867 |
+
"rewards/margins": 12.397500991821289,
|
| 1868 |
+
"rewards/rejected": -13.081140518188477,
|
| 1869 |
+
"step": 1230
|
| 1870 |
+
},
|
| 1871 |
+
{
|
| 1872 |
+
"epoch": 9.92,
|
| 1873 |
+
"grad_norm": 0.0002417004626132451,
|
| 1874 |
+
"learning_rate": 1.9494247982282386e-09,
|
| 1875 |
+
"logits/chosen": -0.42949041724205017,
|
| 1876 |
+
"logits/rejected": -0.9556988477706909,
|
| 1877 |
+
"logps/chosen": -133.76651000976562,
|
| 1878 |
+
"logps/rejected": -1445.5709228515625,
|
| 1879 |
+
"loss": 0.0,
|
| 1880 |
+
"rewards/accuracies": 1.0,
|
| 1881 |
+
"rewards/chosen": -0.8304960131645203,
|
| 1882 |
+
"rewards/margins": 12.607949256896973,
|
| 1883 |
+
"rewards/rejected": -13.438446044921875,
|
| 1884 |
+
"step": 1240
|
| 1885 |
+
},
|
| 1886 |
+
{
|
| 1887 |
+
"epoch": 10.0,
|
| 1888 |
+
"grad_norm": 0.0008051537929778158,
|
| 1889 |
+
"learning_rate": 0.0,
|
| 1890 |
+
"logits/chosen": -0.3784290850162506,
|
| 1891 |
+
"logits/rejected": -0.879416823387146,
|
| 1892 |
+
"logps/chosen": -144.29238891601562,
|
| 1893 |
+
"logps/rejected": -1541.6170654296875,
|
| 1894 |
+
"loss": 0.0,
|
| 1895 |
+
"rewards/accuracies": 1.0,
|
| 1896 |
+
"rewards/chosen": -0.9475828409194946,
|
| 1897 |
+
"rewards/margins": 13.463716506958008,
|
| 1898 |
+
"rewards/rejected": -14.411298751831055,
|
| 1899 |
+
"step": 1250
|
| 1900 |
+
},
|
| 1901 |
+
{
|
| 1902 |
+
"epoch": 10.0,
|
| 1903 |
+
"step": 1250,
|
| 1904 |
+
"total_flos": 0.0,
|
| 1905 |
+
"train_loss": 0.015015173686749768,
|
| 1906 |
+
"train_runtime": 115157.0828,
|
| 1907 |
+
"train_samples_per_second": 0.695,
|
| 1908 |
+
"train_steps_per_second": 0.011
|
| 1909 |
+
}
|
| 1910 |
+
],
|
| 1911 |
+
"logging_steps": 10,
|
| 1912 |
+
"max_steps": 1250,
|
| 1913 |
+
"num_input_tokens_seen": 0,
|
| 1914 |
+
"num_train_epochs": 10,
|
| 1915 |
+
"save_steps": 1000,
|
| 1916 |
+
"stateful_callbacks": {
|
| 1917 |
+
"TrainerControl": {
|
| 1918 |
+
"args": {
|
| 1919 |
+
"should_epoch_stop": false,
|
| 1920 |
+
"should_evaluate": false,
|
| 1921 |
+
"should_log": false,
|
| 1922 |
+
"should_save": true,
|
| 1923 |
+
"should_training_stop": true
|
| 1924 |
+
},
|
| 1925 |
+
"attributes": {}
|
| 1926 |
+
}
|
| 1927 |
+
},
|
| 1928 |
+
"total_flos": 0.0,
|
| 1929 |
+
"train_batch_size": 4,
|
| 1930 |
+
"trial_name": null,
|
| 1931 |
+
"trial_params": null
|
| 1932 |
+
}
|