Model save
Browse files- README.md +79 -0
- adapter_model.safetensors +1 -1
- all_results.json +21 -0
- eval_results.json +16 -0
- train_results.json +8 -0
- trainer_state.json +1516 -0
README.md
ADDED
|
@@ -0,0 +1,79 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
base_model: mistralai/Mistral-7B-v0.1
|
| 4 |
+
tags:
|
| 5 |
+
- generated_from_trainer
|
| 6 |
+
model-index:
|
| 7 |
+
- name: zephyr-7b-dpo-qlora
|
| 8 |
+
results: []
|
| 9 |
+
---
|
| 10 |
+
|
| 11 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 12 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 13 |
+
|
| 14 |
+
# zephyr-7b-dpo-qlora
|
| 15 |
+
|
| 16 |
+
This model is a fine-tuned version of [mistralai/Mistral-7B-v0.1](https://huggingface.co/mistralai/Mistral-7B-v0.1) on the None dataset.
|
| 17 |
+
It achieves the following results on the evaluation set:
|
| 18 |
+
- Loss: 0.0136
|
| 19 |
+
- Rewards/chosen: -376.5464
|
| 20 |
+
- Rewards/rejected: -330.4243
|
| 21 |
+
- Rewards/accuracies: 0.4544
|
| 22 |
+
- Rewards/margins: -46.1221
|
| 23 |
+
- Logps/rejected: -33295.5859
|
| 24 |
+
- Logps/chosen: -37927.6367
|
| 25 |
+
- Neglected: 256.0
|
| 26 |
+
- Selected: 0.0
|
| 27 |
+
|
| 28 |
+
## Model description
|
| 29 |
+
|
| 30 |
+
More information needed
|
| 31 |
+
|
| 32 |
+
## Intended uses & limitations
|
| 33 |
+
|
| 34 |
+
More information needed
|
| 35 |
+
|
| 36 |
+
## Training and evaluation data
|
| 37 |
+
|
| 38 |
+
More information needed
|
| 39 |
+
|
| 40 |
+
## Training procedure
|
| 41 |
+
|
| 42 |
+
### Training hyperparameters
|
| 43 |
+
|
| 44 |
+
The following hyperparameters were used during training:
|
| 45 |
+
- learning_rate: 5e-06
|
| 46 |
+
- train_batch_size: 4
|
| 47 |
+
- eval_batch_size: 8
|
| 48 |
+
- seed: 42
|
| 49 |
+
- distributed_type: multi-GPU
|
| 50 |
+
- num_devices: 4
|
| 51 |
+
- gradient_accumulation_steps: 4
|
| 52 |
+
- total_train_batch_size: 64
|
| 53 |
+
- total_eval_batch_size: 32
|
| 54 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 55 |
+
- lr_scheduler_type: cosine
|
| 56 |
+
- lr_scheduler_warmup_ratio: 0.1
|
| 57 |
+
- num_epochs: 1
|
| 58 |
+
|
| 59 |
+
### Training results
|
| 60 |
+
|
| 61 |
+
| Training Loss | Epoch | Step | Validation Loss | Rewards/chosen | Rewards/rejected | Rewards/accuracies | Rewards/margins | Logps/rejected | Logps/chosen | Neglected | Selected |
|
| 62 |
+
|:-------------:|:-----:|:----:|:---------------:|:--------------:|:----------------:|:------------------:|:---------------:|:--------------:|:------------:|:---------:|:--------:|
|
| 63 |
+
| 0.6727 | 0.1 | 100 | 0.6631 | 0.0074 | -0.0332 | 0.7024 | 0.0405 | -256.4745 | -272.2623 | 256.0 | 0.0 |
|
| 64 |
+
| 0.0392 | 0.21 | 200 | 0.0276 | -119.9914 | -105.4188 | 0.4464 | -14.5726 | -10795.0420 | -12272.1426 | 256.0 | 0.0 |
|
| 65 |
+
| 0.0208 | 0.31 | 300 | 0.0199 | -281.3865 | -245.2151 | 0.4444 | -36.1714 | -24774.6660 | -28411.6465 | 256.0 | 0.0 |
|
| 66 |
+
| 0.0157 | 0.42 | 400 | 0.0161 | -353.7562 | -307.1862 | 0.4563 | -46.5699 | -30971.7832 | -35648.6172 | 256.0 | 0.0 |
|
| 67 |
+
| 0.0182 | 0.52 | 500 | 0.0148 | -331.5956 | -289.6645 | 0.4464 | -41.9311 | -29219.6113 | -33432.5625 | 256.0 | 0.0 |
|
| 68 |
+
| 0.013 | 0.63 | 600 | 0.0143 | -356.6841 | -312.4188 | 0.4544 | -44.2654 | -31495.0312 | -35941.4141 | 256.0 | 0.0 |
|
| 69 |
+
| 0.0165 | 0.73 | 700 | 0.0143 | -353.6940 | -310.5345 | 0.4504 | -43.1595 | -31306.6094 | -35642.4023 | 256.0 | 0.0 |
|
| 70 |
+
| 0.0145 | 0.84 | 800 | 0.0135 | -374.0797 | -328.2772 | 0.4544 | -45.8026 | -33080.8789 | -37680.9766 | 256.0 | 0.0 |
|
| 71 |
+
| 0.0195 | 0.94 | 900 | 0.0137 | -376.5184 | -330.4032 | 0.4544 | -46.1152 | -33293.4727 | -37924.8398 | 256.0 | 0.0 |
|
| 72 |
+
|
| 73 |
+
|
| 74 |
+
### Framework versions
|
| 75 |
+
|
| 76 |
+
- Transformers 4.35.0
|
| 77 |
+
- Pytorch 2.1.1+cu121
|
| 78 |
+
- Datasets 2.14.6
|
| 79 |
+
- Tokenizers 0.14.1
|
adapter_model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 83945744
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9cb5c6fc7d4bdfeb0a0cf7462bf351aef0423bfa34b4b3b3c04ea9d593fe5fa3
|
| 3 |
size 83945744
|
all_results.json
ADDED
|
@@ -0,0 +1,21 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"eval_logps/chosen": -37927.63671875,
|
| 4 |
+
"eval_logps/rejected": -33295.5859375,
|
| 5 |
+
"eval_loss": 0.013621850870549679,
|
| 6 |
+
"eval_neglected": 256.0,
|
| 7 |
+
"eval_rewards/accuracies": 0.454365074634552,
|
| 8 |
+
"eval_rewards/chosen": -376.54638671875,
|
| 9 |
+
"eval_rewards/margins": -46.1220817565918,
|
| 10 |
+
"eval_rewards/rejected": -330.4242858886719,
|
| 11 |
+
"eval_runtime": 968.279,
|
| 12 |
+
"eval_samples": 2000,
|
| 13 |
+
"eval_samples_per_second": 2.066,
|
| 14 |
+
"eval_selected": 0.0,
|
| 15 |
+
"eval_steps_per_second": 0.065,
|
| 16 |
+
"train_loss": 0.11865393293933718,
|
| 17 |
+
"train_runtime": 42177.7253,
|
| 18 |
+
"train_samples": 61135,
|
| 19 |
+
"train_samples_per_second": 1.449,
|
| 20 |
+
"train_steps_per_second": 0.023
|
| 21 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"eval_logps/chosen": -37927.63671875,
|
| 4 |
+
"eval_logps/rejected": -33295.5859375,
|
| 5 |
+
"eval_loss": 0.013621850870549679,
|
| 6 |
+
"eval_neglected": 256.0,
|
| 7 |
+
"eval_rewards/accuracies": 0.454365074634552,
|
| 8 |
+
"eval_rewards/chosen": -376.54638671875,
|
| 9 |
+
"eval_rewards/margins": -46.1220817565918,
|
| 10 |
+
"eval_rewards/rejected": -330.4242858886719,
|
| 11 |
+
"eval_runtime": 968.279,
|
| 12 |
+
"eval_samples": 2000,
|
| 13 |
+
"eval_samples_per_second": 2.066,
|
| 14 |
+
"eval_selected": 0.0,
|
| 15 |
+
"eval_steps_per_second": 0.065
|
| 16 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"train_loss": 0.11865393293933718,
|
| 4 |
+
"train_runtime": 42177.7253,
|
| 5 |
+
"train_samples": 61135,
|
| 6 |
+
"train_samples_per_second": 1.449,
|
| 7 |
+
"train_steps_per_second": 0.023
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1516 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.9997382884061764,
|
| 5 |
+
"eval_steps": 100,
|
| 6 |
+
"global_step": 955,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.0,
|
| 13 |
+
"learning_rate": 5.208333333333333e-08,
|
| 14 |
+
"logps/chosen": -304.8013916015625,
|
| 15 |
+
"logps/rejected": -229.5030517578125,
|
| 16 |
+
"loss": 0.6931,
|
| 17 |
+
"neglected": 10.0,
|
| 18 |
+
"rewards/accuracies": 0.0,
|
| 19 |
+
"rewards/chosen": 0.0,
|
| 20 |
+
"rewards/margins": 0.0,
|
| 21 |
+
"rewards/rejected": 0.0,
|
| 22 |
+
"selected": 0.0,
|
| 23 |
+
"step": 1
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.01,
|
| 27 |
+
"learning_rate": 5.208333333333334e-07,
|
| 28 |
+
"logps/chosen": -313.4251708984375,
|
| 29 |
+
"logps/rejected": -277.2637023925781,
|
| 30 |
+
"loss": 0.693,
|
| 31 |
+
"neglected": 90.0,
|
| 32 |
+
"rewards/accuracies": 0.4861111044883728,
|
| 33 |
+
"rewards/chosen": 8.512949716532603e-05,
|
| 34 |
+
"rewards/margins": 0.0002041187253780663,
|
| 35 |
+
"rewards/rejected": -0.00011898923548869789,
|
| 36 |
+
"selected": 0.0,
|
| 37 |
+
"step": 10
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.02,
|
| 41 |
+
"learning_rate": 1.0416666666666667e-06,
|
| 42 |
+
"logps/chosen": -229.010986328125,
|
| 43 |
+
"logps/rejected": -232.58932495117188,
|
| 44 |
+
"loss": 0.6931,
|
| 45 |
+
"neglected": 242.0,
|
| 46 |
+
"rewards/accuracies": 0.4749999940395355,
|
| 47 |
+
"rewards/chosen": 0.00024582125479355454,
|
| 48 |
+
"rewards/margins": 0.0001970421290025115,
|
| 49 |
+
"rewards/rejected": 4.8779074859339744e-05,
|
| 50 |
+
"selected": 0.0,
|
| 51 |
+
"step": 20
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.03,
|
| 55 |
+
"learning_rate": 1.5625e-06,
|
| 56 |
+
"logps/chosen": -270.5987548828125,
|
| 57 |
+
"logps/rejected": -244.8210906982422,
|
| 58 |
+
"loss": 0.693,
|
| 59 |
+
"neglected": 402.0,
|
| 60 |
+
"rewards/accuracies": 0.5,
|
| 61 |
+
"rewards/chosen": -2.5821285817073658e-05,
|
| 62 |
+
"rewards/margins": 0.00013345989282242954,
|
| 63 |
+
"rewards/rejected": -0.00015928114589769393,
|
| 64 |
+
"selected": 0.0,
|
| 65 |
+
"step": 30
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.04,
|
| 69 |
+
"learning_rate": 2.0833333333333334e-06,
|
| 70 |
+
"logps/chosen": -270.58099365234375,
|
| 71 |
+
"logps/rejected": -263.6953125,
|
| 72 |
+
"loss": 0.6928,
|
| 73 |
+
"neglected": 562.0,
|
| 74 |
+
"rewards/accuracies": 0.5375000238418579,
|
| 75 |
+
"rewards/chosen": 0.0004904457600787282,
|
| 76 |
+
"rewards/margins": 0.0006916436250321567,
|
| 77 |
+
"rewards/rejected": -0.00020119785040151328,
|
| 78 |
+
"selected": 0.0,
|
| 79 |
+
"step": 40
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.05,
|
| 83 |
+
"learning_rate": 2.604166666666667e-06,
|
| 84 |
+
"logps/chosen": -255.3240509033203,
|
| 85 |
+
"logps/rejected": -249.5677490234375,
|
| 86 |
+
"loss": 0.6925,
|
| 87 |
+
"neglected": 722.0,
|
| 88 |
+
"rewards/accuracies": 0.581250011920929,
|
| 89 |
+
"rewards/chosen": 0.0003948220401071012,
|
| 90 |
+
"rewards/margins": 0.001298406976275146,
|
| 91 |
+
"rewards/rejected": -0.0009035851107910275,
|
| 92 |
+
"selected": 0.0,
|
| 93 |
+
"step": 50
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.06,
|
| 97 |
+
"learning_rate": 3.125e-06,
|
| 98 |
+
"logps/chosen": -278.1262512207031,
|
| 99 |
+
"logps/rejected": -257.60858154296875,
|
| 100 |
+
"loss": 0.6921,
|
| 101 |
+
"neglected": 882.0,
|
| 102 |
+
"rewards/accuracies": 0.6187499761581421,
|
| 103 |
+
"rewards/chosen": 0.001432361314073205,
|
| 104 |
+
"rewards/margins": 0.0019829857628792524,
|
| 105 |
+
"rewards/rejected": -0.000550624099560082,
|
| 106 |
+
"selected": 0.0,
|
| 107 |
+
"step": 60
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.07,
|
| 111 |
+
"learning_rate": 3.6458333333333333e-06,
|
| 112 |
+
"logps/chosen": -285.159423828125,
|
| 113 |
+
"logps/rejected": -263.88519287109375,
|
| 114 |
+
"loss": 0.6909,
|
| 115 |
+
"neglected": 1042.0,
|
| 116 |
+
"rewards/accuracies": 0.7124999761581421,
|
| 117 |
+
"rewards/chosen": 0.0027831769548356533,
|
| 118 |
+
"rewards/margins": 0.004454310052096844,
|
| 119 |
+
"rewards/rejected": -0.0016711335629224777,
|
| 120 |
+
"selected": 0.0,
|
| 121 |
+
"step": 70
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.08,
|
| 125 |
+
"learning_rate": 4.166666666666667e-06,
|
| 126 |
+
"logps/chosen": -292.46826171875,
|
| 127 |
+
"logps/rejected": -263.4407043457031,
|
| 128 |
+
"loss": 0.6879,
|
| 129 |
+
"neglected": 1202.0,
|
| 130 |
+
"rewards/accuracies": 0.706250011920929,
|
| 131 |
+
"rewards/chosen": 0.0038231350481510162,
|
| 132 |
+
"rewards/margins": 0.009836939163506031,
|
| 133 |
+
"rewards/rejected": -0.006013805046677589,
|
| 134 |
+
"selected": 0.0,
|
| 135 |
+
"step": 80
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.09,
|
| 139 |
+
"learning_rate": 4.6875000000000004e-06,
|
| 140 |
+
"logps/chosen": -280.253662109375,
|
| 141 |
+
"logps/rejected": -257.877197265625,
|
| 142 |
+
"loss": 0.6827,
|
| 143 |
+
"neglected": 1362.0,
|
| 144 |
+
"rewards/accuracies": 0.71875,
|
| 145 |
+
"rewards/chosen": 0.007530213333666325,
|
| 146 |
+
"rewards/margins": 0.01851402223110199,
|
| 147 |
+
"rewards/rejected": -0.010983810760080814,
|
| 148 |
+
"selected": 0.0,
|
| 149 |
+
"step": 90
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.1,
|
| 153 |
+
"learning_rate": 4.9997324926814375e-06,
|
| 154 |
+
"logps/chosen": -262.7320251464844,
|
| 155 |
+
"logps/rejected": -284.05206298828125,
|
| 156 |
+
"loss": 0.6727,
|
| 157 |
+
"neglected": 1522.0,
|
| 158 |
+
"rewards/accuracies": 0.699999988079071,
|
| 159 |
+
"rewards/chosen": 0.008031011559069157,
|
| 160 |
+
"rewards/margins": 0.026352444663643837,
|
| 161 |
+
"rewards/rejected": -0.018321430310606956,
|
| 162 |
+
"selected": 0.0,
|
| 163 |
+
"step": 100
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.1,
|
| 167 |
+
"eval_logps/chosen": -272.2622985839844,
|
| 168 |
+
"eval_logps/rejected": -256.4744873046875,
|
| 169 |
+
"eval_loss": 0.6630775332450867,
|
| 170 |
+
"eval_neglected": 256.0,
|
| 171 |
+
"eval_rewards/accuracies": 0.7023809552192688,
|
| 172 |
+
"eval_rewards/chosen": 0.0073691424913704395,
|
| 173 |
+
"eval_rewards/margins": 0.040533702820539474,
|
| 174 |
+
"eval_rewards/rejected": -0.03316456079483032,
|
| 175 |
+
"eval_runtime": 667.065,
|
| 176 |
+
"eval_samples_per_second": 2.998,
|
| 177 |
+
"eval_selected": 0.0,
|
| 178 |
+
"eval_steps_per_second": 0.094,
|
| 179 |
+
"step": 100
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 0.12,
|
| 183 |
+
"learning_rate": 4.996723692767927e-06,
|
| 184 |
+
"logps/chosen": -256.07342529296875,
|
| 185 |
+
"logps/rejected": -245.7953643798828,
|
| 186 |
+
"loss": 0.6539,
|
| 187 |
+
"neglected": 586.0,
|
| 188 |
+
"rewards/accuracies": 0.6937500238418579,
|
| 189 |
+
"rewards/chosen": -0.0058533623814582825,
|
| 190 |
+
"rewards/margins": 0.04092119634151459,
|
| 191 |
+
"rewards/rejected": -0.04677455872297287,
|
| 192 |
+
"selected": 0.0,
|
| 193 |
+
"step": 110
|
| 194 |
+
},
|
| 195 |
+
{
|
| 196 |
+
"epoch": 0.13,
|
| 197 |
+
"learning_rate": 4.9903757462135984e-06,
|
| 198 |
+
"logps/chosen": -270.7391357421875,
|
| 199 |
+
"logps/rejected": -248.25411987304688,
|
| 200 |
+
"loss": 0.6261,
|
| 201 |
+
"neglected": 746.0,
|
| 202 |
+
"rewards/accuracies": 0.675000011920929,
|
| 203 |
+
"rewards/chosen": -0.039820872247219086,
|
| 204 |
+
"rewards/margins": 0.06930369138717651,
|
| 205 |
+
"rewards/rejected": -0.1091245636343956,
|
| 206 |
+
"selected": 0.0,
|
| 207 |
+
"step": 120
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 0.14,
|
| 211 |
+
"learning_rate": 4.980697142834315e-06,
|
| 212 |
+
"logps/chosen": -312.620849609375,
|
| 213 |
+
"logps/rejected": -286.03900146484375,
|
| 214 |
+
"loss": 0.588,
|
| 215 |
+
"neglected": 906.0,
|
| 216 |
+
"rewards/accuracies": 0.6499999761581421,
|
| 217 |
+
"rewards/chosen": -0.1690162718296051,
|
| 218 |
+
"rewards/margins": 0.05675806850194931,
|
| 219 |
+
"rewards/rejected": -0.2257743626832962,
|
| 220 |
+
"selected": 0.0,
|
| 221 |
+
"step": 130
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 0.15,
|
| 225 |
+
"learning_rate": 4.967700826904229e-06,
|
| 226 |
+
"logps/chosen": -297.30609130859375,
|
| 227 |
+
"logps/rejected": -288.460693359375,
|
| 228 |
+
"loss": 0.5194,
|
| 229 |
+
"neglected": 1066.0,
|
| 230 |
+
"rewards/accuracies": 0.6312500238418579,
|
| 231 |
+
"rewards/chosen": -0.4745141863822937,
|
| 232 |
+
"rewards/margins": 0.12637066841125488,
|
| 233 |
+
"rewards/rejected": -0.6008848547935486,
|
| 234 |
+
"selected": 0.0,
|
| 235 |
+
"step": 140
|
| 236 |
+
},
|
| 237 |
+
{
|
| 238 |
+
"epoch": 0.16,
|
| 239 |
+
"learning_rate": 4.951404179843963e-06,
|
| 240 |
+
"logps/chosen": -450.36004638671875,
|
| 241 |
+
"logps/rejected": -442.3370056152344,
|
| 242 |
+
"loss": 0.3643,
|
| 243 |
+
"neglected": 1226.0,
|
| 244 |
+
"rewards/accuracies": 0.606249988079071,
|
| 245 |
+
"rewards/chosen": -1.6727848052978516,
|
| 246 |
+
"rewards/margins": 0.14853505790233612,
|
| 247 |
+
"rewards/rejected": -1.821319818496704,
|
| 248 |
+
"selected": 0.0,
|
| 249 |
+
"step": 150
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 0.17,
|
| 253 |
+
"learning_rate": 4.931828996974498e-06,
|
| 254 |
+
"logps/chosen": -864.3917846679688,
|
| 255 |
+
"logps/rejected": -883.9728393554688,
|
| 256 |
+
"loss": 0.1924,
|
| 257 |
+
"neglected": 1386.0,
|
| 258 |
+
"rewards/accuracies": 0.48750001192092896,
|
| 259 |
+
"rewards/chosen": -6.123291492462158,
|
| 260 |
+
"rewards/margins": 0.1330125331878662,
|
| 261 |
+
"rewards/rejected": -6.2563042640686035,
|
| 262 |
+
"selected": 0.0,
|
| 263 |
+
"step": 160
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 0.18,
|
| 267 |
+
"learning_rate": 4.909001458367867e-06,
|
| 268 |
+
"logps/chosen": -2515.188232421875,
|
| 269 |
+
"logps/rejected": -2068.07666015625,
|
| 270 |
+
"loss": 0.0935,
|
| 271 |
+
"neglected": 1546.0,
|
| 272 |
+
"rewards/accuracies": 0.5,
|
| 273 |
+
"rewards/chosen": -22.28314781188965,
|
| 274 |
+
"rewards/margins": -4.024808406829834,
|
| 275 |
+
"rewards/rejected": -18.25834083557129,
|
| 276 |
+
"selected": 0.0,
|
| 277 |
+
"step": 170
|
| 278 |
+
},
|
| 279 |
+
{
|
| 280 |
+
"epoch": 0.19,
|
| 281 |
+
"learning_rate": 4.882952093833628e-06,
|
| 282 |
+
"logps/chosen": -4048.93603515625,
|
| 283 |
+
"logps/rejected": -3594.38134765625,
|
| 284 |
+
"loss": 0.0695,
|
| 285 |
+
"neglected": 1706.0,
|
| 286 |
+
"rewards/accuracies": 0.42500001192092896,
|
| 287 |
+
"rewards/chosen": -38.03619384765625,
|
| 288 |
+
"rewards/margins": -4.711598873138428,
|
| 289 |
+
"rewards/rejected": -33.32460021972656,
|
| 290 |
+
"selected": 0.0,
|
| 291 |
+
"step": 180
|
| 292 |
+
},
|
| 293 |
+
{
|
| 294 |
+
"epoch": 0.2,
|
| 295 |
+
"learning_rate": 4.853715742087947e-06,
|
| 296 |
+
"logps/chosen": -5398.57421875,
|
| 297 |
+
"logps/rejected": -4498.2265625,
|
| 298 |
+
"loss": 0.0402,
|
| 299 |
+
"neglected": 1866.0,
|
| 300 |
+
"rewards/accuracies": 0.4375,
|
| 301 |
+
"rewards/chosen": -51.07048416137695,
|
| 302 |
+
"rewards/margins": -8.596471786499023,
|
| 303 |
+
"rewards/rejected": -42.47401428222656,
|
| 304 |
+
"selected": 0.0,
|
| 305 |
+
"step": 190
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 0.21,
|
| 309 |
+
"learning_rate": 4.821331504159906e-06,
|
| 310 |
+
"logps/chosen": -8471.2490234375,
|
| 311 |
+
"logps/rejected": -7905.1455078125,
|
| 312 |
+
"loss": 0.0392,
|
| 313 |
+
"neglected": 2026.0,
|
| 314 |
+
"rewards/accuracies": 0.4375,
|
| 315 |
+
"rewards/chosen": -82.20130920410156,
|
| 316 |
+
"rewards/margins": -5.739879131317139,
|
| 317 |
+
"rewards/rejected": -76.46143341064453,
|
| 318 |
+
"selected": 0.0,
|
| 319 |
+
"step": 200
|
| 320 |
+
},
|
| 321 |
+
{
|
| 322 |
+
"epoch": 0.21,
|
| 323 |
+
"eval_logps/chosen": -12272.142578125,
|
| 324 |
+
"eval_logps/rejected": -10795.0419921875,
|
| 325 |
+
"eval_loss": 0.027584508061408997,
|
| 326 |
+
"eval_neglected": 256.0,
|
| 327 |
+
"eval_rewards/accuracies": 0.4464285671710968,
|
| 328 |
+
"eval_rewards/chosen": -119.99142456054688,
|
| 329 |
+
"eval_rewards/margins": -14.572598457336426,
|
| 330 |
+
"eval_rewards/rejected": -105.4188232421875,
|
| 331 |
+
"eval_runtime": 670.5193,
|
| 332 |
+
"eval_samples_per_second": 2.983,
|
| 333 |
+
"eval_selected": 0.0,
|
| 334 |
+
"eval_steps_per_second": 0.094,
|
| 335 |
+
"step": 200
|
| 336 |
+
},
|
| 337 |
+
{
|
| 338 |
+
"epoch": 0.22,
|
| 339 |
+
"learning_rate": 4.7858426910973435e-06,
|
| 340 |
+
"logps/chosen": -12588.5732421875,
|
| 341 |
+
"logps/rejected": -11012.65625,
|
| 342 |
+
"loss": 0.0212,
|
| 343 |
+
"neglected": 586.0,
|
| 344 |
+
"rewards/accuracies": 0.44999998807907104,
|
| 345 |
+
"rewards/chosen": -122.9972152709961,
|
| 346 |
+
"rewards/margins": -15.338798522949219,
|
| 347 |
+
"rewards/rejected": -107.6584243774414,
|
| 348 |
+
"selected": 0.0,
|
| 349 |
+
"step": 210
|
| 350 |
+
},
|
| 351 |
+
{
|
| 352 |
+
"epoch": 0.23,
|
| 353 |
+
"learning_rate": 4.747296766042161e-06,
|
| 354 |
+
"logps/chosen": -9843.9521484375,
|
| 355 |
+
"logps/rejected": -9481.474609375,
|
| 356 |
+
"loss": 0.0355,
|
| 357 |
+
"neglected": 746.0,
|
| 358 |
+
"rewards/accuracies": 0.48750001192092896,
|
| 359 |
+
"rewards/chosen": -95.80561065673828,
|
| 360 |
+
"rewards/margins": -3.6035759449005127,
|
| 361 |
+
"rewards/rejected": -92.20204162597656,
|
| 362 |
+
"selected": 0.0,
|
| 363 |
+
"step": 220
|
| 364 |
+
},
|
| 365 |
+
{
|
| 366 |
+
"epoch": 0.24,
|
| 367 |
+
"learning_rate": 4.705745280752586e-06,
|
| 368 |
+
"logps/chosen": -9174.8818359375,
|
| 369 |
+
"logps/rejected": -8592.8603515625,
|
| 370 |
+
"loss": 0.0293,
|
| 371 |
+
"neglected": 906.0,
|
| 372 |
+
"rewards/accuracies": 0.46875,
|
| 373 |
+
"rewards/chosen": -88.80033874511719,
|
| 374 |
+
"rewards/margins": -5.414186000823975,
|
| 375 |
+
"rewards/rejected": -83.38614654541016,
|
| 376 |
+
"selected": 0.0,
|
| 377 |
+
"step": 230
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 0.25,
|
| 381 |
+
"learning_rate": 4.661243806657256e-06,
|
| 382 |
+
"logps/chosen": -10734.61328125,
|
| 383 |
+
"logps/rejected": -10024.224609375,
|
| 384 |
+
"loss": 0.0237,
|
| 385 |
+
"neglected": 1066.0,
|
| 386 |
+
"rewards/accuracies": 0.4437499940395355,
|
| 387 |
+
"rewards/chosen": -104.76686096191406,
|
| 388 |
+
"rewards/margins": -6.675856113433838,
|
| 389 |
+
"rewards/rejected": -98.09098815917969,
|
| 390 |
+
"selected": 0.0,
|
| 391 |
+
"step": 240
|
| 392 |
+
},
|
| 393 |
+
{
|
| 394 |
+
"epoch": 0.26,
|
| 395 |
+
"learning_rate": 4.613851860533367e-06,
|
| 396 |
+
"logps/chosen": -12696.328125,
|
| 397 |
+
"logps/rejected": -12253.8046875,
|
| 398 |
+
"loss": 0.0216,
|
| 399 |
+
"neglected": 1226.0,
|
| 400 |
+
"rewards/accuracies": 0.5062500238418579,
|
| 401 |
+
"rewards/chosen": -124.18550872802734,
|
| 402 |
+
"rewards/margins": -4.2357869148254395,
|
| 403 |
+
"rewards/rejected": -119.94972229003906,
|
| 404 |
+
"selected": 0.0,
|
| 405 |
+
"step": 250
|
| 406 |
+
},
|
| 407 |
+
{
|
| 408 |
+
"epoch": 0.27,
|
| 409 |
+
"learning_rate": 4.563632824908252e-06,
|
| 410 |
+
"logps/chosen": -16011.0830078125,
|
| 411 |
+
"logps/rejected": -15822.650390625,
|
| 412 |
+
"loss": 0.0224,
|
| 413 |
+
"neglected": 1386.0,
|
| 414 |
+
"rewards/accuracies": 0.4937500059604645,
|
| 415 |
+
"rewards/chosen": -157.71505737304688,
|
| 416 |
+
"rewards/margins": -1.8953163623809814,
|
| 417 |
+
"rewards/rejected": -155.81973266601562,
|
| 418 |
+
"selected": 0.0,
|
| 419 |
+
"step": 260
|
| 420 |
+
},
|
| 421 |
+
{
|
| 422 |
+
"epoch": 0.28,
|
| 423 |
+
"learning_rate": 4.510653863290871e-06,
|
| 424 |
+
"logps/chosen": -21082.57421875,
|
| 425 |
+
"logps/rejected": -18679.11328125,
|
| 426 |
+
"loss": 0.0214,
|
| 427 |
+
"neglected": 1546.0,
|
| 428 |
+
"rewards/accuracies": 0.48124998807907104,
|
| 429 |
+
"rewards/chosen": -208.0191192626953,
|
| 430 |
+
"rewards/margins": -23.887981414794922,
|
| 431 |
+
"rewards/rejected": -184.1311492919922,
|
| 432 |
+
"selected": 0.0,
|
| 433 |
+
"step": 270
|
| 434 |
+
},
|
| 435 |
+
{
|
| 436 |
+
"epoch": 0.29,
|
| 437 |
+
"learning_rate": 4.454985830346574e-06,
|
| 438 |
+
"logps/chosen": -24180.8984375,
|
| 439 |
+
"logps/rejected": -22673.47265625,
|
| 440 |
+
"loss": 0.0205,
|
| 441 |
+
"neglected": 1706.0,
|
| 442 |
+
"rewards/accuracies": 0.4312500059604645,
|
| 443 |
+
"rewards/chosen": -238.95950317382812,
|
| 444 |
+
"rewards/margins": -14.777926445007324,
|
| 445 |
+
"rewards/rejected": -224.18154907226562,
|
| 446 |
+
"selected": 0.0,
|
| 447 |
+
"step": 280
|
| 448 |
+
},
|
| 449 |
+
{
|
| 450 |
+
"epoch": 0.3,
|
| 451 |
+
"learning_rate": 4.396703177135262e-06,
|
| 452 |
+
"logps/chosen": -25520.37890625,
|
| 453 |
+
"logps/rejected": -20765.65234375,
|
| 454 |
+
"loss": 0.0181,
|
| 455 |
+
"neglected": 1866.0,
|
| 456 |
+
"rewards/accuracies": 0.375,
|
| 457 |
+
"rewards/chosen": -252.4154815673828,
|
| 458 |
+
"rewards/margins": -46.95801544189453,
|
| 459 |
+
"rewards/rejected": -205.4574432373047,
|
| 460 |
+
"selected": 0.0,
|
| 461 |
+
"step": 290
|
| 462 |
+
},
|
| 463 |
+
{
|
| 464 |
+
"epoch": 0.31,
|
| 465 |
+
"learning_rate": 4.335883851539693e-06,
|
| 466 |
+
"logps/chosen": -26487.619140625,
|
| 467 |
+
"logps/rejected": -21971.16796875,
|
| 468 |
+
"loss": 0.0208,
|
| 469 |
+
"neglected": 2026.0,
|
| 470 |
+
"rewards/accuracies": 0.4124999940395355,
|
| 471 |
+
"rewards/chosen": -262.32843017578125,
|
| 472 |
+
"rewards/margins": -44.91776657104492,
|
| 473 |
+
"rewards/rejected": -217.41067504882812,
|
| 474 |
+
"selected": 0.0,
|
| 475 |
+
"step": 300
|
| 476 |
+
},
|
| 477 |
+
{
|
| 478 |
+
"epoch": 0.31,
|
| 479 |
+
"eval_logps/chosen": -28411.646484375,
|
| 480 |
+
"eval_logps/rejected": -24774.666015625,
|
| 481 |
+
"eval_loss": 0.019909363240003586,
|
| 482 |
+
"eval_neglected": 256.0,
|
| 483 |
+
"eval_rewards/accuracies": 0.4444444477558136,
|
| 484 |
+
"eval_rewards/chosen": -281.386474609375,
|
| 485 |
+
"eval_rewards/margins": -36.1713981628418,
|
| 486 |
+
"eval_rewards/rejected": -245.21505737304688,
|
| 487 |
+
"eval_runtime": 505.6466,
|
| 488 |
+
"eval_samples_per_second": 3.955,
|
| 489 |
+
"eval_selected": 0.0,
|
| 490 |
+
"eval_steps_per_second": 0.125,
|
| 491 |
+
"step": 300
|
| 492 |
+
},
|
| 493 |
+
{
|
| 494 |
+
"epoch": 0.32,
|
| 495 |
+
"learning_rate": 4.2726091940171055e-06,
|
| 496 |
+
"logps/chosen": -23575.23046875,
|
| 497 |
+
"logps/rejected": -23384.05078125,
|
| 498 |
+
"loss": 0.0186,
|
| 499 |
+
"neglected": 586.0,
|
| 500 |
+
"rewards/accuracies": 0.5062500238418579,
|
| 501 |
+
"rewards/chosen": -233.19082641601562,
|
| 502 |
+
"rewards/margins": -2.1331238746643066,
|
| 503 |
+
"rewards/rejected": -231.0576934814453,
|
| 504 |
+
"selected": 0.0,
|
| 505 |
+
"step": 310
|
| 506 |
+
},
|
| 507 |
+
{
|
| 508 |
+
"epoch": 0.33,
|
| 509 |
+
"learning_rate": 4.206963828813555e-06,
|
| 510 |
+
"logps/chosen": -25525.6015625,
|
| 511 |
+
"logps/rejected": -22454.76953125,
|
| 512 |
+
"loss": 0.0147,
|
| 513 |
+
"neglected": 746.0,
|
| 514 |
+
"rewards/accuracies": 0.4375,
|
| 515 |
+
"rewards/chosen": -252.456787109375,
|
| 516 |
+
"rewards/margins": -30.628952026367188,
|
| 517 |
+
"rewards/rejected": -221.8278350830078,
|
| 518 |
+
"selected": 0.0,
|
| 519 |
+
"step": 320
|
| 520 |
+
},
|
| 521 |
+
{
|
| 522 |
+
"epoch": 0.35,
|
| 523 |
+
"learning_rate": 4.139035550786495e-06,
|
| 524 |
+
"logps/chosen": -23703.537109375,
|
| 525 |
+
"logps/rejected": -19463.728515625,
|
| 526 |
+
"loss": 0.022,
|
| 527 |
+
"neglected": 906.0,
|
| 528 |
+
"rewards/accuracies": 0.4749999940395355,
|
| 529 |
+
"rewards/chosen": -234.44473266601562,
|
| 530 |
+
"rewards/margins": -42.07261657714844,
|
| 531 |
+
"rewards/rejected": -192.3721160888672,
|
| 532 |
+
"selected": 0.0,
|
| 533 |
+
"step": 330
|
| 534 |
+
},
|
| 535 |
+
{
|
| 536 |
+
"epoch": 0.36,
|
| 537 |
+
"learning_rate": 4.068915207986931e-06,
|
| 538 |
+
"logps/chosen": -25306.37109375,
|
| 539 |
+
"logps/rejected": -21047.587890625,
|
| 540 |
+
"loss": 0.0215,
|
| 541 |
+
"neglected": 1066.0,
|
| 542 |
+
"rewards/accuracies": 0.39375001192092896,
|
| 543 |
+
"rewards/chosen": -250.39120483398438,
|
| 544 |
+
"rewards/margins": -42.19620132446289,
|
| 545 |
+
"rewards/rejected": -208.1950225830078,
|
| 546 |
+
"selected": 0.0,
|
| 547 |
+
"step": 340
|
| 548 |
+
},
|
| 549 |
+
{
|
| 550 |
+
"epoch": 0.37,
|
| 551 |
+
"learning_rate": 3.996696580158211e-06,
|
| 552 |
+
"logps/chosen": -30706.958984375,
|
| 553 |
+
"logps/rejected": -28025.75390625,
|
| 554 |
+
"loss": 0.0174,
|
| 555 |
+
"neglected": 1226.0,
|
| 556 |
+
"rewards/accuracies": 0.4124999940395355,
|
| 557 |
+
"rewards/chosen": -304.2236328125,
|
| 558 |
+
"rewards/margins": -26.65300941467285,
|
| 559 |
+
"rewards/rejected": -277.57061767578125,
|
| 560 |
+
"selected": 0.0,
|
| 561 |
+
"step": 350
|
| 562 |
+
},
|
| 563 |
+
{
|
| 564 |
+
"epoch": 0.38,
|
| 565 |
+
"learning_rate": 3.922476253313921e-06,
|
| 566 |
+
"logps/chosen": -30064.09375,
|
| 567 |
+
"logps/rejected": -27983.521484375,
|
| 568 |
+
"loss": 0.0134,
|
| 569 |
+
"neglected": 1386.0,
|
| 570 |
+
"rewards/accuracies": 0.4749999940395355,
|
| 571 |
+
"rewards/chosen": -298.0165710449219,
|
| 572 |
+
"rewards/margins": -20.747739791870117,
|
| 573 |
+
"rewards/rejected": -277.268798828125,
|
| 574 |
+
"selected": 0.0,
|
| 575 |
+
"step": 360
|
| 576 |
+
},
|
| 577 |
+
{
|
| 578 |
+
"epoch": 0.39,
|
| 579 |
+
"learning_rate": 3.846353490562664e-06,
|
| 580 |
+
"logps/chosen": -31143.537109375,
|
| 581 |
+
"logps/rejected": -27110.509765625,
|
| 582 |
+
"loss": 0.016,
|
| 583 |
+
"neglected": 1546.0,
|
| 584 |
+
"rewards/accuracies": 0.4625000059604645,
|
| 585 |
+
"rewards/chosen": -308.7967529296875,
|
| 586 |
+
"rewards/margins": -40.30244827270508,
|
| 587 |
+
"rewards/rejected": -268.49432373046875,
|
| 588 |
+
"selected": 0.0,
|
| 589 |
+
"step": 370
|
| 590 |
+
},
|
| 591 |
+
{
|
| 592 |
+
"epoch": 0.4,
|
| 593 |
+
"learning_rate": 3.768430099352445e-06,
|
| 594 |
+
"logps/chosen": -32500.40234375,
|
| 595 |
+
"logps/rejected": -29200.681640625,
|
| 596 |
+
"loss": 0.0166,
|
| 597 |
+
"neglected": 1706.0,
|
| 598 |
+
"rewards/accuracies": 0.4375,
|
| 599 |
+
"rewards/chosen": -322.29559326171875,
|
| 600 |
+
"rewards/margins": -32.805747985839844,
|
| 601 |
+
"rewards/rejected": -289.4898376464844,
|
| 602 |
+
"selected": 0.0,
|
| 603 |
+
"step": 380
|
| 604 |
+
},
|
| 605 |
+
{
|
| 606 |
+
"epoch": 0.41,
|
| 607 |
+
"learning_rate": 3.6888102953122307e-06,
|
| 608 |
+
"logps/chosen": -30651.95703125,
|
| 609 |
+
"logps/rejected": -26855.337890625,
|
| 610 |
+
"loss": 0.0167,
|
| 611 |
+
"neglected": 1866.0,
|
| 612 |
+
"rewards/accuracies": 0.4312500059604645,
|
| 613 |
+
"rewards/chosen": -303.84075927734375,
|
| 614 |
+
"rewards/margins": -37.63256072998047,
|
| 615 |
+
"rewards/rejected": -266.208251953125,
|
| 616 |
+
"selected": 0.0,
|
| 617 |
+
"step": 390
|
| 618 |
+
},
|
| 619 |
+
{
|
| 620 |
+
"epoch": 0.42,
|
| 621 |
+
"learning_rate": 3.607600562872785e-06,
|
| 622 |
+
"logps/chosen": -34714.859375,
|
| 623 |
+
"logps/rejected": -29928.291015625,
|
| 624 |
+
"loss": 0.0157,
|
| 625 |
+
"neglected": 2026.0,
|
| 626 |
+
"rewards/accuracies": 0.41874998807907104,
|
| 627 |
+
"rewards/chosen": -344.1798400878906,
|
| 628 |
+
"rewards/margins": -47.40506362915039,
|
| 629 |
+
"rewards/rejected": -296.77471923828125,
|
| 630 |
+
"selected": 0.0,
|
| 631 |
+
"step": 400
|
| 632 |
+
},
|
| 633 |
+
{
|
| 634 |
+
"epoch": 0.42,
|
| 635 |
+
"eval_logps/chosen": -35648.6171875,
|
| 636 |
+
"eval_logps/rejected": -30971.783203125,
|
| 637 |
+
"eval_loss": 0.01606718823313713,
|
| 638 |
+
"eval_neglected": 256.0,
|
| 639 |
+
"eval_rewards/accuracies": 0.4563491940498352,
|
| 640 |
+
"eval_rewards/chosen": -353.75616455078125,
|
| 641 |
+
"eval_rewards/margins": -46.569923400878906,
|
| 642 |
+
"eval_rewards/rejected": -307.1862487792969,
|
| 643 |
+
"eval_runtime": 506.2643,
|
| 644 |
+
"eval_samples_per_second": 3.951,
|
| 645 |
+
"eval_selected": 0.0,
|
| 646 |
+
"eval_steps_per_second": 0.124,
|
| 647 |
+
"step": 400
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 0.43,
|
| 651 |
+
"learning_rate": 3.5249095128531863e-06,
|
| 652 |
+
"logps/chosen": -35328.84375,
|
| 653 |
+
"logps/rejected": -30211.0,
|
| 654 |
+
"loss": 0.0123,
|
| 655 |
+
"neglected": 586.0,
|
| 656 |
+
"rewards/accuracies": 0.4437499940395355,
|
| 657 |
+
"rewards/chosen": -350.4105529785156,
|
| 658 |
+
"rewards/margins": -50.80678176879883,
|
| 659 |
+
"rewards/rejected": -299.60382080078125,
|
| 660 |
+
"selected": 0.0,
|
| 661 |
+
"step": 410
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 0.44,
|
| 665 |
+
"learning_rate": 3.4408477372034743e-06,
|
| 666 |
+
"logps/chosen": -28428.931640625,
|
| 667 |
+
"logps/rejected": -24432.90234375,
|
| 668 |
+
"loss": 0.0282,
|
| 669 |
+
"neglected": 746.0,
|
| 670 |
+
"rewards/accuracies": 0.4437499940395355,
|
| 671 |
+
"rewards/chosen": -281.9709777832031,
|
| 672 |
+
"rewards/margins": -40.07168960571289,
|
| 673 |
+
"rewards/rejected": -241.89932250976562,
|
| 674 |
+
"selected": 0.0,
|
| 675 |
+
"step": 420
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 0.45,
|
| 679 |
+
"learning_rate": 3.355527661097728e-06,
|
| 680 |
+
"logps/chosen": -32917.82421875,
|
| 681 |
+
"logps/rejected": -31499.80078125,
|
| 682 |
+
"loss": 0.0179,
|
| 683 |
+
"neglected": 906.0,
|
| 684 |
+
"rewards/accuracies": 0.4749999940395355,
|
| 685 |
+
"rewards/chosen": -326.5057678222656,
|
| 686 |
+
"rewards/margins": -14.0189790725708,
|
| 687 |
+
"rewards/rejected": -312.4867858886719,
|
| 688 |
+
"selected": 0.0,
|
| 689 |
+
"step": 430
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 0.46,
|
| 693 |
+
"learning_rate": 3.269063392575352e-06,
|
| 694 |
+
"logps/chosen": -32941.2421875,
|
| 695 |
+
"logps/rejected": -28788.244140625,
|
| 696 |
+
"loss": 0.018,
|
| 697 |
+
"neglected": 1066.0,
|
| 698 |
+
"rewards/accuracies": 0.45625001192092896,
|
| 699 |
+
"rewards/chosen": -326.6683349609375,
|
| 700 |
+
"rewards/margins": -41.23744583129883,
|
| 701 |
+
"rewards/rejected": -285.4308776855469,
|
| 702 |
+
"selected": 0.0,
|
| 703 |
+
"step": 440
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 0.47,
|
| 707 |
+
"learning_rate": 3.181570569931697e-06,
|
| 708 |
+
"logps/chosen": -28051.92578125,
|
| 709 |
+
"logps/rejected": -26678.359375,
|
| 710 |
+
"loss": 0.025,
|
| 711 |
+
"neglected": 1226.0,
|
| 712 |
+
"rewards/accuracies": 0.41874998807907104,
|
| 713 |
+
"rewards/chosen": -278.347900390625,
|
| 714 |
+
"rewards/margins": -13.726778984069824,
|
| 715 |
+
"rewards/rejected": -264.62115478515625,
|
| 716 |
+
"selected": 0.0,
|
| 717 |
+
"step": 450
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 0.48,
|
| 721 |
+
"learning_rate": 3.09316620706208e-06,
|
| 722 |
+
"logps/chosen": -38179.28515625,
|
| 723 |
+
"logps/rejected": -31792.712890625,
|
| 724 |
+
"loss": 0.0135,
|
| 725 |
+
"neglected": 1386.0,
|
| 726 |
+
"rewards/accuracies": 0.4124999940395355,
|
| 727 |
+
"rewards/chosen": -378.8819885253906,
|
| 728 |
+
"rewards/margins": -63.55814743041992,
|
| 729 |
+
"rewards/rejected": -315.3238220214844,
|
| 730 |
+
"selected": 0.0,
|
| 731 |
+
"step": 460
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 0.49,
|
| 735 |
+
"learning_rate": 3.0039685369660785e-06,
|
| 736 |
+
"logps/chosen": -34295.48046875,
|
| 737 |
+
"logps/rejected": -30349.291015625,
|
| 738 |
+
"loss": 0.0128,
|
| 739 |
+
"neglected": 1546.0,
|
| 740 |
+
"rewards/accuracies": 0.48124998807907104,
|
| 741 |
+
"rewards/chosen": -340.40753173828125,
|
| 742 |
+
"rewards/margins": -39.258506774902344,
|
| 743 |
+
"rewards/rejected": -301.1490173339844,
|
| 744 |
+
"selected": 0.0,
|
| 745 |
+
"step": 470
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 0.5,
|
| 749 |
+
"learning_rate": 2.91409685362137e-06,
|
| 750 |
+
"logps/chosen": -30752.1875,
|
| 751 |
+
"logps/rejected": -30045.68359375,
|
| 752 |
+
"loss": 0.0118,
|
| 753 |
+
"neglected": 1706.0,
|
| 754 |
+
"rewards/accuracies": 0.45625001192092896,
|
| 755 |
+
"rewards/chosen": -305.1645202636719,
|
| 756 |
+
"rewards/margins": -7.10043478012085,
|
| 757 |
+
"rewards/rejected": -298.06414794921875,
|
| 758 |
+
"selected": 0.0,
|
| 759 |
+
"step": 480
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 0.51,
|
| 763 |
+
"learning_rate": 2.8236713524386085e-06,
|
| 764 |
+
"logps/chosen": -32674.708984375,
|
| 765 |
+
"logps/rejected": -27154.64453125,
|
| 766 |
+
"loss": 0.0173,
|
| 767 |
+
"neglected": 1866.0,
|
| 768 |
+
"rewards/accuracies": 0.42500001192092896,
|
| 769 |
+
"rewards/chosen": -324.16522216796875,
|
| 770 |
+
"rewards/margins": -54.92888641357422,
|
| 771 |
+
"rewards/rejected": -269.236328125,
|
| 772 |
+
"selected": 0.0,
|
| 773 |
+
"step": 490
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 0.52,
|
| 777 |
+
"learning_rate": 2.7328129695107205e-06,
|
| 778 |
+
"logps/chosen": -33942.0390625,
|
| 779 |
+
"logps/rejected": -26798.068359375,
|
| 780 |
+
"loss": 0.0182,
|
| 781 |
+
"neglected": 2026.0,
|
| 782 |
+
"rewards/accuracies": 0.39375001192092896,
|
| 783 |
+
"rewards/chosen": -336.67034912109375,
|
| 784 |
+
"rewards/margins": -71.16191864013672,
|
| 785 |
+
"rewards/rejected": -265.5083923339844,
|
| 786 |
+
"selected": 0.0,
|
| 787 |
+
"step": 500
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 0.52,
|
| 791 |
+
"eval_logps/chosen": -33432.5625,
|
| 792 |
+
"eval_logps/rejected": -29219.611328125,
|
| 793 |
+
"eval_loss": 0.014793259091675282,
|
| 794 |
+
"eval_neglected": 256.0,
|
| 795 |
+
"eval_rewards/accuracies": 0.4464285671710968,
|
| 796 |
+
"eval_rewards/chosen": -331.59564208984375,
|
| 797 |
+
"eval_rewards/margins": -41.931129455566406,
|
| 798 |
+
"eval_rewards/rejected": -289.6645202636719,
|
| 799 |
+
"eval_runtime": 505.8824,
|
| 800 |
+
"eval_samples_per_second": 3.953,
|
| 801 |
+
"eval_selected": 0.0,
|
| 802 |
+
"eval_steps_per_second": 0.125,
|
| 803 |
+
"step": 500
|
| 804 |
+
},
|
| 805 |
+
{
|
| 806 |
+
"epoch": 0.53,
|
| 807 |
+
"learning_rate": 2.641643219871597e-06,
|
| 808 |
+
"logps/chosen": -31474.45703125,
|
| 809 |
+
"logps/rejected": -25789.703125,
|
| 810 |
+
"loss": 0.0149,
|
| 811 |
+
"neglected": 586.0,
|
| 812 |
+
"rewards/accuracies": 0.46875,
|
| 813 |
+
"rewards/chosen": -312.119384765625,
|
| 814 |
+
"rewards/margins": -56.47794723510742,
|
| 815 |
+
"rewards/rejected": -255.6414337158203,
|
| 816 |
+
"selected": 0.0,
|
| 817 |
+
"step": 510
|
| 818 |
+
},
|
| 819 |
+
{
|
| 820 |
+
"epoch": 0.54,
|
| 821 |
+
"learning_rate": 2.5502840349805074e-06,
|
| 822 |
+
"logps/chosen": -32054.20703125,
|
| 823 |
+
"logps/rejected": -27586.98046875,
|
| 824 |
+
"loss": 0.0173,
|
| 825 |
+
"neglected": 746.0,
|
| 826 |
+
"rewards/accuracies": 0.4437499940395355,
|
| 827 |
+
"rewards/chosen": -317.7613830566406,
|
| 828 |
+
"rewards/margins": -44.480018615722656,
|
| 829 |
+
"rewards/rejected": -273.2813720703125,
|
| 830 |
+
"selected": 0.0,
|
| 831 |
+
"step": 520
|
| 832 |
+
},
|
| 833 |
+
{
|
| 834 |
+
"epoch": 0.55,
|
| 835 |
+
"learning_rate": 2.4588575996495797e-06,
|
| 836 |
+
"logps/chosen": -34159.5703125,
|
| 837 |
+
"logps/rejected": -30448.79296875,
|
| 838 |
+
"loss": 0.0138,
|
| 839 |
+
"neglected": 906.0,
|
| 840 |
+
"rewards/accuracies": 0.40625,
|
| 841 |
+
"rewards/chosen": -338.7040710449219,
|
| 842 |
+
"rewards/margins": -37.0126953125,
|
| 843 |
+
"rewards/rejected": -301.69134521484375,
|
| 844 |
+
"selected": 0.0,
|
| 845 |
+
"step": 530
|
| 846 |
+
},
|
| 847 |
+
{
|
| 848 |
+
"epoch": 0.57,
|
| 849 |
+
"learning_rate": 2.367486188632446e-06,
|
| 850 |
+
"logps/chosen": -33792.76171875,
|
| 851 |
+
"logps/rejected": -29700.40625,
|
| 852 |
+
"loss": 0.0078,
|
| 853 |
+
"neglected": 1066.0,
|
| 854 |
+
"rewards/accuracies": 0.40625,
|
| 855 |
+
"rewards/chosen": -335.1407775878906,
|
| 856 |
+
"rewards/margins": -40.700645446777344,
|
| 857 |
+
"rewards/rejected": -294.440185546875,
|
| 858 |
+
"selected": 0.0,
|
| 859 |
+
"step": 540
|
| 860 |
+
},
|
| 861 |
+
{
|
| 862 |
+
"epoch": 0.58,
|
| 863 |
+
"learning_rate": 2.276292003092593e-06,
|
| 864 |
+
"logps/chosen": -33080.16796875,
|
| 865 |
+
"logps/rejected": -29123.927734375,
|
| 866 |
+
"loss": 0.0145,
|
| 867 |
+
"neglected": 1226.0,
|
| 868 |
+
"rewards/accuracies": 0.4312500059604645,
|
| 869 |
+
"rewards/chosen": -328.01007080078125,
|
| 870 |
+
"rewards/margins": -39.278717041015625,
|
| 871 |
+
"rewards/rejected": -288.7312927246094,
|
| 872 |
+
"selected": 0.0,
|
| 873 |
+
"step": 550
|
| 874 |
+
},
|
| 875 |
+
{
|
| 876 |
+
"epoch": 0.59,
|
| 877 |
+
"learning_rate": 2.1853970071701415e-06,
|
| 878 |
+
"logps/chosen": -29881.09375,
|
| 879 |
+
"logps/rejected": -24806.671875,
|
| 880 |
+
"loss": 0.0136,
|
| 881 |
+
"neglected": 1386.0,
|
| 882 |
+
"rewards/accuracies": 0.4437499940395355,
|
| 883 |
+
"rewards/chosen": -296.17803955078125,
|
| 884 |
+
"rewards/margins": -50.19990539550781,
|
| 885 |
+
"rewards/rejected": -245.9781494140625,
|
| 886 |
+
"selected": 0.0,
|
| 887 |
+
"step": 560
|
| 888 |
+
},
|
| 889 |
+
{
|
| 890 |
+
"epoch": 0.6,
|
| 891 |
+
"learning_rate": 2.0949227648656194e-06,
|
| 892 |
+
"logps/chosen": -34481.3203125,
|
| 893 |
+
"logps/rejected": -30493.02734375,
|
| 894 |
+
"loss": 0.0172,
|
| 895 |
+
"neglected": 1546.0,
|
| 896 |
+
"rewards/accuracies": 0.4312500059604645,
|
| 897 |
+
"rewards/chosen": -342.2419738769531,
|
| 898 |
+
"rewards/margins": -39.841697692871094,
|
| 899 |
+
"rewards/rejected": -302.4002685546875,
|
| 900 |
+
"selected": 0.0,
|
| 901 |
+
"step": 570
|
| 902 |
+
},
|
| 903 |
+
{
|
| 904 |
+
"epoch": 0.61,
|
| 905 |
+
"learning_rate": 2.00499027745888e-06,
|
| 906 |
+
"logps/chosen": -34967.98828125,
|
| 907 |
+
"logps/rejected": -29546.712890625,
|
| 908 |
+
"loss": 0.0141,
|
| 909 |
+
"neglected": 1706.0,
|
| 910 |
+
"rewards/accuracies": 0.4000000059604645,
|
| 911 |
+
"rewards/chosen": -347.0220031738281,
|
| 912 |
+
"rewards/margins": -54.10175323486328,
|
| 913 |
+
"rewards/rejected": -292.9202575683594,
|
| 914 |
+
"selected": 0.0,
|
| 915 |
+
"step": 580
|
| 916 |
+
},
|
| 917 |
+
{
|
| 918 |
+
"epoch": 0.62,
|
| 919 |
+
"learning_rate": 1.915719821680624e-06,
|
| 920 |
+
"logps/chosen": -32696.44921875,
|
| 921 |
+
"logps/rejected": -30430.255859375,
|
| 922 |
+
"loss": 0.0167,
|
| 923 |
+
"neglected": 1866.0,
|
| 924 |
+
"rewards/accuracies": 0.45625001192092896,
|
| 925 |
+
"rewards/chosen": -324.4429626464844,
|
| 926 |
+
"rewards/margins": -22.577922821044922,
|
| 927 |
+
"rewards/rejected": -301.8650207519531,
|
| 928 |
+
"selected": 0.0,
|
| 929 |
+
"step": 590
|
| 930 |
+
},
|
| 931 |
+
{
|
| 932 |
+
"epoch": 0.63,
|
| 933 |
+
"learning_rate": 1.8272307888529276e-06,
|
| 934 |
+
"logps/chosen": -36048.4453125,
|
| 935 |
+
"logps/rejected": -33494.70703125,
|
| 936 |
+
"loss": 0.013,
|
| 937 |
+
"neglected": 2026.0,
|
| 938 |
+
"rewards/accuracies": 0.4749999940395355,
|
| 939 |
+
"rewards/chosen": -357.53070068359375,
|
| 940 |
+
"rewards/margins": -25.526748657226562,
|
| 941 |
+
"rewards/rejected": -332.0039978027344,
|
| 942 |
+
"selected": 0.0,
|
| 943 |
+
"step": 600
|
| 944 |
+
},
|
| 945 |
+
{
|
| 946 |
+
"epoch": 0.63,
|
| 947 |
+
"eval_logps/chosen": -35941.4140625,
|
| 948 |
+
"eval_logps/rejected": -31495.03125,
|
| 949 |
+
"eval_loss": 0.014261237345635891,
|
| 950 |
+
"eval_neglected": 256.0,
|
| 951 |
+
"eval_rewards/accuracies": 0.454365074634552,
|
| 952 |
+
"eval_rewards/chosen": -356.68414306640625,
|
| 953 |
+
"eval_rewards/margins": -44.265377044677734,
|
| 954 |
+
"eval_rewards/rejected": -312.41876220703125,
|
| 955 |
+
"eval_runtime": 505.9708,
|
| 956 |
+
"eval_samples_per_second": 3.953,
|
| 957 |
+
"eval_selected": 0.0,
|
| 958 |
+
"eval_steps_per_second": 0.125,
|
| 959 |
+
"step": 600
|
| 960 |
+
},
|
| 961 |
+
{
|
| 962 |
+
"epoch": 0.64,
|
| 963 |
+
"learning_rate": 1.739641525213929e-06,
|
| 964 |
+
"logps/chosen": -36224.6953125,
|
| 965 |
+
"logps/rejected": -28970.59765625,
|
| 966 |
+
"loss": 0.0178,
|
| 967 |
+
"neglected": 586.0,
|
| 968 |
+
"rewards/accuracies": 0.4124999940395355,
|
| 969 |
+
"rewards/chosen": -359.3634948730469,
|
| 970 |
+
"rewards/margins": -71.96965789794922,
|
| 971 |
+
"rewards/rejected": -287.39385986328125,
|
| 972 |
+
"selected": 0.0,
|
| 973 |
+
"step": 610
|
| 974 |
+
},
|
| 975 |
+
{
|
| 976 |
+
"epoch": 0.65,
|
| 977 |
+
"learning_rate": 1.6530691736402317e-06,
|
| 978 |
+
"logps/chosen": -33294.37109375,
|
| 979 |
+
"logps/rejected": -29921.0,
|
| 980 |
+
"loss": 0.0107,
|
| 981 |
+
"neglected": 746.0,
|
| 982 |
+
"rewards/accuracies": 0.45625001192092896,
|
| 983 |
+
"rewards/chosen": -330.2255859375,
|
| 984 |
+
"rewards/margins": -33.42811965942383,
|
| 985 |
+
"rewards/rejected": -296.7974548339844,
|
| 986 |
+
"selected": 0.0,
|
| 987 |
+
"step": 620
|
| 988 |
+
},
|
| 989 |
+
{
|
| 990 |
+
"epoch": 0.66,
|
| 991 |
+
"learning_rate": 1.5676295169786864e-06,
|
| 992 |
+
"logps/chosen": -33129.5703125,
|
| 993 |
+
"logps/rejected": -29490.744140625,
|
| 994 |
+
"loss": 0.0122,
|
| 995 |
+
"neglected": 906.0,
|
| 996 |
+
"rewards/accuracies": 0.4437499940395355,
|
| 997 |
+
"rewards/chosen": -328.56658935546875,
|
| 998 |
+
"rewards/margins": -36.11338806152344,
|
| 999 |
+
"rewards/rejected": -292.45318603515625,
|
| 1000 |
+
"selected": 0.0,
|
| 1001 |
+
"step": 630
|
| 1002 |
+
},
|
| 1003 |
+
{
|
| 1004 |
+
"epoch": 0.67,
|
| 1005 |
+
"learning_rate": 1.4834368231970922e-06,
|
| 1006 |
+
"logps/chosen": -31400.677734375,
|
| 1007 |
+
"logps/rejected": -29454.287109375,
|
| 1008 |
+
"loss": 0.0158,
|
| 1009 |
+
"neglected": 1066.0,
|
| 1010 |
+
"rewards/accuracies": 0.46875,
|
| 1011 |
+
"rewards/chosen": -311.5920104980469,
|
| 1012 |
+
"rewards/margins": -19.44215202331543,
|
| 1013 |
+
"rewards/rejected": -292.14984130859375,
|
| 1014 |
+
"selected": 0.0,
|
| 1015 |
+
"step": 640
|
| 1016 |
+
},
|
| 1017 |
+
{
|
| 1018 |
+
"epoch": 0.68,
|
| 1019 |
+
"learning_rate": 1.4006036925609245e-06,
|
| 1020 |
+
"logps/chosen": -37346.015625,
|
| 1021 |
+
"logps/rejected": -32676.51171875,
|
| 1022 |
+
"loss": 0.0115,
|
| 1023 |
+
"neglected": 1226.0,
|
| 1024 |
+
"rewards/accuracies": 0.4124999940395355,
|
| 1025 |
+
"rewards/chosen": -370.58929443359375,
|
| 1026 |
+
"rewards/margins": -46.472129821777344,
|
| 1027 |
+
"rewards/rejected": -324.11724853515625,
|
| 1028 |
+
"selected": 0.0,
|
| 1029 |
+
"step": 650
|
| 1030 |
+
},
|
| 1031 |
+
{
|
| 1032 |
+
"epoch": 0.69,
|
| 1033 |
+
"learning_rate": 1.3192409070404582e-06,
|
| 1034 |
+
"logps/chosen": -35501.5234375,
|
| 1035 |
+
"logps/rejected": -31327.859375,
|
| 1036 |
+
"loss": 0.0195,
|
| 1037 |
+
"neglected": 1386.0,
|
| 1038 |
+
"rewards/accuracies": 0.4312500059604645,
|
| 1039 |
+
"rewards/chosen": -352.1575622558594,
|
| 1040 |
+
"rewards/margins": -41.29918670654297,
|
| 1041 |
+
"rewards/rejected": -310.8583984375,
|
| 1042 |
+
"selected": 0.0,
|
| 1043 |
+
"step": 660
|
| 1044 |
+
},
|
| 1045 |
+
{
|
| 1046 |
+
"epoch": 0.7,
|
| 1047 |
+
"learning_rate": 1.2394572821496953e-06,
|
| 1048 |
+
"logps/chosen": -34126.65234375,
|
| 1049 |
+
"logps/rejected": -28228.837890625,
|
| 1050 |
+
"loss": 0.0213,
|
| 1051 |
+
"neglected": 1546.0,
|
| 1052 |
+
"rewards/accuracies": 0.4375,
|
| 1053 |
+
"rewards/chosen": -338.5702209472656,
|
| 1054 |
+
"rewards/margins": -58.62493896484375,
|
| 1055 |
+
"rewards/rejected": -279.9453125,
|
| 1056 |
+
"selected": 0.0,
|
| 1057 |
+
"step": 670
|
| 1058 |
+
},
|
| 1059 |
+
{
|
| 1060 |
+
"epoch": 0.71,
|
| 1061 |
+
"learning_rate": 1.1613595214152713e-06,
|
| 1062 |
+
"logps/chosen": -33340.77734375,
|
| 1063 |
+
"logps/rejected": -30065.52734375,
|
| 1064 |
+
"loss": 0.0155,
|
| 1065 |
+
"neglected": 1706.0,
|
| 1066 |
+
"rewards/accuracies": 0.45625001192092896,
|
| 1067 |
+
"rewards/chosen": -330.6611633300781,
|
| 1068 |
+
"rewards/margins": -32.5056037902832,
|
| 1069 |
+
"rewards/rejected": -298.15557861328125,
|
| 1070 |
+
"selected": 0.0,
|
| 1071 |
+
"step": 680
|
| 1072 |
+
},
|
| 1073 |
+
{
|
| 1074 |
+
"epoch": 0.72,
|
| 1075 |
+
"learning_rate": 1.0850520736699362e-06,
|
| 1076 |
+
"logps/chosen": -30460.240234375,
|
| 1077 |
+
"logps/rejected": -24525.88671875,
|
| 1078 |
+
"loss": 0.0166,
|
| 1079 |
+
"neglected": 1866.0,
|
| 1080 |
+
"rewards/accuracies": 0.38749998807907104,
|
| 1081 |
+
"rewards/chosen": -302.1998291015625,
|
| 1082 |
+
"rewards/margins": -58.95698165893555,
|
| 1083 |
+
"rewards/rejected": -243.2428436279297,
|
| 1084 |
+
"selected": 0.0,
|
| 1085 |
+
"step": 690
|
| 1086 |
+
},
|
| 1087 |
+
{
|
| 1088 |
+
"epoch": 0.73,
|
| 1089 |
+
"learning_rate": 1.0106369933615043e-06,
|
| 1090 |
+
"logps/chosen": -34674.8515625,
|
| 1091 |
+
"logps/rejected": -27218.890625,
|
| 1092 |
+
"loss": 0.0165,
|
| 1093 |
+
"neglected": 2026.0,
|
| 1094 |
+
"rewards/accuracies": 0.40625,
|
| 1095 |
+
"rewards/chosen": -344.1193542480469,
|
| 1096 |
+
"rewards/margins": -74.25706481933594,
|
| 1097 |
+
"rewards/rejected": -269.8622741699219,
|
| 1098 |
+
"selected": 0.0,
|
| 1099 |
+
"step": 700
|
| 1100 |
+
},
|
| 1101 |
+
{
|
| 1102 |
+
"epoch": 0.73,
|
| 1103 |
+
"eval_logps/chosen": -35642.40234375,
|
| 1104 |
+
"eval_logps/rejected": -31306.609375,
|
| 1105 |
+
"eval_loss": 0.014253102242946625,
|
| 1106 |
+
"eval_neglected": 256.0,
|
| 1107 |
+
"eval_rewards/accuracies": 0.4503968358039856,
|
| 1108 |
+
"eval_rewards/chosen": -353.6939697265625,
|
| 1109 |
+
"eval_rewards/margins": -43.15950012207031,
|
| 1110 |
+
"eval_rewards/rejected": -310.53448486328125,
|
| 1111 |
+
"eval_runtime": 505.4591,
|
| 1112 |
+
"eval_samples_per_second": 3.957,
|
| 1113 |
+
"eval_selected": 0.0,
|
| 1114 |
+
"eval_steps_per_second": 0.125,
|
| 1115 |
+
"step": 700
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 0.74,
|
| 1119 |
+
"learning_rate": 9.382138040640714e-07,
|
| 1120 |
+
"logps/chosen": -35340.9296875,
|
| 1121 |
+
"logps/rejected": -29694.728515625,
|
| 1122 |
+
"loss": 0.0173,
|
| 1123 |
+
"neglected": 586.0,
|
| 1124 |
+
"rewards/accuracies": 0.41874998807907104,
|
| 1125 |
+
"rewards/chosen": -350.610595703125,
|
| 1126 |
+
"rewards/margins": -55.991127014160156,
|
| 1127 |
+
"rewards/rejected": -294.61944580078125,
|
| 1128 |
+
"selected": 0.0,
|
| 1129 |
+
"step": 710
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 0.75,
|
| 1133 |
+
"learning_rate": 8.678793653740633e-07,
|
| 1134 |
+
"logps/chosen": -33239.89453125,
|
| 1135 |
+
"logps/rejected": -29252.65234375,
|
| 1136 |
+
"loss": 0.0194,
|
| 1137 |
+
"neglected": 746.0,
|
| 1138 |
+
"rewards/accuracies": 0.41874998807907104,
|
| 1139 |
+
"rewards/chosen": -329.77923583984375,
|
| 1140 |
+
"rewards/margins": -39.71256637573242,
|
| 1141 |
+
"rewards/rejected": -290.0666809082031,
|
| 1142 |
+
"selected": 0.0,
|
| 1143 |
+
"step": 720
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 0.76,
|
| 1147 |
+
"learning_rate": 7.997277433690984e-07,
|
| 1148 |
+
"logps/chosen": -33704.45703125,
|
| 1149 |
+
"logps/rejected": -27931.458984375,
|
| 1150 |
+
"loss": 0.0181,
|
| 1151 |
+
"neglected": 906.0,
|
| 1152 |
+
"rewards/accuracies": 0.36250001192092896,
|
| 1153 |
+
"rewards/chosen": -334.1654052734375,
|
| 1154 |
+
"rewards/margins": -57.19194793701172,
|
| 1155 |
+
"rewards/rejected": -276.97344970703125,
|
| 1156 |
+
"selected": 0.0,
|
| 1157 |
+
"step": 730
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 0.77,
|
| 1161 |
+
"learning_rate": 7.338500848029603e-07,
|
| 1162 |
+
"logps/chosen": -38638.05078125,
|
| 1163 |
+
"logps/rejected": -33758.7890625,
|
| 1164 |
+
"loss": 0.0091,
|
| 1165 |
+
"neglected": 1066.0,
|
| 1166 |
+
"rewards/accuracies": 0.4625000059604645,
|
| 1167 |
+
"rewards/chosen": -383.42559814453125,
|
| 1168 |
+
"rewards/margins": -48.62553024291992,
|
| 1169 |
+
"rewards/rejected": -334.800048828125,
|
| 1170 |
+
"selected": 0.0,
|
| 1171 |
+
"step": 740
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 0.79,
|
| 1175 |
+
"learning_rate": 6.70334495204884e-07,
|
| 1176 |
+
"logps/chosen": -35638.7421875,
|
| 1177 |
+
"logps/rejected": -31301.58984375,
|
| 1178 |
+
"loss": 0.0156,
|
| 1179 |
+
"neglected": 1226.0,
|
| 1180 |
+
"rewards/accuracies": 0.39375001192092896,
|
| 1181 |
+
"rewards/chosen": -353.8833312988281,
|
| 1182 |
+
"rewards/margins": -43.371212005615234,
|
| 1183 |
+
"rewards/rejected": -310.51214599609375,
|
| 1184 |
+
"selected": 0.0,
|
| 1185 |
+
"step": 750
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 0.8,
|
| 1189 |
+
"learning_rate": 6.092659210462232e-07,
|
| 1190 |
+
"logps/chosen": -38116.8828125,
|
| 1191 |
+
"logps/rejected": -33497.234375,
|
| 1192 |
+
"loss": 0.012,
|
| 1193 |
+
"neglected": 1386.0,
|
| 1194 |
+
"rewards/accuracies": 0.4437499940395355,
|
| 1195 |
+
"rewards/chosen": -378.3582763671875,
|
| 1196 |
+
"rewards/margins": -45.90812301635742,
|
| 1197 |
+
"rewards/rejected": -332.45013427734375,
|
| 1198 |
+
"selected": 0.0,
|
| 1199 |
+
"step": 760
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 0.81,
|
| 1203 |
+
"learning_rate": 5.507260361320738e-07,
|
| 1204 |
+
"logps/chosen": -36556.7890625,
|
| 1205 |
+
"logps/rejected": -33627.23828125,
|
| 1206 |
+
"loss": 0.0106,
|
| 1207 |
+
"neglected": 1546.0,
|
| 1208 |
+
"rewards/accuracies": 0.4312500059604645,
|
| 1209 |
+
"rewards/chosen": -362.6523742675781,
|
| 1210 |
+
"rewards/margins": -29.13754653930664,
|
| 1211 |
+
"rewards/rejected": -333.51483154296875,
|
| 1212 |
+
"selected": 0.0,
|
| 1213 |
+
"step": 770
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 0.82,
|
| 1217 |
+
"learning_rate": 4.947931323697983e-07,
|
| 1218 |
+
"logps/chosen": -33210.11328125,
|
| 1219 |
+
"logps/rejected": -29063.90234375,
|
| 1220 |
+
"loss": 0.013,
|
| 1221 |
+
"neglected": 1706.0,
|
| 1222 |
+
"rewards/accuracies": 0.41874998807907104,
|
| 1223 |
+
"rewards/chosen": -329.4010925292969,
|
| 1224 |
+
"rewards/margins": -41.132415771484375,
|
| 1225 |
+
"rewards/rejected": -288.2686462402344,
|
| 1226 |
+
"selected": 0.0,
|
| 1227 |
+
"step": 780
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 0.83,
|
| 1231 |
+
"learning_rate": 4.4154201506053985e-07,
|
| 1232 |
+
"logps/chosen": -35076.3515625,
|
| 1233 |
+
"logps/rejected": -33320.109375,
|
| 1234 |
+
"loss": 0.0146,
|
| 1235 |
+
"neglected": 1866.0,
|
| 1236 |
+
"rewards/accuracies": 0.45625001192092896,
|
| 1237 |
+
"rewards/chosen": -348.1772155761719,
|
| 1238 |
+
"rewards/margins": -17.483577728271484,
|
| 1239 |
+
"rewards/rejected": -330.693603515625,
|
| 1240 |
+
"selected": 0.0,
|
| 1241 |
+
"step": 790
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 0.84,
|
| 1245 |
+
"learning_rate": 3.910439028537638e-07,
|
| 1246 |
+
"logps/chosen": -37705.09375,
|
| 1247 |
+
"logps/rejected": -33144.11328125,
|
| 1248 |
+
"loss": 0.0145,
|
| 1249 |
+
"neglected": 2026.0,
|
| 1250 |
+
"rewards/accuracies": 0.42500001192092896,
|
| 1251 |
+
"rewards/chosen": -374.12744140625,
|
| 1252 |
+
"rewards/margins": -45.31665802001953,
|
| 1253 |
+
"rewards/rejected": -328.810791015625,
|
| 1254 |
+
"selected": 0.0,
|
| 1255 |
+
"step": 800
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 0.84,
|
| 1259 |
+
"eval_logps/chosen": -37680.9765625,
|
| 1260 |
+
"eval_logps/rejected": -33080.87890625,
|
| 1261 |
+
"eval_loss": 0.013520442880690098,
|
| 1262 |
+
"eval_neglected": 256.0,
|
| 1263 |
+
"eval_rewards/accuracies": 0.454365074634552,
|
| 1264 |
+
"eval_rewards/chosen": -374.0797424316406,
|
| 1265 |
+
"eval_rewards/margins": -45.80255889892578,
|
| 1266 |
+
"eval_rewards/rejected": -328.27716064453125,
|
| 1267 |
+
"eval_runtime": 506.1329,
|
| 1268 |
+
"eval_samples_per_second": 3.952,
|
| 1269 |
+
"eval_selected": 0.0,
|
| 1270 |
+
"eval_steps_per_second": 0.124,
|
| 1271 |
+
"step": 800
|
| 1272 |
+
},
|
| 1273 |
+
{
|
| 1274 |
+
"epoch": 0.85,
|
| 1275 |
+
"learning_rate": 3.4336633249862084e-07,
|
| 1276 |
+
"logps/chosen": -34459.21484375,
|
| 1277 |
+
"logps/rejected": -27397.837890625,
|
| 1278 |
+
"loss": 0.0064,
|
| 1279 |
+
"neglected": 586.0,
|
| 1280 |
+
"rewards/accuracies": 0.375,
|
| 1281 |
+
"rewards/chosen": -341.8682556152344,
|
| 1282 |
+
"rewards/margins": -70.08294677734375,
|
| 1283 |
+
"rewards/rejected": -271.78533935546875,
|
| 1284 |
+
"selected": 0.0,
|
| 1285 |
+
"step": 810
|
| 1286 |
+
},
|
| 1287 |
+
{
|
| 1288 |
+
"epoch": 0.86,
|
| 1289 |
+
"learning_rate": 2.98573068519539e-07,
|
| 1290 |
+
"logps/chosen": -37666.25390625,
|
| 1291 |
+
"logps/rejected": -29870.04296875,
|
| 1292 |
+
"loss": 0.0225,
|
| 1293 |
+
"neglected": 746.0,
|
| 1294 |
+
"rewards/accuracies": 0.4124999940395355,
|
| 1295 |
+
"rewards/chosen": -373.8772888183594,
|
| 1296 |
+
"rewards/margins": -77.50922393798828,
|
| 1297 |
+
"rewards/rejected": -296.3680725097656,
|
| 1298 |
+
"selected": 0.0,
|
| 1299 |
+
"step": 820
|
| 1300 |
+
},
|
| 1301 |
+
{
|
| 1302 |
+
"epoch": 0.87,
|
| 1303 |
+
"learning_rate": 2.5672401793681854e-07,
|
| 1304 |
+
"logps/chosen": -35177.1796875,
|
| 1305 |
+
"logps/rejected": -35886.7734375,
|
| 1306 |
+
"loss": 0.0127,
|
| 1307 |
+
"neglected": 906.0,
|
| 1308 |
+
"rewards/accuracies": 0.550000011920929,
|
| 1309 |
+
"rewards/chosen": -349.24237060546875,
|
| 1310 |
+
"rewards/margins": 6.9106035232543945,
|
| 1311 |
+
"rewards/rejected": -356.1529235839844,
|
| 1312 |
+
"selected": 0.0,
|
| 1313 |
+
"step": 830
|
| 1314 |
+
},
|
| 1315 |
+
{
|
| 1316 |
+
"epoch": 0.88,
|
| 1317 |
+
"learning_rate": 2.178751501463036e-07,
|
| 1318 |
+
"logps/chosen": -35737.97265625,
|
| 1319 |
+
"logps/rejected": -33345.86328125,
|
| 1320 |
+
"loss": 0.0107,
|
| 1321 |
+
"neglected": 1066.0,
|
| 1322 |
+
"rewards/accuracies": 0.45625001192092896,
|
| 1323 |
+
"rewards/chosen": -354.8319396972656,
|
| 1324 |
+
"rewards/margins": -23.900835037231445,
|
| 1325 |
+
"rewards/rejected": -330.93109130859375,
|
| 1326 |
+
"selected": 0.0,
|
| 1327 |
+
"step": 840
|
| 1328 |
+
},
|
| 1329 |
+
{
|
| 1330 |
+
"epoch": 0.89,
|
| 1331 |
+
"learning_rate": 1.820784220652766e-07,
|
| 1332 |
+
"logps/chosen": -36996.6171875,
|
| 1333 |
+
"logps/rejected": -31543.29296875,
|
| 1334 |
+
"loss": 0.0169,
|
| 1335 |
+
"neglected": 1226.0,
|
| 1336 |
+
"rewards/accuracies": 0.40625,
|
| 1337 |
+
"rewards/chosen": -367.1126708984375,
|
| 1338 |
+
"rewards/margins": -54.04352951049805,
|
| 1339 |
+
"rewards/rejected": -313.0691833496094,
|
| 1340 |
+
"selected": 0.0,
|
| 1341 |
+
"step": 850
|
| 1342 |
+
},
|
| 1343 |
+
{
|
| 1344 |
+
"epoch": 0.9,
|
| 1345 |
+
"learning_rate": 1.4938170864468636e-07,
|
| 1346 |
+
"logps/chosen": -35994.3359375,
|
| 1347 |
+
"logps/rejected": -32652.32421875,
|
| 1348 |
+
"loss": 0.0158,
|
| 1349 |
+
"neglected": 1386.0,
|
| 1350 |
+
"rewards/accuracies": 0.4312500059604645,
|
| 1351 |
+
"rewards/chosen": -357.20001220703125,
|
| 1352 |
+
"rewards/margins": -33.3302001953125,
|
| 1353 |
+
"rewards/rejected": -323.8697814941406,
|
| 1354 |
+
"selected": 0.0,
|
| 1355 |
+
"step": 860
|
| 1356 |
+
},
|
| 1357 |
+
{
|
| 1358 |
+
"epoch": 0.91,
|
| 1359 |
+
"learning_rate": 1.1982873884064466e-07,
|
| 1360 |
+
"logps/chosen": -31830.9375,
|
| 1361 |
+
"logps/rejected": -30581.083984375,
|
| 1362 |
+
"loss": 0.0158,
|
| 1363 |
+
"neglected": 1546.0,
|
| 1364 |
+
"rewards/accuracies": 0.543749988079071,
|
| 1365 |
+
"rewards/chosen": -315.9645690917969,
|
| 1366 |
+
"rewards/margins": -12.488165855407715,
|
| 1367 |
+
"rewards/rejected": -303.4764404296875,
|
| 1368 |
+
"selected": 0.0,
|
| 1369 |
+
"step": 870
|
| 1370 |
+
},
|
| 1371 |
+
{
|
| 1372 |
+
"epoch": 0.92,
|
| 1373 |
+
"learning_rate": 9.345903713082305e-08,
|
| 1374 |
+
"logps/chosen": -37537.7265625,
|
| 1375 |
+
"logps/rejected": -32952.95703125,
|
| 1376 |
+
"loss": 0.0108,
|
| 1377 |
+
"neglected": 1706.0,
|
| 1378 |
+
"rewards/accuracies": 0.48124998807907104,
|
| 1379 |
+
"rewards/chosen": -372.6186218261719,
|
| 1380 |
+
"rewards/margins": -45.61473846435547,
|
| 1381 |
+
"rewards/rejected": -327.00390625,
|
| 1382 |
+
"selected": 0.0,
|
| 1383 |
+
"step": 880
|
| 1384 |
+
},
|
| 1385 |
+
{
|
| 1386 |
+
"epoch": 0.93,
|
| 1387 |
+
"learning_rate": 7.030787065396866e-08,
|
| 1388 |
+
"logps/chosen": -39424.8828125,
|
| 1389 |
+
"logps/rejected": -34691.7265625,
|
| 1390 |
+
"loss": 0.0135,
|
| 1391 |
+
"neglected": 1866.0,
|
| 1392 |
+
"rewards/accuracies": 0.4312500059604645,
|
| 1393 |
+
"rewards/chosen": -391.5462341308594,
|
| 1394 |
+
"rewards/margins": -47.24760437011719,
|
| 1395 |
+
"rewards/rejected": -344.29864501953125,
|
| 1396 |
+
"selected": 0.0,
|
| 1397 |
+
"step": 890
|
| 1398 |
+
},
|
| 1399 |
+
{
|
| 1400 |
+
"epoch": 0.94,
|
| 1401 |
+
"learning_rate": 5.0406202043228604e-08,
|
| 1402 |
+
"logps/chosen": -36448.4765625,
|
| 1403 |
+
"logps/rejected": -30919.54296875,
|
| 1404 |
+
"loss": 0.0195,
|
| 1405 |
+
"neglected": 2026.0,
|
| 1406 |
+
"rewards/accuracies": 0.44999998807907104,
|
| 1407 |
+
"rewards/chosen": -361.8363037109375,
|
| 1408 |
+
"rewards/margins": -55.14497756958008,
|
| 1409 |
+
"rewards/rejected": -306.69134521484375,
|
| 1410 |
+
"selected": 0.0,
|
| 1411 |
+
"step": 900
|
| 1412 |
+
},
|
| 1413 |
+
{
|
| 1414 |
+
"epoch": 0.94,
|
| 1415 |
+
"eval_logps/chosen": -37924.83984375,
|
| 1416 |
+
"eval_logps/rejected": -33293.47265625,
|
| 1417 |
+
"eval_loss": 0.013714035972952843,
|
| 1418 |
+
"eval_neglected": 256.0,
|
| 1419 |
+
"eval_rewards/accuracies": 0.454365074634552,
|
| 1420 |
+
"eval_rewards/chosen": -376.51837158203125,
|
| 1421 |
+
"eval_rewards/margins": -46.11524200439453,
|
| 1422 |
+
"eval_rewards/rejected": -330.4031677246094,
|
| 1423 |
+
"eval_runtime": 967.1825,
|
| 1424 |
+
"eval_samples_per_second": 2.068,
|
| 1425 |
+
"eval_selected": 0.0,
|
| 1426 |
+
"eval_steps_per_second": 0.065,
|
| 1427 |
+
"step": 900
|
| 1428 |
+
},
|
| 1429 |
+
{
|
| 1430 |
+
"epoch": 0.95,
|
| 1431 |
+
"learning_rate": 3.378064801637687e-08,
|
| 1432 |
+
"logps/chosen": -37312.3515625,
|
| 1433 |
+
"logps/rejected": -31717.978515625,
|
| 1434 |
+
"loss": 0.0123,
|
| 1435 |
+
"neglected": 586.0,
|
| 1436 |
+
"rewards/accuracies": 0.4375,
|
| 1437 |
+
"rewards/chosen": -370.2989196777344,
|
| 1438 |
+
"rewards/margins": -55.649620056152344,
|
| 1439 |
+
"rewards/rejected": -314.6493225097656,
|
| 1440 |
+
"selected": 0.0,
|
| 1441 |
+
"step": 910
|
| 1442 |
+
},
|
| 1443 |
+
{
|
| 1444 |
+
"epoch": 0.96,
|
| 1445 |
+
"learning_rate": 2.0453443778310766e-08,
|
| 1446 |
+
"logps/chosen": -38356.51953125,
|
| 1447 |
+
"logps/rejected": -32435.177734375,
|
| 1448 |
+
"loss": 0.0172,
|
| 1449 |
+
"neglected": 746.0,
|
| 1450 |
+
"rewards/accuracies": 0.36250001192092896,
|
| 1451 |
+
"rewards/chosen": -380.7662048339844,
|
| 1452 |
+
"rewards/margins": -58.904090881347656,
|
| 1453 |
+
"rewards/rejected": -321.86212158203125,
|
| 1454 |
+
"selected": 0.0,
|
| 1455 |
+
"step": 920
|
| 1456 |
+
},
|
| 1457 |
+
{
|
| 1458 |
+
"epoch": 0.97,
|
| 1459 |
+
"learning_rate": 1.0442413283435759e-08,
|
| 1460 |
+
"logps/chosen": -37030.9140625,
|
| 1461 |
+
"logps/rejected": -29036.62109375,
|
| 1462 |
+
"loss": 0.0174,
|
| 1463 |
+
"neglected": 906.0,
|
| 1464 |
+
"rewards/accuracies": 0.45625001192092896,
|
| 1465 |
+
"rewards/chosen": -367.7157897949219,
|
| 1466 |
+
"rewards/margins": -79.69428253173828,
|
| 1467 |
+
"rewards/rejected": -288.02154541015625,
|
| 1468 |
+
"selected": 0.0,
|
| 1469 |
+
"step": 930
|
| 1470 |
+
},
|
| 1471 |
+
{
|
| 1472 |
+
"epoch": 0.98,
|
| 1473 |
+
"learning_rate": 3.760945397705828e-09,
|
| 1474 |
+
"logps/chosen": -39078.89453125,
|
| 1475 |
+
"logps/rejected": -32606.041015625,
|
| 1476 |
+
"loss": 0.0123,
|
| 1477 |
+
"neglected": 1066.0,
|
| 1478 |
+
"rewards/accuracies": 0.4312500059604645,
|
| 1479 |
+
"rewards/chosen": -388.01318359375,
|
| 1480 |
+
"rewards/margins": -64.55225372314453,
|
| 1481 |
+
"rewards/rejected": -323.4609375,
|
| 1482 |
+
"selected": 0.0,
|
| 1483 |
+
"step": 940
|
| 1484 |
+
},
|
| 1485 |
+
{
|
| 1486 |
+
"epoch": 0.99,
|
| 1487 |
+
"learning_rate": 4.1797599220405605e-10,
|
| 1488 |
+
"logps/chosen": -34646.1640625,
|
| 1489 |
+
"logps/rejected": -31544.458984375,
|
| 1490 |
+
"loss": 0.0168,
|
| 1491 |
+
"neglected": 1226.0,
|
| 1492 |
+
"rewards/accuracies": 0.48750001192092896,
|
| 1493 |
+
"rewards/chosen": -343.90411376953125,
|
| 1494 |
+
"rewards/margins": -30.896869659423828,
|
| 1495 |
+
"rewards/rejected": -313.00726318359375,
|
| 1496 |
+
"selected": 0.0,
|
| 1497 |
+
"step": 950
|
| 1498 |
+
},
|
| 1499 |
+
{
|
| 1500 |
+
"epoch": 1.0,
|
| 1501 |
+
"step": 955,
|
| 1502 |
+
"total_flos": 0.0,
|
| 1503 |
+
"train_loss": 0.11865393293933718,
|
| 1504 |
+
"train_runtime": 42177.7253,
|
| 1505 |
+
"train_samples_per_second": 1.449,
|
| 1506 |
+
"train_steps_per_second": 0.023
|
| 1507 |
+
}
|
| 1508 |
+
],
|
| 1509 |
+
"logging_steps": 10,
|
| 1510 |
+
"max_steps": 955,
|
| 1511 |
+
"num_train_epochs": 1,
|
| 1512 |
+
"save_steps": 100,
|
| 1513 |
+
"total_flos": 0.0,
|
| 1514 |
+
"trial_name": null,
|
| 1515 |
+
"trial_params": null
|
| 1516 |
+
}
|