Commit
•
4af0852
1
Parent(s):
7311b52
Model save
Browse files- README.md +62 -0
- adapter_model.safetensors +1 -1
- all_results.json +8 -0
- emissions.csv +2 -0
- runs/Oct01_02-41-51_gpu4-119-5/events.out.tfevents.1727714563.gpu4-119-5.2350551.0 +2 -2
- train_results.json +8 -0
- trainer_state.json +702 -0
README.md
ADDED
|
@@ -0,0 +1,62 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
library_name: peft
|
| 4 |
+
tags:
|
| 5 |
+
- trl
|
| 6 |
+
- dpo
|
| 7 |
+
- generated_from_trainer
|
| 8 |
+
base_model: alignment-handbook/zephyr-7b-sft-full
|
| 9 |
+
model-index:
|
| 10 |
+
- name: phi-2-kto-i0-v2
|
| 11 |
+
results: []
|
| 12 |
+
---
|
| 13 |
+
|
| 14 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 15 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 16 |
+
|
| 17 |
+
# phi-2-kto-i0-v2
|
| 18 |
+
|
| 19 |
+
This model is a fine-tuned version of [alignment-handbook/zephyr-7b-sft-full](https://huggingface.co/alignment-handbook/zephyr-7b-sft-full) on the None dataset.
|
| 20 |
+
|
| 21 |
+
## Model description
|
| 22 |
+
|
| 23 |
+
More information needed
|
| 24 |
+
|
| 25 |
+
## Intended uses & limitations
|
| 26 |
+
|
| 27 |
+
More information needed
|
| 28 |
+
|
| 29 |
+
## Training and evaluation data
|
| 30 |
+
|
| 31 |
+
More information needed
|
| 32 |
+
|
| 33 |
+
## Training procedure
|
| 34 |
+
|
| 35 |
+
### Training hyperparameters
|
| 36 |
+
|
| 37 |
+
The following hyperparameters were used during training:
|
| 38 |
+
- learning_rate: 5e-06
|
| 39 |
+
- train_batch_size: 4
|
| 40 |
+
- eval_batch_size: 4
|
| 41 |
+
- seed: 42
|
| 42 |
+
- distributed_type: multi-GPU
|
| 43 |
+
- num_devices: 4
|
| 44 |
+
- gradient_accumulation_steps: 8
|
| 45 |
+
- total_train_batch_size: 128
|
| 46 |
+
- total_eval_batch_size: 16
|
| 47 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 48 |
+
- lr_scheduler_type: cosine
|
| 49 |
+
- lr_scheduler_warmup_ratio: 0.1
|
| 50 |
+
- num_epochs: 1
|
| 51 |
+
|
| 52 |
+
### Training results
|
| 53 |
+
|
| 54 |
+
|
| 55 |
+
|
| 56 |
+
### Framework versions
|
| 57 |
+
|
| 58 |
+
- PEFT 0.7.1
|
| 59 |
+
- Transformers 4.36.2
|
| 60 |
+
- Pytorch 2.1.2
|
| 61 |
+
- Datasets 2.14.6
|
| 62 |
+
- Tokenizers 0.15.2
|
adapter_model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 151021328
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2c0ce51029bfb7b18b4686a27bf88b772be099319635ab957846b1daefedf920
|
| 3 |
size 151021328
|
all_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"train_loss": 0.38682256204777044,
|
| 4 |
+
"train_runtime": 17349.9683,
|
| 5 |
+
"train_samples": 61135,
|
| 6 |
+
"train_samples_per_second": 3.524,
|
| 7 |
+
"train_steps_per_second": 0.027
|
| 8 |
+
}
|
emissions.csv
ADDED
|
@@ -0,0 +1,2 @@
|
|
|
|
|
|
|
|
|
|
| 1 |
+
timestamp,project_name,run_id,duration,emissions,emissions_rate,cpu_power,gpu_power,ram_power,cpu_energy,gpu_energy,ram_energy,energy_consumed,country_name,country_iso_code,region,cloud_provider,cloud_region,os,python_version,codecarbon_version,cpu_count,cpu_model,gpu_count,gpu_model,longitude,latitude,ram_total_size,tracking_mode,on_cloud,pue
|
| 2 |
+
2024-10-01T07:31:53,codecarbon,2c979beb-03ea-46da-b77a-c2b317ee11a7,17350.095675706863,3.5915123695560958,0.00020700245328242396,42.5,785.197,188.74309015274048,0.20482500590946942,5.649975200325763,0.9057318429925282,6.760532049227754,Australia,AUS,new south wales,,,Linux-5.15.0-84-generic-x86_64-with-glibc2.35,3.10.14,2.2.3,32,Intel(R) Xeon(R) W-3335 CPU @ 3.40GHz,4,4 x NVIDIA GeForce RTX 4090,151.1919,-33.887,503.3149070739746,machine,N,1.0
|
runs/Oct01_02-41-51_gpu4-119-5/events.out.tfevents.1727714563.gpu4-119-5.2350551.0
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e5909fb7c834761f01670575d344f3d805fbd30c2de31cb563435147d13bb973
|
| 3 |
+
size 28361
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"train_loss": 0.38682256204777044,
|
| 4 |
+
"train_runtime": 17349.9683,
|
| 5 |
+
"train_samples": 61135,
|
| 6 |
+
"train_samples_per_second": 3.524,
|
| 7 |
+
"train_steps_per_second": 0.027
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,702 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.998691442030882,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 477,
|
| 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": 1.0416666666666667e-07,
|
| 14 |
+
"logits/chosen": -2.7967724800109863,
|
| 15 |
+
"logits/rejected": -2.805750846862793,
|
| 16 |
+
"logps/chosen": -270.3558654785156,
|
| 17 |
+
"logps/rejected": -243.22396850585938,
|
| 18 |
+
"loss": 0.5,
|
| 19 |
+
"rewards/accuracies": 0.0,
|
| 20 |
+
"rewards/chosen": 0.0,
|
| 21 |
+
"rewards/margins": 0.0,
|
| 22 |
+
"rewards/rejected": 0.0,
|
| 23 |
+
"step": 1
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.02,
|
| 27 |
+
"learning_rate": 1.0416666666666667e-06,
|
| 28 |
+
"logits/chosen": -2.671762466430664,
|
| 29 |
+
"logits/rejected": -2.713010549545288,
|
| 30 |
+
"logps/chosen": -228.7700958251953,
|
| 31 |
+
"logps/rejected": -213.14617919921875,
|
| 32 |
+
"loss": 0.5,
|
| 33 |
+
"rewards/accuracies": 0.3680555522441864,
|
| 34 |
+
"rewards/chosen": 6.792499334551394e-06,
|
| 35 |
+
"rewards/margins": -1.524862182122888e-05,
|
| 36 |
+
"rewards/rejected": 2.204112388426438e-05,
|
| 37 |
+
"step": 10
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.04,
|
| 41 |
+
"learning_rate": 2.0833333333333334e-06,
|
| 42 |
+
"logits/chosen": -2.7296969890594482,
|
| 43 |
+
"logits/rejected": -2.7327723503112793,
|
| 44 |
+
"logps/chosen": -242.0876922607422,
|
| 45 |
+
"logps/rejected": -224.0583038330078,
|
| 46 |
+
"loss": 0.5,
|
| 47 |
+
"rewards/accuracies": 0.4781250059604645,
|
| 48 |
+
"rewards/chosen": 0.00022381536837201566,
|
| 49 |
+
"rewards/margins": 7.298699347302318e-05,
|
| 50 |
+
"rewards/rejected": 0.0001508283894509077,
|
| 51 |
+
"step": 20
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.06,
|
| 55 |
+
"learning_rate": 3.125e-06,
|
| 56 |
+
"logits/chosen": -2.7316231727600098,
|
| 57 |
+
"logits/rejected": -2.7253520488739014,
|
| 58 |
+
"logps/chosen": -253.94223022460938,
|
| 59 |
+
"logps/rejected": -235.99728393554688,
|
| 60 |
+
"loss": 0.4999,
|
| 61 |
+
"rewards/accuracies": 0.512499988079071,
|
| 62 |
+
"rewards/chosen": 0.000914489384740591,
|
| 63 |
+
"rewards/margins": 0.0003135653678327799,
|
| 64 |
+
"rewards/rejected": 0.0006009239004924893,
|
| 65 |
+
"step": 30
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.08,
|
| 69 |
+
"learning_rate": 4.166666666666667e-06,
|
| 70 |
+
"logits/chosen": -2.6916933059692383,
|
| 71 |
+
"logits/rejected": -2.690849781036377,
|
| 72 |
+
"logps/chosen": -252.69357299804688,
|
| 73 |
+
"logps/rejected": -234.3971710205078,
|
| 74 |
+
"loss": 0.4998,
|
| 75 |
+
"rewards/accuracies": 0.5874999761581421,
|
| 76 |
+
"rewards/chosen": 0.0020142460707575083,
|
| 77 |
+
"rewards/margins": 0.0008797285263426602,
|
| 78 |
+
"rewards/rejected": 0.0011345174862071872,
|
| 79 |
+
"step": 40
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.1,
|
| 83 |
+
"learning_rate": 4.999731868769027e-06,
|
| 84 |
+
"logits/chosen": -2.6373677253723145,
|
| 85 |
+
"logits/rejected": -2.641324281692505,
|
| 86 |
+
"logps/chosen": -224.9910430908203,
|
| 87 |
+
"logps/rejected": -220.6387939453125,
|
| 88 |
+
"loss": 0.4996,
|
| 89 |
+
"rewards/accuracies": 0.59375,
|
| 90 |
+
"rewards/chosen": 0.003733439836651087,
|
| 91 |
+
"rewards/margins": 0.0014786701649427414,
|
| 92 |
+
"rewards/rejected": 0.0022547696717083454,
|
| 93 |
+
"step": 50
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.13,
|
| 97 |
+
"learning_rate": 4.9903533134293035e-06,
|
| 98 |
+
"logits/chosen": -2.6569573879241943,
|
| 99 |
+
"logits/rejected": -2.6791322231292725,
|
| 100 |
+
"logps/chosen": -221.05709838867188,
|
| 101 |
+
"logps/rejected": -203.5032501220703,
|
| 102 |
+
"loss": 0.4994,
|
| 103 |
+
"rewards/accuracies": 0.606249988079071,
|
| 104 |
+
"rewards/chosen": 0.005443854723125696,
|
| 105 |
+
"rewards/margins": 0.0024864792358130217,
|
| 106 |
+
"rewards/rejected": 0.0029573754873126745,
|
| 107 |
+
"step": 60
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.15,
|
| 111 |
+
"learning_rate": 4.967625656594782e-06,
|
| 112 |
+
"logits/chosen": -2.707890748977661,
|
| 113 |
+
"logits/rejected": -2.6958038806915283,
|
| 114 |
+
"logps/chosen": -227.9226531982422,
|
| 115 |
+
"logps/rejected": -209.75619506835938,
|
| 116 |
+
"loss": 0.4992,
|
| 117 |
+
"rewards/accuracies": 0.6031249761581421,
|
| 118 |
+
"rewards/chosen": 0.006226530764251947,
|
| 119 |
+
"rewards/margins": 0.0026825761888176203,
|
| 120 |
+
"rewards/rejected": 0.0035439543426036835,
|
| 121 |
+
"step": 70
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.17,
|
| 125 |
+
"learning_rate": 4.93167072587771e-06,
|
| 126 |
+
"logits/chosen": -2.691779375076294,
|
| 127 |
+
"logits/rejected": -2.6711440086364746,
|
| 128 |
+
"logps/chosen": -218.3474578857422,
|
| 129 |
+
"logps/rejected": -207.0499725341797,
|
| 130 |
+
"loss": 0.499,
|
| 131 |
+
"rewards/accuracies": 0.609375,
|
| 132 |
+
"rewards/chosen": 0.007510344497859478,
|
| 133 |
+
"rewards/margins": 0.003996443003416061,
|
| 134 |
+
"rewards/rejected": 0.0035139017272740602,
|
| 135 |
+
"step": 80
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.19,
|
| 139 |
+
"learning_rate": 4.882681251368549e-06,
|
| 140 |
+
"logits/chosen": -2.6946892738342285,
|
| 141 |
+
"logits/rejected": -2.6706433296203613,
|
| 142 |
+
"logps/chosen": -233.44540405273438,
|
| 143 |
+
"logps/rejected": -225.4489288330078,
|
| 144 |
+
"loss": 0.4988,
|
| 145 |
+
"rewards/accuracies": 0.606249988079071,
|
| 146 |
+
"rewards/chosen": 0.007306075654923916,
|
| 147 |
+
"rewards/margins": 0.004931028466671705,
|
| 148 |
+
"rewards/rejected": 0.002375046955421567,
|
| 149 |
+
"step": 90
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.21,
|
| 153 |
+
"learning_rate": 4.8209198325401815e-06,
|
| 154 |
+
"logits/chosen": -2.741804838180542,
|
| 155 |
+
"logits/rejected": -2.722029447555542,
|
| 156 |
+
"logps/chosen": -244.57785034179688,
|
| 157 |
+
"logps/rejected": -229.181884765625,
|
| 158 |
+
"loss": 0.4986,
|
| 159 |
+
"rewards/accuracies": 0.675000011920929,
|
| 160 |
+
"rewards/chosen": 0.00877557136118412,
|
| 161 |
+
"rewards/margins": 0.007159523665904999,
|
| 162 |
+
"rewards/rejected": 0.0016160461818799376,
|
| 163 |
+
"step": 100
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.23,
|
| 167 |
+
"learning_rate": 4.746717530629565e-06,
|
| 168 |
+
"logits/chosen": -2.6642231941223145,
|
| 169 |
+
"logits/rejected": -2.682945966720581,
|
| 170 |
+
"logps/chosen": -227.15109252929688,
|
| 171 |
+
"logps/rejected": -215.261474609375,
|
| 172 |
+
"loss": 0.4986,
|
| 173 |
+
"rewards/accuracies": 0.6000000238418579,
|
| 174 |
+
"rewards/chosen": 0.007662790361791849,
|
| 175 |
+
"rewards/margins": 0.007442783564329147,
|
| 176 |
+
"rewards/rejected": 0.00022000684111844748,
|
| 177 |
+
"step": 110
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.25,
|
| 181 |
+
"learning_rate": 4.660472094042121e-06,
|
| 182 |
+
"logits/chosen": -2.6481804847717285,
|
| 183 |
+
"logits/rejected": -2.6557509899139404,
|
| 184 |
+
"logps/chosen": -243.96139526367188,
|
| 185 |
+
"logps/rejected": -208.84384155273438,
|
| 186 |
+
"loss": 0.4984,
|
| 187 |
+
"rewards/accuracies": 0.5874999761581421,
|
| 188 |
+
"rewards/chosen": 0.007970081642270088,
|
| 189 |
+
"rewards/margins": 0.007513949181884527,
|
| 190 |
+
"rewards/rejected": 0.00045613135444000363,
|
| 191 |
+
"step": 120
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.27,
|
| 195 |
+
"learning_rate": 4.5626458262912745e-06,
|
| 196 |
+
"logits/chosen": -2.607055187225342,
|
| 197 |
+
"logits/rejected": -2.6165382862091064,
|
| 198 |
+
"logps/chosen": -214.7068634033203,
|
| 199 |
+
"logps/rejected": -214.4553680419922,
|
| 200 |
+
"loss": 0.4985,
|
| 201 |
+
"rewards/accuracies": 0.5625,
|
| 202 |
+
"rewards/chosen": 0.007824478670954704,
|
| 203 |
+
"rewards/margins": 0.007803040556609631,
|
| 204 |
+
"rewards/rejected": 2.143829988199286e-05,
|
| 205 |
+
"step": 130
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.29,
|
| 209 |
+
"learning_rate": 4.453763107901676e-06,
|
| 210 |
+
"logits/chosen": -2.662113666534424,
|
| 211 |
+
"logits/rejected": -2.6899476051330566,
|
| 212 |
+
"logps/chosen": -230.07870483398438,
|
| 213 |
+
"logps/rejected": -218.9852752685547,
|
| 214 |
+
"loss": 0.4981,
|
| 215 |
+
"rewards/accuracies": 0.675000011920929,
|
| 216 |
+
"rewards/chosen": 0.008939255960285664,
|
| 217 |
+
"rewards/margins": 0.00983688049018383,
|
| 218 |
+
"rewards/rejected": -0.000897624995559454,
|
| 219 |
+
"step": 140
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.31,
|
| 223 |
+
"learning_rate": 4.33440758555951e-06,
|
| 224 |
+
"logits/chosen": -2.6915247440338135,
|
| 225 |
+
"logits/rejected": -2.684671640396118,
|
| 226 |
+
"logps/chosen": -239.7102813720703,
|
| 227 |
+
"logps/rejected": -196.85520935058594,
|
| 228 |
+
"loss": 0.498,
|
| 229 |
+
"rewards/accuracies": 0.625,
|
| 230 |
+
"rewards/chosen": 0.006214095279574394,
|
| 231 |
+
"rewards/margins": 0.008828171528875828,
|
| 232 |
+
"rewards/rejected": -0.0026140757836401463,
|
| 233 |
+
"step": 150
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.33,
|
| 237 |
+
"learning_rate": 4.205219043576955e-06,
|
| 238 |
+
"logits/chosen": -2.6351287364959717,
|
| 239 |
+
"logits/rejected": -2.6126668453216553,
|
| 240 |
+
"logps/chosen": -217.2763214111328,
|
| 241 |
+
"logps/rejected": -232.38308715820312,
|
| 242 |
+
"loss": 0.4977,
|
| 243 |
+
"rewards/accuracies": 0.612500011920929,
|
| 244 |
+
"rewards/chosen": 0.005867967382073402,
|
| 245 |
+
"rewards/margins": 0.012363599613308907,
|
| 246 |
+
"rewards/rejected": -0.006495633628219366,
|
| 247 |
+
"step": 160
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.36,
|
| 251 |
+
"learning_rate": 4.066889974440757e-06,
|
| 252 |
+
"logits/chosen": -2.651954412460327,
|
| 253 |
+
"logits/rejected": -2.6270413398742676,
|
| 254 |
+
"logps/chosen": -230.48202514648438,
|
| 255 |
+
"logps/rejected": -207.2578887939453,
|
| 256 |
+
"loss": 0.4981,
|
| 257 |
+
"rewards/accuracies": 0.5687500238418579,
|
| 258 |
+
"rewards/chosen": 0.005082262679934502,
|
| 259 |
+
"rewards/margins": 0.009124400094151497,
|
| 260 |
+
"rewards/rejected": -0.004042136482894421,
|
| 261 |
+
"step": 170
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.38,
|
| 265 |
+
"learning_rate": 3.92016186682789e-06,
|
| 266 |
+
"logits/chosen": -2.623335838317871,
|
| 267 |
+
"logits/rejected": -2.6311707496643066,
|
| 268 |
+
"logps/chosen": -240.3880157470703,
|
| 269 |
+
"logps/rejected": -239.96383666992188,
|
| 270 |
+
"loss": 0.4978,
|
| 271 |
+
"rewards/accuracies": 0.65625,
|
| 272 |
+
"rewards/chosen": 0.004731752909719944,
|
| 273 |
+
"rewards/margins": 0.013921832665801048,
|
| 274 |
+
"rewards/rejected": -0.009190080687403679,
|
| 275 |
+
"step": 180
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.4,
|
| 279 |
+
"learning_rate": 3.7658212309857576e-06,
|
| 280 |
+
"logits/chosen": -2.6556308269500732,
|
| 281 |
+
"logits/rejected": -2.635357618331909,
|
| 282 |
+
"logps/chosen": -227.9666748046875,
|
| 283 |
+
"logps/rejected": -235.2104949951172,
|
| 284 |
+
"loss": 0.4976,
|
| 285 |
+
"rewards/accuracies": 0.6343749761581421,
|
| 286 |
+
"rewards/chosen": 0.005550178233534098,
|
| 287 |
+
"rewards/margins": 0.015345620922744274,
|
| 288 |
+
"rewards/rejected": -0.009795443154871464,
|
| 289 |
+
"step": 190
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.42,
|
| 293 |
+
"learning_rate": 3.604695382782159e-06,
|
| 294 |
+
"logits/chosen": -2.613764524459839,
|
| 295 |
+
"logits/rejected": -2.5981905460357666,
|
| 296 |
+
"logps/chosen": -244.93527221679688,
|
| 297 |
+
"logps/rejected": -219.70816040039062,
|
| 298 |
+
"loss": 0.4975,
|
| 299 |
+
"rewards/accuracies": 0.621874988079071,
|
| 300 |
+
"rewards/chosen": 0.005905141122639179,
|
| 301 |
+
"rewards/margins": 0.01656787097454071,
|
| 302 |
+
"rewards/rejected": -0.010662728920578957,
|
| 303 |
+
"step": 200
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.44,
|
| 307 |
+
"learning_rate": 3.437648009023905e-06,
|
| 308 |
+
"logits/chosen": -2.6281652450561523,
|
| 309 |
+
"logits/rejected": -2.6158878803253174,
|
| 310 |
+
"logps/chosen": -230.68093872070312,
|
| 311 |
+
"logps/rejected": -229.5253448486328,
|
| 312 |
+
"loss": 0.4974,
|
| 313 |
+
"rewards/accuracies": 0.5843750238418579,
|
| 314 |
+
"rewards/chosen": 0.0013032301794737577,
|
| 315 |
+
"rewards/margins": 0.012972685508430004,
|
| 316 |
+
"rewards/rejected": -0.011669456958770752,
|
| 317 |
+
"step": 210
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.46,
|
| 321 |
+
"learning_rate": 3.265574537815398e-06,
|
| 322 |
+
"logits/chosen": -2.6179089546203613,
|
| 323 |
+
"logits/rejected": -2.600205898284912,
|
| 324 |
+
"logps/chosen": -220.5909881591797,
|
| 325 |
+
"logps/rejected": -219.4180450439453,
|
| 326 |
+
"loss": 0.4975,
|
| 327 |
+
"rewards/accuracies": 0.574999988079071,
|
| 328 |
+
"rewards/chosen": 0.00289526185952127,
|
| 329 |
+
"rewards/margins": 0.01316472701728344,
|
| 330 |
+
"rewards/rejected": -0.0102694658562541,
|
| 331 |
+
"step": 220
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.48,
|
| 335 |
+
"learning_rate": 3.089397338773569e-06,
|
| 336 |
+
"logits/chosen": -2.5429725646972656,
|
| 337 |
+
"logits/rejected": -2.5193886756896973,
|
| 338 |
+
"logps/chosen": -219.8972930908203,
|
| 339 |
+
"logps/rejected": -224.34597778320312,
|
| 340 |
+
"loss": 0.4974,
|
| 341 |
+
"rewards/accuracies": 0.578125,
|
| 342 |
+
"rewards/chosen": 0.001055281376466155,
|
| 343 |
+
"rewards/margins": 0.015370063483715057,
|
| 344 |
+
"rewards/rejected": -0.014314780943095684,
|
| 345 |
+
"step": 230
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.5,
|
| 349 |
+
"learning_rate": 2.9100607788275547e-06,
|
| 350 |
+
"logits/chosen": -2.5410211086273193,
|
| 351 |
+
"logits/rejected": -2.5228562355041504,
|
| 352 |
+
"logps/chosen": -221.77243041992188,
|
| 353 |
+
"logps/rejected": -222.72091674804688,
|
| 354 |
+
"loss": 0.4972,
|
| 355 |
+
"rewards/accuracies": 0.5562499761581421,
|
| 356 |
+
"rewards/chosen": -0.0065743401646614075,
|
| 357 |
+
"rewards/margins": 0.014264288358390331,
|
| 358 |
+
"rewards/rejected": -0.020838629454374313,
|
| 359 |
+
"step": 240
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.52,
|
| 363 |
+
"learning_rate": 2.72852616010567e-06,
|
| 364 |
+
"logits/chosen": -2.5719101428985596,
|
| 365 |
+
"logits/rejected": -2.5223493576049805,
|
| 366 |
+
"logps/chosen": -226.51950073242188,
|
| 367 |
+
"logps/rejected": -222.7196502685547,
|
| 368 |
+
"loss": 0.4969,
|
| 369 |
+
"rewards/accuracies": 0.590624988079071,
|
| 370 |
+
"rewards/chosen": -0.0035572790075093508,
|
| 371 |
+
"rewards/margins": 0.023138266056776047,
|
| 372 |
+
"rewards/rejected": -0.02669554576277733,
|
| 373 |
+
"step": 250
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.54,
|
| 377 |
+
"learning_rate": 2.5457665670441937e-06,
|
| 378 |
+
"logits/chosen": -2.6053082942962646,
|
| 379 |
+
"logits/rejected": -2.5733938217163086,
|
| 380 |
+
"logps/chosen": -249.0303955078125,
|
| 381 |
+
"logps/rejected": -240.6870574951172,
|
| 382 |
+
"loss": 0.4962,
|
| 383 |
+
"rewards/accuracies": 0.6156250238418579,
|
| 384 |
+
"rewards/chosen": -0.005832755006849766,
|
| 385 |
+
"rewards/margins": 0.026431718841195107,
|
| 386 |
+
"rewards/rejected": -0.03226447105407715,
|
| 387 |
+
"step": 260
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.57,
|
| 391 |
+
"learning_rate": 2.3627616503391813e-06,
|
| 392 |
+
"logits/chosen": -2.6097915172576904,
|
| 393 |
+
"logits/rejected": -2.565622329711914,
|
| 394 |
+
"logps/chosen": -272.3103942871094,
|
| 395 |
+
"logps/rejected": -283.4029846191406,
|
| 396 |
+
"loss": 0.4962,
|
| 397 |
+
"rewards/accuracies": 0.637499988079071,
|
| 398 |
+
"rewards/chosen": -0.009318219497799873,
|
| 399 |
+
"rewards/margins": 0.03199902921915054,
|
| 400 |
+
"rewards/rejected": -0.041317250579595566,
|
| 401 |
+
"step": 270
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.59,
|
| 405 |
+
"learning_rate": 2.1804923757009885e-06,
|
| 406 |
+
"logits/chosen": -2.5360941886901855,
|
| 407 |
+
"logits/rejected": -2.502972364425659,
|
| 408 |
+
"logps/chosen": -253.5753173828125,
|
| 409 |
+
"logps/rejected": -242.55581665039062,
|
| 410 |
+
"loss": 0.4965,
|
| 411 |
+
"rewards/accuracies": 0.5874999761581421,
|
| 412 |
+
"rewards/chosen": -0.025225955992937088,
|
| 413 |
+
"rewards/margins": 0.025721151381731033,
|
| 414 |
+
"rewards/rejected": -0.05094710737466812,
|
| 415 |
+
"step": 280
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.61,
|
| 419 |
+
"learning_rate": 1.9999357655598894e-06,
|
| 420 |
+
"logits/chosen": -2.5319244861602783,
|
| 421 |
+
"logits/rejected": -2.465296983718872,
|
| 422 |
+
"logps/chosen": -268.68438720703125,
|
| 423 |
+
"logps/rejected": -286.90582275390625,
|
| 424 |
+
"loss": 0.4959,
|
| 425 |
+
"rewards/accuracies": 0.612500011920929,
|
| 426 |
+
"rewards/chosen": -0.03240751475095749,
|
| 427 |
+
"rewards/margins": 0.035424619913101196,
|
| 428 |
+
"rewards/rejected": -0.06783213466405869,
|
| 429 |
+
"step": 290
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.63,
|
| 433 |
+
"learning_rate": 1.8220596619089576e-06,
|
| 434 |
+
"logits/chosen": -2.4394478797912598,
|
| 435 |
+
"logits/rejected": -2.4182257652282715,
|
| 436 |
+
"logps/chosen": -272.16973876953125,
|
| 437 |
+
"logps/rejected": -293.65020751953125,
|
| 438 |
+
"loss": 0.4947,
|
| 439 |
+
"rewards/accuracies": 0.5562499761581421,
|
| 440 |
+
"rewards/chosen": -0.03661135956645012,
|
| 441 |
+
"rewards/margins": 0.03990130499005318,
|
| 442 |
+
"rewards/rejected": -0.07651267200708389,
|
| 443 |
+
"step": 300
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.65,
|
| 447 |
+
"learning_rate": 1.647817538357072e-06,
|
| 448 |
+
"logits/chosen": -2.297726631164551,
|
| 449 |
+
"logits/rejected": -2.2101216316223145,
|
| 450 |
+
"logps/chosen": -390.9844665527344,
|
| 451 |
+
"logps/rejected": -431.4691467285156,
|
| 452 |
+
"loss": 0.4919,
|
| 453 |
+
"rewards/accuracies": 0.528124988079071,
|
| 454 |
+
"rewards/chosen": -0.15482836961746216,
|
| 455 |
+
"rewards/margins": 0.07953239977359772,
|
| 456 |
+
"rewards/rejected": -0.23436078429222107,
|
| 457 |
+
"step": 310
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.67,
|
| 461 |
+
"learning_rate": 1.4781433892011132e-06,
|
| 462 |
+
"logits/chosen": -1.8453972339630127,
|
| 463 |
+
"logits/rejected": -1.7317909002304077,
|
| 464 |
+
"logps/chosen": -681.6414184570312,
|
| 465 |
+
"logps/rejected": -987.8673706054688,
|
| 466 |
+
"loss": 0.4842,
|
| 467 |
+
"rewards/accuracies": 0.5531250238418579,
|
| 468 |
+
"rewards/chosen": -0.4435897767543793,
|
| 469 |
+
"rewards/margins": 0.3242936134338379,
|
| 470 |
+
"rewards/rejected": -0.7678834199905396,
|
| 471 |
+
"step": 320
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.69,
|
| 475 |
+
"learning_rate": 1.3139467229135999e-06,
|
| 476 |
+
"logits/chosen": -1.9067537784576416,
|
| 477 |
+
"logits/rejected": -1.8014500141143799,
|
| 478 |
+
"logps/chosen": -885.9420166015625,
|
| 479 |
+
"logps/rejected": -1086.572509765625,
|
| 480 |
+
"loss": 0.482,
|
| 481 |
+
"rewards/accuracies": 0.48750001192092896,
|
| 482 |
+
"rewards/chosen": -0.6387772560119629,
|
| 483 |
+
"rewards/margins": 0.23005250096321106,
|
| 484 |
+
"rewards/rejected": -0.8688297271728516,
|
| 485 |
+
"step": 330
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.71,
|
| 489 |
+
"learning_rate": 1.1561076868822756e-06,
|
| 490 |
+
"logits/chosen": -2.052692413330078,
|
| 491 |
+
"logits/rejected": -1.9470455646514893,
|
| 492 |
+
"logps/chosen": -739.042236328125,
|
| 493 |
+
"logps/rejected": -992.4461059570312,
|
| 494 |
+
"loss": 0.4809,
|
| 495 |
+
"rewards/accuracies": 0.559374988079071,
|
| 496 |
+
"rewards/chosen": -0.49481409788131714,
|
| 497 |
+
"rewards/margins": 0.2876318097114563,
|
| 498 |
+
"rewards/rejected": -0.7824459075927734,
|
| 499 |
+
"step": 340
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.73,
|
| 503 |
+
"learning_rate": 1.0054723495346484e-06,
|
| 504 |
+
"logits/chosen": -1.9866969585418701,
|
| 505 |
+
"logits/rejected": -1.8636195659637451,
|
| 506 |
+
"logps/chosen": -883.5921630859375,
|
| 507 |
+
"logps/rejected": -1097.7490234375,
|
| 508 |
+
"loss": 0.4832,
|
| 509 |
+
"rewards/accuracies": 0.5406249761581421,
|
| 510 |
+
"rewards/chosen": -0.658557116985321,
|
| 511 |
+
"rewards/margins": 0.2439090460538864,
|
| 512 |
+
"rewards/rejected": -0.9024661779403687,
|
| 513 |
+
"step": 350
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.75,
|
| 517 |
+
"learning_rate": 8.628481651367876e-07,
|
| 518 |
+
"logits/chosen": -2.103574275970459,
|
| 519 |
+
"logits/rejected": -1.9690395593643188,
|
| 520 |
+
"logps/chosen": -854.9332885742188,
|
| 521 |
+
"logps/rejected": -1295.6673583984375,
|
| 522 |
+
"loss": 0.4805,
|
| 523 |
+
"rewards/accuracies": 0.5874999761581421,
|
| 524 |
+
"rewards/chosen": -0.6121799945831299,
|
| 525 |
+
"rewards/margins": 0.4702150821685791,
|
| 526 |
+
"rewards/rejected": -1.082395076751709,
|
| 527 |
+
"step": 360
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.77,
|
| 531 |
+
"learning_rate": 7.289996455765749e-07,
|
| 532 |
+
"logits/chosen": -2.1500675678253174,
|
| 533 |
+
"logits/rejected": -2.044593572616577,
|
| 534 |
+
"logps/chosen": -781.8174438476562,
|
| 535 |
+
"logps/rejected": -1134.3828125,
|
| 536 |
+
"loss": 0.4804,
|
| 537 |
+
"rewards/accuracies": 0.49687498807907104,
|
| 538 |
+
"rewards/chosen": -0.5213514566421509,
|
| 539 |
+
"rewards/margins": 0.39176231622695923,
|
| 540 |
+
"rewards/rejected": -0.9131137132644653,
|
| 541 |
+
"step": 370
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.8,
|
| 545 |
+
"learning_rate": 6.046442623320145e-07,
|
| 546 |
+
"logits/chosen": -2.1524083614349365,
|
| 547 |
+
"logits/rejected": -2.063347816467285,
|
| 548 |
+
"logps/chosen": -762.1702270507812,
|
| 549 |
+
"logps/rejected": -1244.992431640625,
|
| 550 |
+
"loss": 0.477,
|
| 551 |
+
"rewards/accuracies": 0.6156250238418579,
|
| 552 |
+
"rewards/chosen": -0.5212429761886597,
|
| 553 |
+
"rewards/margins": 0.4973062574863434,
|
| 554 |
+
"rewards/rejected": -1.0185492038726807,
|
| 555 |
+
"step": 380
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.82,
|
| 559 |
+
"learning_rate": 4.904486005914027e-07,
|
| 560 |
+
"logits/chosen": -2.1940252780914307,
|
| 561 |
+
"logits/rejected": -2.106301784515381,
|
| 562 |
+
"logps/chosen": -789.7882690429688,
|
| 563 |
+
"logps/rejected": -1301.745361328125,
|
| 564 |
+
"loss": 0.4771,
|
| 565 |
+
"rewards/accuracies": 0.612500011920929,
|
| 566 |
+
"rewards/chosen": -0.5382066965103149,
|
| 567 |
+
"rewards/margins": 0.5309630632400513,
|
| 568 |
+
"rewards/rejected": -1.0691697597503662,
|
| 569 |
+
"step": 390
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.84,
|
| 573 |
+
"learning_rate": 3.8702478614051353e-07,
|
| 574 |
+
"logits/chosen": -2.1056671142578125,
|
| 575 |
+
"logits/rejected": -2.0197396278381348,
|
| 576 |
+
"logps/chosen": -846.3111572265625,
|
| 577 |
+
"logps/rejected": -1285.551025390625,
|
| 578 |
+
"loss": 0.478,
|
| 579 |
+
"rewards/accuracies": 0.518750011920929,
|
| 580 |
+
"rewards/chosen": -0.6085025072097778,
|
| 581 |
+
"rewards/margins": 0.4548709988594055,
|
| 582 |
+
"rewards/rejected": -1.0633734464645386,
|
| 583 |
+
"step": 400
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.86,
|
| 587 |
+
"learning_rate": 2.9492720416985004e-07,
|
| 588 |
+
"logits/chosen": -2.086305618286133,
|
| 589 |
+
"logits/rejected": -1.996105432510376,
|
| 590 |
+
"logps/chosen": -932.2883911132812,
|
| 591 |
+
"logps/rejected": -1177.2127685546875,
|
| 592 |
+
"loss": 0.4807,
|
| 593 |
+
"rewards/accuracies": 0.534375011920929,
|
| 594 |
+
"rewards/chosen": -0.6851298213005066,
|
| 595 |
+
"rewards/margins": 0.29829445481300354,
|
| 596 |
+
"rewards/rejected": -0.983424186706543,
|
| 597 |
+
"step": 410
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.88,
|
| 601 |
+
"learning_rate": 2.1464952759020857e-07,
|
| 602 |
+
"logits/chosen": -2.1438374519348145,
|
| 603 |
+
"logits/rejected": -2.0642104148864746,
|
| 604 |
+
"logps/chosen": -878.0729370117188,
|
| 605 |
+
"logps/rejected": -1302.420654296875,
|
| 606 |
+
"loss": 0.4798,
|
| 607 |
+
"rewards/accuracies": 0.606249988079071,
|
| 608 |
+
"rewards/chosen": -0.6570709347724915,
|
| 609 |
+
"rewards/margins": 0.4223295748233795,
|
| 610 |
+
"rewards/rejected": -1.0794004201889038,
|
| 611 |
+
"step": 420
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.9,
|
| 615 |
+
"learning_rate": 1.4662207078575685e-07,
|
| 616 |
+
"logits/chosen": -2.1188504695892334,
|
| 617 |
+
"logits/rejected": -2.0422720909118652,
|
| 618 |
+
"logps/chosen": -986.8173828125,
|
| 619 |
+
"logps/rejected": -1470.625244140625,
|
| 620 |
+
"loss": 0.4786,
|
| 621 |
+
"rewards/accuracies": 0.546875,
|
| 622 |
+
"rewards/chosen": -0.7492085695266724,
|
| 623 |
+
"rewards/margins": 0.5096833109855652,
|
| 624 |
+
"rewards/rejected": -1.2588918209075928,
|
| 625 |
+
"step": 430
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.92,
|
| 629 |
+
"learning_rate": 9.120948298936422e-08,
|
| 630 |
+
"logits/chosen": -2.160806894302368,
|
| 631 |
+
"logits/rejected": -2.0456414222717285,
|
| 632 |
+
"logps/chosen": -828.1990356445312,
|
| 633 |
+
"logps/rejected": -1279.345703125,
|
| 634 |
+
"loss": 0.4768,
|
| 635 |
+
"rewards/accuracies": 0.5874999761581421,
|
| 636 |
+
"rewards/chosen": -0.6060695052146912,
|
| 637 |
+
"rewards/margins": 0.45984458923339844,
|
| 638 |
+
"rewards/rejected": -1.0659140348434448,
|
| 639 |
+
"step": 440
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.94,
|
| 643 |
+
"learning_rate": 4.870879364444109e-08,
|
| 644 |
+
"logits/chosen": -2.1377763748168945,
|
| 645 |
+
"logits/rejected": -2.0168356895446777,
|
| 646 |
+
"logps/chosen": -823.4695434570312,
|
| 647 |
+
"logps/rejected": -1253.1837158203125,
|
| 648 |
+
"loss": 0.4787,
|
| 649 |
+
"rewards/accuracies": 0.5218750238418579,
|
| 650 |
+
"rewards/chosen": -0.591966986656189,
|
| 651 |
+
"rewards/margins": 0.44343581795692444,
|
| 652 |
+
"rewards/rejected": -1.0354026556015015,
|
| 653 |
+
"step": 450
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.96,
|
| 657 |
+
"learning_rate": 1.93478202307823e-08,
|
| 658 |
+
"logits/chosen": -2.1620395183563232,
|
| 659 |
+
"logits/rejected": -2.011014223098755,
|
| 660 |
+
"logps/chosen": -961.3709106445312,
|
| 661 |
+
"logps/rejected": -1517.798828125,
|
| 662 |
+
"loss": 0.4762,
|
| 663 |
+
"rewards/accuracies": 0.6000000238418579,
|
| 664 |
+
"rewards/chosen": -0.7009885907173157,
|
| 665 |
+
"rewards/margins": 0.592383086681366,
|
| 666 |
+
"rewards/rejected": -1.2933716773986816,
|
| 667 |
+
"step": 460
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.98,
|
| 671 |
+
"learning_rate": 3.283947088983663e-09,
|
| 672 |
+
"logits/chosen": -2.198997735977173,
|
| 673 |
+
"logits/rejected": -2.040200710296631,
|
| 674 |
+
"logps/chosen": -834.0213623046875,
|
| 675 |
+
"logps/rejected": -1428.2811279296875,
|
| 676 |
+
"loss": 0.4748,
|
| 677 |
+
"rewards/accuracies": 0.5718749761581421,
|
| 678 |
+
"rewards/chosen": -0.594894289970398,
|
| 679 |
+
"rewards/margins": 0.6237603425979614,
|
| 680 |
+
"rewards/rejected": -1.2186545133590698,
|
| 681 |
+
"step": 470
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 1.0,
|
| 685 |
+
"step": 477,
|
| 686 |
+
"total_flos": 0.0,
|
| 687 |
+
"train_loss": 0.38682256204777044,
|
| 688 |
+
"train_runtime": 17349.9683,
|
| 689 |
+
"train_samples_per_second": 3.524,
|
| 690 |
+
"train_steps_per_second": 0.027
|
| 691 |
+
}
|
| 692 |
+
],
|
| 693 |
+
"logging_steps": 10,
|
| 694 |
+
"max_steps": 477,
|
| 695 |
+
"num_input_tokens_seen": 0,
|
| 696 |
+
"num_train_epochs": 1,
|
| 697 |
+
"save_steps": 20,
|
| 698 |
+
"total_flos": 0.0,
|
| 699 |
+
"train_batch_size": 4,
|
| 700 |
+
"trial_name": null,
|
| 701 |
+
"trial_params": null
|
| 702 |
+
}
|