Model save
Browse files- README.md +61 -0
- adapter_model.safetensors +1 -1
- all_results.json +8 -0
- emissions.csv +2 -0
- runs/Sep29_00-10-54_gpu4-119-5/events.out.tfevents.1727533197.gpu4-119-5.124621.0 +2 -2
- train_results.json +8 -0
- trainer_state.json +912 -0
README.md
ADDED
|
@@ -0,0 +1,61 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
library_name: peft
|
| 3 |
+
tags:
|
| 4 |
+
- trl
|
| 5 |
+
- dpo
|
| 6 |
+
- generated_from_trainer
|
| 7 |
+
base_model: DUAL-GPO/phi-2-kto-i0-merged
|
| 8 |
+
model-index:
|
| 9 |
+
- name: phi-2-kto-0-20-i1
|
| 10 |
+
results: []
|
| 11 |
+
---
|
| 12 |
+
|
| 13 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 14 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 15 |
+
|
| 16 |
+
# phi-2-kto-0-20-i1
|
| 17 |
+
|
| 18 |
+
This model is a fine-tuned version of [DUAL-GPO/phi-2-kto-i0-merged](https://huggingface.co/DUAL-GPO/phi-2-kto-i0-merged) on the None dataset.
|
| 19 |
+
|
| 20 |
+
## Model description
|
| 21 |
+
|
| 22 |
+
More information needed
|
| 23 |
+
|
| 24 |
+
## Intended uses & limitations
|
| 25 |
+
|
| 26 |
+
More information needed
|
| 27 |
+
|
| 28 |
+
## Training and evaluation data
|
| 29 |
+
|
| 30 |
+
More information needed
|
| 31 |
+
|
| 32 |
+
## Training procedure
|
| 33 |
+
|
| 34 |
+
### Training hyperparameters
|
| 35 |
+
|
| 36 |
+
The following hyperparameters were used during training:
|
| 37 |
+
- learning_rate: 5e-06
|
| 38 |
+
- train_batch_size: 4
|
| 39 |
+
- eval_batch_size: 4
|
| 40 |
+
- seed: 42
|
| 41 |
+
- distributed_type: multi-GPU
|
| 42 |
+
- num_devices: 2
|
| 43 |
+
- gradient_accumulation_steps: 4
|
| 44 |
+
- total_train_batch_size: 32
|
| 45 |
+
- total_eval_batch_size: 8
|
| 46 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 47 |
+
- lr_scheduler_type: cosine
|
| 48 |
+
- lr_scheduler_warmup_ratio: 0.1
|
| 49 |
+
- num_epochs: 1
|
| 50 |
+
|
| 51 |
+
### Training results
|
| 52 |
+
|
| 53 |
+
|
| 54 |
+
|
| 55 |
+
### Framework versions
|
| 56 |
+
|
| 57 |
+
- PEFT 0.7.1
|
| 58 |
+
- Transformers 4.36.2
|
| 59 |
+
- Pytorch 2.1.2
|
| 60 |
+
- Datasets 2.14.6
|
| 61 |
+
- Tokenizers 0.15.2
|
adapter_model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 335579632
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5d5a7118ef5b274a6210af96430efc6a9dd5529150edc6a47d1bfac76872523c
|
| 3 |
size 335579632
|
all_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"train_loss": 0.4739649757385254,
|
| 4 |
+
"train_runtime": 8016.6369,
|
| 5 |
+
"train_samples": 20000,
|
| 6 |
+
"train_samples_per_second": 2.495,
|
| 7 |
+
"train_steps_per_second": 0.078
|
| 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 |
+
2024-09-29T02:33:34,codecarbon,8e6e2722-1b9b-4deb-bf00-adf70582a21e,8016.799459695816,1.0154088204292226,0.00012666012484585098,42.5,228.80100000000004,188.74309015274048,0.09464160994788015,1.3980325706075303,0.4186943926868365,1.911368573242245,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/Sep29_00-10-54_gpu4-119-5/events.out.tfevents.1727533197.gpu4-119-5.124621.0
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:ebc824ea36ebfdc37ff7e960cad18daa16985eab1f024dbdb04d96ca9fafb7b3
|
| 3 |
+
size 45039
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"train_loss": 0.4739649757385254,
|
| 4 |
+
"train_runtime": 8016.6369,
|
| 5 |
+
"train_samples": 20000,
|
| 6 |
+
"train_samples_per_second": 2.495,
|
| 7 |
+
"train_steps_per_second": 0.078
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,912 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 1.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 625,
|
| 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": 7.936507936507937e-08,
|
| 14 |
+
"logits/chosen": 0.3974232077598572,
|
| 15 |
+
"logits/rejected": 0.3553540110588074,
|
| 16 |
+
"logps/chosen": -777.8718872070312,
|
| 17 |
+
"logps/rejected": -1263.3857421875,
|
| 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": 7.936507936507937e-07,
|
| 28 |
+
"logits/chosen": 0.2189813256263733,
|
| 29 |
+
"logits/rejected": 0.11798671633005142,
|
| 30 |
+
"logps/chosen": -1287.5775146484375,
|
| 31 |
+
"logps/rejected": -2137.94189453125,
|
| 32 |
+
"loss": 0.4998,
|
| 33 |
+
"rewards/accuracies": 0.4236111044883728,
|
| 34 |
+
"rewards/chosen": 0.001081566559150815,
|
| 35 |
+
"rewards/margins": 0.002832952421158552,
|
| 36 |
+
"rewards/rejected": -0.0017513858620077372,
|
| 37 |
+
"step": 10
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.03,
|
| 41 |
+
"learning_rate": 1.5873015873015873e-06,
|
| 42 |
+
"logits/chosen": 0.254694402217865,
|
| 43 |
+
"logits/rejected": 0.17226830124855042,
|
| 44 |
+
"logps/chosen": -1060.751708984375,
|
| 45 |
+
"logps/rejected": -1972.5423583984375,
|
| 46 |
+
"loss": 0.4987,
|
| 47 |
+
"rewards/accuracies": 0.574999988079071,
|
| 48 |
+
"rewards/chosen": -0.01862182281911373,
|
| 49 |
+
"rewards/margins": 0.011560038663446903,
|
| 50 |
+
"rewards/rejected": -0.03018186055123806,
|
| 51 |
+
"step": 20
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.05,
|
| 55 |
+
"learning_rate": 2.380952380952381e-06,
|
| 56 |
+
"logits/chosen": 0.24693536758422852,
|
| 57 |
+
"logits/rejected": 0.14691275358200073,
|
| 58 |
+
"logps/chosen": -1164.8314208984375,
|
| 59 |
+
"logps/rejected": -2265.688232421875,
|
| 60 |
+
"loss": 0.4942,
|
| 61 |
+
"rewards/accuracies": 0.5375000238418579,
|
| 62 |
+
"rewards/chosen": -0.06745004653930664,
|
| 63 |
+
"rewards/margins": 0.06839548051357269,
|
| 64 |
+
"rewards/rejected": -0.13584552705287933,
|
| 65 |
+
"step": 30
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.06,
|
| 69 |
+
"learning_rate": 3.1746031746031746e-06,
|
| 70 |
+
"logits/chosen": 0.2351571023464203,
|
| 71 |
+
"logits/rejected": 0.08826713263988495,
|
| 72 |
+
"logps/chosen": -1165.0933837890625,
|
| 73 |
+
"logps/rejected": -2538.54443359375,
|
| 74 |
+
"loss": 0.4882,
|
| 75 |
+
"rewards/accuracies": 0.5249999761581421,
|
| 76 |
+
"rewards/chosen": -0.18789814412593842,
|
| 77 |
+
"rewards/margins": 0.15809166431427002,
|
| 78 |
+
"rewards/rejected": -0.34598982334136963,
|
| 79 |
+
"step": 40
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.08,
|
| 83 |
+
"learning_rate": 3.968253968253968e-06,
|
| 84 |
+
"logits/chosen": 0.21212966740131378,
|
| 85 |
+
"logits/rejected": 0.055394046008586884,
|
| 86 |
+
"logps/chosen": -1518.728271484375,
|
| 87 |
+
"logps/rejected": -2687.27783203125,
|
| 88 |
+
"loss": 0.4803,
|
| 89 |
+
"rewards/accuracies": 0.581250011920929,
|
| 90 |
+
"rewards/chosen": -0.5128198862075806,
|
| 91 |
+
"rewards/margins": 0.28584352135658264,
|
| 92 |
+
"rewards/rejected": -0.7986633777618408,
|
| 93 |
+
"step": 50
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.1,
|
| 97 |
+
"learning_rate": 4.761904761904762e-06,
|
| 98 |
+
"logits/chosen": 0.14459244906902313,
|
| 99 |
+
"logits/rejected": 0.03840586543083191,
|
| 100 |
+
"logps/chosen": -2544.2294921875,
|
| 101 |
+
"logps/rejected": -3382.06640625,
|
| 102 |
+
"loss": 0.4881,
|
| 103 |
+
"rewards/accuracies": 0.5,
|
| 104 |
+
"rewards/chosen": -1.1230318546295166,
|
| 105 |
+
"rewards/margins": 0.08744711428880692,
|
| 106 |
+
"rewards/rejected": -1.2104789018630981,
|
| 107 |
+
"step": 60
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.11,
|
| 111 |
+
"learning_rate": 4.998086282661188e-06,
|
| 112 |
+
"logits/chosen": 0.03217538818717003,
|
| 113 |
+
"logits/rejected": -0.029030317440629005,
|
| 114 |
+
"logps/chosen": -2578.112060546875,
|
| 115 |
+
"logps/rejected": -3333.111328125,
|
| 116 |
+
"loss": 0.4889,
|
| 117 |
+
"rewards/accuracies": 0.48124998807907104,
|
| 118 |
+
"rewards/chosen": -1.2647404670715332,
|
| 119 |
+
"rewards/margins": 0.06161295250058174,
|
| 120 |
+
"rewards/rejected": -1.3263534307479858,
|
| 121 |
+
"step": 70
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.13,
|
| 125 |
+
"learning_rate": 4.988720025682995e-06,
|
| 126 |
+
"logits/chosen": 0.1540413349866867,
|
| 127 |
+
"logits/rejected": 0.00599607964977622,
|
| 128 |
+
"logps/chosen": -2171.90966796875,
|
| 129 |
+
"logps/rejected": -3310.91259765625,
|
| 130 |
+
"loss": 0.4818,
|
| 131 |
+
"rewards/accuracies": 0.53125,
|
| 132 |
+
"rewards/chosen": -1.0253573656082153,
|
| 133 |
+
"rewards/margins": 0.32645487785339355,
|
| 134 |
+
"rewards/rejected": -1.3518123626708984,
|
| 135 |
+
"step": 80
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.14,
|
| 139 |
+
"learning_rate": 4.9715789537359126e-06,
|
| 140 |
+
"logits/chosen": 0.18023057281970978,
|
| 141 |
+
"logits/rejected": -0.0267815999686718,
|
| 142 |
+
"logps/chosen": -2072.573974609375,
|
| 143 |
+
"logps/rejected": -3439.62744140625,
|
| 144 |
+
"loss": 0.477,
|
| 145 |
+
"rewards/accuracies": 0.5687500238418579,
|
| 146 |
+
"rewards/chosen": -0.79749995470047,
|
| 147 |
+
"rewards/margins": 0.3549983501434326,
|
| 148 |
+
"rewards/rejected": -1.152498483657837,
|
| 149 |
+
"step": 90
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.16,
|
| 153 |
+
"learning_rate": 4.946716615897932e-06,
|
| 154 |
+
"logits/chosen": 0.27958863973617554,
|
| 155 |
+
"logits/rejected": 0.08261282742023468,
|
| 156 |
+
"logps/chosen": -1780.5172119140625,
|
| 157 |
+
"logps/rejected": -2961.440185546875,
|
| 158 |
+
"loss": 0.4749,
|
| 159 |
+
"rewards/accuracies": 0.518750011920929,
|
| 160 |
+
"rewards/chosen": -0.6027330160140991,
|
| 161 |
+
"rewards/margins": 0.38158130645751953,
|
| 162 |
+
"rewards/rejected": -0.9843141436576843,
|
| 163 |
+
"step": 100
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.18,
|
| 167 |
+
"learning_rate": 4.9142106826480114e-06,
|
| 168 |
+
"logits/chosen": 0.139088436961174,
|
| 169 |
+
"logits/rejected": 0.03682307153940201,
|
| 170 |
+
"logps/chosen": -2555.55810546875,
|
| 171 |
+
"logps/rejected": -3190.896240234375,
|
| 172 |
+
"loss": 0.4867,
|
| 173 |
+
"rewards/accuracies": 0.44999998807907104,
|
| 174 |
+
"rewards/chosen": -0.9442774653434753,
|
| 175 |
+
"rewards/margins": 0.06666886806488037,
|
| 176 |
+
"rewards/rejected": -1.0109463930130005,
|
| 177 |
+
"step": 110
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.19,
|
| 181 |
+
"learning_rate": 4.874162703221823e-06,
|
| 182 |
+
"logits/chosen": 0.1611190289258957,
|
| 183 |
+
"logits/rejected": -0.012361553497612476,
|
| 184 |
+
"logps/chosen": -2050.61865234375,
|
| 185 |
+
"logps/rejected": -3838.018798828125,
|
| 186 |
+
"loss": 0.4732,
|
| 187 |
+
"rewards/accuracies": 0.59375,
|
| 188 |
+
"rewards/chosen": -0.9395005106925964,
|
| 189 |
+
"rewards/margins": 0.6280331015586853,
|
| 190 |
+
"rewards/rejected": -1.5675336122512817,
|
| 191 |
+
"step": 120
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.21,
|
| 195 |
+
"learning_rate": 4.826697788369752e-06,
|
| 196 |
+
"logits/chosen": 0.09378266334533691,
|
| 197 |
+
"logits/rejected": -0.05030001327395439,
|
| 198 |
+
"logps/chosen": -2598.26953125,
|
| 199 |
+
"logps/rejected": -3382.542236328125,
|
| 200 |
+
"loss": 0.479,
|
| 201 |
+
"rewards/accuracies": 0.512499988079071,
|
| 202 |
+
"rewards/chosen": -1.1976029872894287,
|
| 203 |
+
"rewards/margins": 0.14105060696601868,
|
| 204 |
+
"rewards/rejected": -1.338653564453125,
|
| 205 |
+
"step": 130
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.22,
|
| 209 |
+
"learning_rate": 4.7719642195082224e-06,
|
| 210 |
+
"logits/chosen": 0.14611801505088806,
|
| 211 |
+
"logits/rejected": 0.03608284890651703,
|
| 212 |
+
"logps/chosen": -1731.415771484375,
|
| 213 |
+
"logps/rejected": -2871.3349609375,
|
| 214 |
+
"loss": 0.4783,
|
| 215 |
+
"rewards/accuracies": 0.5062500238418579,
|
| 216 |
+
"rewards/chosen": -0.7977763414382935,
|
| 217 |
+
"rewards/margins": 0.4263841211795807,
|
| 218 |
+
"rewards/rejected": -1.2241604328155518,
|
| 219 |
+
"step": 140
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.24,
|
| 223 |
+
"learning_rate": 4.710132985485355e-06,
|
| 224 |
+
"logits/chosen": 0.1052960604429245,
|
| 225 |
+
"logits/rejected": -0.002506089163944125,
|
| 226 |
+
"logps/chosen": -1893.218505859375,
|
| 227 |
+
"logps/rejected": -2854.30859375,
|
| 228 |
+
"loss": 0.4777,
|
| 229 |
+
"rewards/accuracies": 0.574999988079071,
|
| 230 |
+
"rewards/chosen": -0.6661534905433655,
|
| 231 |
+
"rewards/margins": 0.35293903946876526,
|
| 232 |
+
"rewards/rejected": -1.0190925598144531,
|
| 233 |
+
"step": 150
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.26,
|
| 237 |
+
"learning_rate": 4.641397248408122e-06,
|
| 238 |
+
"logits/chosen": 0.017198827117681503,
|
| 239 |
+
"logits/rejected": -0.11535916477441788,
|
| 240 |
+
"logps/chosen": -2269.91650390625,
|
| 241 |
+
"logps/rejected": -3919.4375,
|
| 242 |
+
"loss": 0.4746,
|
| 243 |
+
"rewards/accuracies": 0.5062500238418579,
|
| 244 |
+
"rewards/chosen": -0.9236265420913696,
|
| 245 |
+
"rewards/margins": 0.3831055164337158,
|
| 246 |
+
"rewards/rejected": -1.306731939315796,
|
| 247 |
+
"step": 160
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.27,
|
| 251 |
+
"learning_rate": 4.5659717401997655e-06,
|
| 252 |
+
"logits/chosen": 0.01966998353600502,
|
| 253 |
+
"logits/rejected": -0.08529923856258392,
|
| 254 |
+
"logps/chosen": -2126.437255859375,
|
| 255 |
+
"logps/rejected": -3403.6328125,
|
| 256 |
+
"loss": 0.4795,
|
| 257 |
+
"rewards/accuracies": 0.543749988079071,
|
| 258 |
+
"rewards/chosen": -1.0373507738113403,
|
| 259 |
+
"rewards/margins": 0.41469430923461914,
|
| 260 |
+
"rewards/rejected": -1.452045202255249,
|
| 261 |
+
"step": 170
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.29,
|
| 265 |
+
"learning_rate": 4.4840920917726425e-06,
|
| 266 |
+
"logits/chosen": 0.05112285539507866,
|
| 267 |
+
"logits/rejected": -0.052227288484573364,
|
| 268 |
+
"logps/chosen": -2470.00537109375,
|
| 269 |
+
"logps/rejected": -3359.12109375,
|
| 270 |
+
"loss": 0.4825,
|
| 271 |
+
"rewards/accuracies": 0.550000011920929,
|
| 272 |
+
"rewards/chosen": -1.1501070261001587,
|
| 273 |
+
"rewards/margins": 0.24429550766944885,
|
| 274 |
+
"rewards/rejected": -1.3944026231765747,
|
| 275 |
+
"step": 180
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.3,
|
| 279 |
+
"learning_rate": 4.396014096912182e-06,
|
| 280 |
+
"logits/chosen": 0.04223916679620743,
|
| 281 |
+
"logits/rejected": -0.13196751475334167,
|
| 282 |
+
"logps/chosen": -2119.87841796875,
|
| 283 |
+
"logps/rejected": -3257.087158203125,
|
| 284 |
+
"loss": 0.4842,
|
| 285 |
+
"rewards/accuracies": 0.4937500059604645,
|
| 286 |
+
"rewards/chosen": -1.0418171882629395,
|
| 287 |
+
"rewards/margins": 0.2414085417985916,
|
| 288 |
+
"rewards/rejected": -1.2832257747650146,
|
| 289 |
+
"step": 190
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.32,
|
| 293 |
+
"learning_rate": 4.302012913171584e-06,
|
| 294 |
+
"logits/chosen": 0.01791330613195896,
|
| 295 |
+
"logits/rejected": -0.14596834778785706,
|
| 296 |
+
"logps/chosen": -2362.22314453125,
|
| 297 |
+
"logps/rejected": -3574.25927734375,
|
| 298 |
+
"loss": 0.4675,
|
| 299 |
+
"rewards/accuracies": 0.6000000238418579,
|
| 300 |
+
"rewards/chosen": -1.061951994895935,
|
| 301 |
+
"rewards/margins": 0.25661829113960266,
|
| 302 |
+
"rewards/rejected": -1.3185702562332153,
|
| 303 |
+
"step": 200
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.34,
|
| 307 |
+
"learning_rate": 4.202382202273702e-06,
|
| 308 |
+
"logits/chosen": 0.012257062830030918,
|
| 309 |
+
"logits/rejected": -0.15737880766391754,
|
| 310 |
+
"logps/chosen": -2264.546630859375,
|
| 311 |
+
"logps/rejected": -4172.1708984375,
|
| 312 |
+
"loss": 0.4655,
|
| 313 |
+
"rewards/accuracies": 0.637499988079071,
|
| 314 |
+
"rewards/chosen": -1.0106382369995117,
|
| 315 |
+
"rewards/margins": 0.853549599647522,
|
| 316 |
+
"rewards/rejected": -1.8641879558563232,
|
| 317 |
+
"step": 210
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.35,
|
| 321 |
+
"learning_rate": 4.097433212705492e-06,
|
| 322 |
+
"logits/chosen": -0.0625365823507309,
|
| 323 |
+
"logits/rejected": -0.27418404817581177,
|
| 324 |
+
"logps/chosen": -2211.255615234375,
|
| 325 |
+
"logps/rejected": -4339.79638671875,
|
| 326 |
+
"loss": 0.4675,
|
| 327 |
+
"rewards/accuracies": 0.6312500238418579,
|
| 328 |
+
"rewards/chosen": -1.0741349458694458,
|
| 329 |
+
"rewards/margins": 0.7618507742881775,
|
| 330 |
+
"rewards/rejected": -1.835985779762268,
|
| 331 |
+
"step": 220
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.37,
|
| 335 |
+
"learning_rate": 3.987493807371033e-06,
|
| 336 |
+
"logits/chosen": 0.07172416150569916,
|
| 337 |
+
"logits/rejected": 0.0005642950418405235,
|
| 338 |
+
"logps/chosen": -2197.752197265625,
|
| 339 |
+
"logps/rejected": -3114.567626953125,
|
| 340 |
+
"loss": 0.471,
|
| 341 |
+
"rewards/accuracies": 0.5562499761581421,
|
| 342 |
+
"rewards/chosen": -0.9756608009338379,
|
| 343 |
+
"rewards/margins": 0.24378642439842224,
|
| 344 |
+
"rewards/rejected": -1.2194470167160034,
|
| 345 |
+
"step": 230
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.38,
|
| 349 |
+
"learning_rate": 3.872907439340758e-06,
|
| 350 |
+
"logits/chosen": 0.01632564514875412,
|
| 351 |
+
"logits/rejected": -0.18748557567596436,
|
| 352 |
+
"logps/chosen": -2377.099853515625,
|
| 353 |
+
"logps/rejected": -4215.26953125,
|
| 354 |
+
"loss": 0.4747,
|
| 355 |
+
"rewards/accuracies": 0.5625,
|
| 356 |
+
"rewards/chosen": -1.054962396621704,
|
| 357 |
+
"rewards/margins": 0.6837267279624939,
|
| 358 |
+
"rewards/rejected": -1.7386891841888428,
|
| 359 |
+
"step": 240
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.4,
|
| 363 |
+
"learning_rate": 3.75403207889666e-06,
|
| 364 |
+
"logits/chosen": -0.012413917109370232,
|
| 365 |
+
"logits/rejected": -0.13336405158042908,
|
| 366 |
+
"logps/chosen": -2371.0888671875,
|
| 367 |
+
"logps/rejected": -3082.55322265625,
|
| 368 |
+
"loss": 0.4731,
|
| 369 |
+
"rewards/accuracies": 0.518750011920929,
|
| 370 |
+
"rewards/chosen": -0.9990957379341125,
|
| 371 |
+
"rewards/margins": 0.2219332903623581,
|
| 372 |
+
"rewards/rejected": -1.2210289239883423,
|
| 373 |
+
"step": 250
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.42,
|
| 377 |
+
"learning_rate": 3.631239095225417e-06,
|
| 378 |
+
"logits/chosen": 0.007460703607648611,
|
| 379 |
+
"logits/rejected": -0.1302846223115921,
|
| 380 |
+
"logps/chosen": -2123.05810546875,
|
| 381 |
+
"logps/rejected": -3470.561279296875,
|
| 382 |
+
"loss": 0.4727,
|
| 383 |
+
"rewards/accuracies": 0.543749988079071,
|
| 384 |
+
"rewards/chosen": -1.0149245262145996,
|
| 385 |
+
"rewards/margins": 0.44816678762435913,
|
| 386 |
+
"rewards/rejected": -1.463091254234314,
|
| 387 |
+
"step": 260
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.43,
|
| 391 |
+
"learning_rate": 3.5049120962530608e-06,
|
| 392 |
+
"logits/chosen": -0.0549989752471447,
|
| 393 |
+
"logits/rejected": -0.2491796910762787,
|
| 394 |
+
"logps/chosen": -2456.557861328125,
|
| 395 |
+
"logps/rejected": -4269.2412109375,
|
| 396 |
+
"loss": 0.4724,
|
| 397 |
+
"rewards/accuracies": 0.5687500238418579,
|
| 398 |
+
"rewards/chosen": -1.0748237371444702,
|
| 399 |
+
"rewards/margins": 0.7018817067146301,
|
| 400 |
+
"rewards/rejected": -1.7767053842544556,
|
| 401 |
+
"step": 270
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.45,
|
| 405 |
+
"learning_rate": 3.3754457302455464e-06,
|
| 406 |
+
"logits/chosen": 0.028259318321943283,
|
| 407 |
+
"logits/rejected": -0.15549519658088684,
|
| 408 |
+
"logps/chosen": -2343.92724609375,
|
| 409 |
+
"logps/rejected": -3815.20751953125,
|
| 410 |
+
"loss": 0.4775,
|
| 411 |
+
"rewards/accuracies": 0.5562499761581421,
|
| 412 |
+
"rewards/chosen": -0.9266031384468079,
|
| 413 |
+
"rewards/margins": 0.4884551167488098,
|
| 414 |
+
"rewards/rejected": -1.4150583744049072,
|
| 415 |
+
"step": 280
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.46,
|
| 419 |
+
"learning_rate": 3.2432444529190714e-06,
|
| 420 |
+
"logits/chosen": 0.09060511738061905,
|
| 421 |
+
"logits/rejected": -0.13999487459659576,
|
| 422 |
+
"logps/chosen": -1729.5335693359375,
|
| 423 |
+
"logps/rejected": -3556.65771484375,
|
| 424 |
+
"loss": 0.4629,
|
| 425 |
+
"rewards/accuracies": 0.574999988079071,
|
| 426 |
+
"rewards/chosen": -0.5673225522041321,
|
| 427 |
+
"rewards/margins": 0.5148328542709351,
|
| 428 |
+
"rewards/rejected": -1.082155466079712,
|
| 429 |
+
"step": 290
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.48,
|
| 433 |
+
"learning_rate": 3.1087212639117057e-06,
|
| 434 |
+
"logits/chosen": 0.10731784254312515,
|
| 435 |
+
"logits/rejected": -0.12288031727075577,
|
| 436 |
+
"logps/chosen": -1606.456787109375,
|
| 437 |
+
"logps/rejected": -3169.525146484375,
|
| 438 |
+
"loss": 0.4802,
|
| 439 |
+
"rewards/accuracies": 0.612500011920929,
|
| 440 |
+
"rewards/chosen": -0.47551971673965454,
|
| 441 |
+
"rewards/margins": 0.5530696511268616,
|
| 442 |
+
"rewards/rejected": -1.0285893678665161,
|
| 443 |
+
"step": 300
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.5,
|
| 447 |
+
"learning_rate": 2.9722964165636263e-06,
|
| 448 |
+
"logits/chosen": -0.040035147219896317,
|
| 449 |
+
"logits/rejected": -0.13161209225654602,
|
| 450 |
+
"logps/chosen": -2169.19677734375,
|
| 451 |
+
"logps/rejected": -3303.66064453125,
|
| 452 |
+
"loss": 0.4767,
|
| 453 |
+
"rewards/accuracies": 0.574999988079071,
|
| 454 |
+
"rewards/chosen": -0.8213248252868652,
|
| 455 |
+
"rewards/margins": 0.4187691807746887,
|
| 456 |
+
"rewards/rejected": -1.2400939464569092,
|
| 457 |
+
"step": 310
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.51,
|
| 461 |
+
"learning_rate": 2.8343961050366275e-06,
|
| 462 |
+
"logits/chosen": -0.0054204524494707584,
|
| 463 |
+
"logits/rejected": -0.2861900329589844,
|
| 464 |
+
"logps/chosen": -1592.34375,
|
| 465 |
+
"logps/rejected": -3716.65869140625,
|
| 466 |
+
"loss": 0.4721,
|
| 467 |
+
"rewards/accuracies": 0.6187499761581421,
|
| 468 |
+
"rewards/chosen": -0.7492900490760803,
|
| 469 |
+
"rewards/margins": 0.8597043752670288,
|
| 470 |
+
"rewards/rejected": -1.608994483947754,
|
| 471 |
+
"step": 320
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.53,
|
| 475 |
+
"learning_rate": 2.695451132874385e-06,
|
| 476 |
+
"logits/chosen": -0.1356322020292282,
|
| 477 |
+
"logits/rejected": -0.292384535074234,
|
| 478 |
+
"logps/chosen": -1807.0335693359375,
|
| 479 |
+
"logps/rejected": -3166.426513671875,
|
| 480 |
+
"loss": 0.48,
|
| 481 |
+
"rewards/accuracies": 0.5562499761581421,
|
| 482 |
+
"rewards/chosen": -0.7727370858192444,
|
| 483 |
+
"rewards/margins": 0.4709666669368744,
|
| 484 |
+
"rewards/rejected": -1.2437037229537964,
|
| 485 |
+
"step": 330
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.54,
|
| 489 |
+
"learning_rate": 2.5558955671628964e-06,
|
| 490 |
+
"logits/chosen": -0.07628178596496582,
|
| 491 |
+
"logits/rejected": -0.20791587233543396,
|
| 492 |
+
"logps/chosen": -1737.909423828125,
|
| 493 |
+
"logps/rejected": -3203.9970703125,
|
| 494 |
+
"loss": 0.4668,
|
| 495 |
+
"rewards/accuracies": 0.5249999761581421,
|
| 496 |
+
"rewards/chosen": -0.7647561430931091,
|
| 497 |
+
"rewards/margins": 0.49256449937820435,
|
| 498 |
+
"rewards/rejected": -1.2573206424713135,
|
| 499 |
+
"step": 340
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.56,
|
| 503 |
+
"learning_rate": 2.4161653824955654e-06,
|
| 504 |
+
"logits/chosen": -0.05267338082194328,
|
| 505 |
+
"logits/rejected": -0.25987547636032104,
|
| 506 |
+
"logps/chosen": -2673.15380859375,
|
| 507 |
+
"logps/rejected": -3792.9609375,
|
| 508 |
+
"loss": 0.4653,
|
| 509 |
+
"rewards/accuracies": 0.5249999761581421,
|
| 510 |
+
"rewards/chosen": -1.1372772455215454,
|
| 511 |
+
"rewards/margins": 0.37971851229667664,
|
| 512 |
+
"rewards/rejected": -1.5169956684112549,
|
| 513 |
+
"step": 350
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.58,
|
| 517 |
+
"learning_rate": 2.2766970989791697e-06,
|
| 518 |
+
"logits/chosen": -0.15440881252288818,
|
| 519 |
+
"logits/rejected": -0.3189676105976105,
|
| 520 |
+
"logps/chosen": -2074.7294921875,
|
| 521 |
+
"logps/rejected": -3670.177001953125,
|
| 522 |
+
"loss": 0.47,
|
| 523 |
+
"rewards/accuracies": 0.581250011920929,
|
| 524 |
+
"rewards/chosen": -0.8276017904281616,
|
| 525 |
+
"rewards/margins": 0.6911835670471191,
|
| 526 |
+
"rewards/rejected": -1.5187852382659912,
|
| 527 |
+
"step": 360
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.59,
|
| 531 |
+
"learning_rate": 2.1379264185356545e-06,
|
| 532 |
+
"logits/chosen": -0.03989617899060249,
|
| 533 |
+
"logits/rejected": -0.29078492522239685,
|
| 534 |
+
"logps/chosen": -2240.99951171875,
|
| 535 |
+
"logps/rejected": -3693.21875,
|
| 536 |
+
"loss": 0.4675,
|
| 537 |
+
"rewards/accuracies": 0.574999988079071,
|
| 538 |
+
"rewards/chosen": -0.982672393321991,
|
| 539 |
+
"rewards/margins": 0.5129915475845337,
|
| 540 |
+
"rewards/rejected": -1.4956640005111694,
|
| 541 |
+
"step": 370
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.61,
|
| 545 |
+
"learning_rate": 2.000286863759934e-06,
|
| 546 |
+
"logits/chosen": -0.18141961097717285,
|
| 547 |
+
"logits/rejected": -0.34517520666122437,
|
| 548 |
+
"logps/chosen": -2447.95458984375,
|
| 549 |
+
"logps/rejected": -4179.51220703125,
|
| 550 |
+
"loss": 0.4721,
|
| 551 |
+
"rewards/accuracies": 0.550000011920929,
|
| 552 |
+
"rewards/chosen": -1.249446153640747,
|
| 553 |
+
"rewards/margins": 0.7962532639503479,
|
| 554 |
+
"rewards/rejected": -2.04569935798645,
|
| 555 |
+
"step": 380
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.62,
|
| 559 |
+
"learning_rate": 1.8642084235859764e-06,
|
| 560 |
+
"logits/chosen": -0.10085698217153549,
|
| 561 |
+
"logits/rejected": -0.4116978645324707,
|
| 562 |
+
"logps/chosen": -2039.9761962890625,
|
| 563 |
+
"logps/rejected": -4429.87451171875,
|
| 564 |
+
"loss": 0.4625,
|
| 565 |
+
"rewards/accuracies": 0.625,
|
| 566 |
+
"rewards/chosen": -1.0368822813034058,
|
| 567 |
+
"rewards/margins": 1.2251580953598022,
|
| 568 |
+
"rewards/rejected": -2.262040615081787,
|
| 569 |
+
"step": 390
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.64,
|
| 573 |
+
"learning_rate": 1.7301162099921013e-06,
|
| 574 |
+
"logits/chosen": -0.07682979851961136,
|
| 575 |
+
"logits/rejected": -0.35371267795562744,
|
| 576 |
+
"logps/chosen": -2304.14306640625,
|
| 577 |
+
"logps/rejected": -3896.870361328125,
|
| 578 |
+
"loss": 0.4642,
|
| 579 |
+
"rewards/accuracies": 0.5625,
|
| 580 |
+
"rewards/chosen": -1.131152868270874,
|
| 581 |
+
"rewards/margins": 0.7709552645683289,
|
| 582 |
+
"rewards/rejected": -1.9021081924438477,
|
| 583 |
+
"step": 400
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.66,
|
| 587 |
+
"learning_rate": 1.5984291299420117e-06,
|
| 588 |
+
"logits/chosen": -0.07511943578720093,
|
| 589 |
+
"logits/rejected": -0.28092044591903687,
|
| 590 |
+
"logps/chosen": -2233.59814453125,
|
| 591 |
+
"logps/rejected": -3915.47900390625,
|
| 592 |
+
"loss": 0.4691,
|
| 593 |
+
"rewards/accuracies": 0.581250011920929,
|
| 594 |
+
"rewards/chosen": -1.0048208236694336,
|
| 595 |
+
"rewards/margins": 0.6850495934486389,
|
| 596 |
+
"rewards/rejected": -1.6898702383041382,
|
| 597 |
+
"step": 410
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.67,
|
| 601 |
+
"learning_rate": 1.4695585767104092e-06,
|
| 602 |
+
"logits/chosen": -0.005914182402193546,
|
| 603 |
+
"logits/rejected": -0.24251346290111542,
|
| 604 |
+
"logps/chosen": -1424.0343017578125,
|
| 605 |
+
"logps/rejected": -3496.74755859375,
|
| 606 |
+
"loss": 0.4581,
|
| 607 |
+
"rewards/accuracies": 0.59375,
|
| 608 |
+
"rewards/chosen": -0.6118324398994446,
|
| 609 |
+
"rewards/margins": 0.6595968008041382,
|
| 610 |
+
"rewards/rejected": -1.2714293003082275,
|
| 611 |
+
"step": 420
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.69,
|
| 615 |
+
"learning_rate": 1.3439071446815452e-06,
|
| 616 |
+
"logits/chosen": -0.05694418027997017,
|
| 617 |
+
"logits/rejected": -0.27310264110565186,
|
| 618 |
+
"logps/chosen": -1811.8056640625,
|
| 619 |
+
"logps/rejected": -3876.968017578125,
|
| 620 |
+
"loss": 0.4613,
|
| 621 |
+
"rewards/accuracies": 0.6187499761581421,
|
| 622 |
+
"rewards/chosen": -0.7894518971443176,
|
| 623 |
+
"rewards/margins": 0.9580751657485962,
|
| 624 |
+
"rewards/rejected": -1.7475271224975586,
|
| 625 |
+
"step": 430
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.7,
|
| 629 |
+
"learning_rate": 1.2218673716356919e-06,
|
| 630 |
+
"logits/chosen": -0.04861157387495041,
|
| 631 |
+
"logits/rejected": -0.22785380482673645,
|
| 632 |
+
"logps/chosen": -2162.256103515625,
|
| 633 |
+
"logps/rejected": -3135.13330078125,
|
| 634 |
+
"loss": 0.4752,
|
| 635 |
+
"rewards/accuracies": 0.4749999940395355,
|
| 636 |
+
"rewards/chosen": -0.9659271240234375,
|
| 637 |
+
"rewards/margins": 0.3405894339084625,
|
| 638 |
+
"rewards/rejected": -1.3065165281295776,
|
| 639 |
+
"step": 440
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.72,
|
| 643 |
+
"learning_rate": 1.103820512452661e-06,
|
| 644 |
+
"logits/chosen": -0.0886421948671341,
|
| 645 |
+
"logits/rejected": -0.24363021552562714,
|
| 646 |
+
"logps/chosen": -2278.705078125,
|
| 647 |
+
"logps/rejected": -3871.57861328125,
|
| 648 |
+
"loss": 0.4697,
|
| 649 |
+
"rewards/accuracies": 0.574999988079071,
|
| 650 |
+
"rewards/chosen": -1.0452690124511719,
|
| 651 |
+
"rewards/margins": 0.5815836191177368,
|
| 652 |
+
"rewards/rejected": -1.6268523931503296,
|
| 653 |
+
"step": 450
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.74,
|
| 657 |
+
"learning_rate": 9.901353480633468e-07,
|
| 658 |
+
"logits/chosen": -0.01507838536053896,
|
| 659 |
+
"logits/rejected": -0.27143269777297974,
|
| 660 |
+
"logps/chosen": -1932.046875,
|
| 661 |
+
"logps/rejected": -3638.012939453125,
|
| 662 |
+
"loss": 0.4705,
|
| 663 |
+
"rewards/accuracies": 0.6000000238418579,
|
| 664 |
+
"rewards/chosen": -0.8395353555679321,
|
| 665 |
+
"rewards/margins": 0.6771036982536316,
|
| 666 |
+
"rewards/rejected": -1.5166391134262085,
|
| 667 |
+
"step": 460
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.75,
|
| 671 |
+
"learning_rate": 8.811670333701544e-07,
|
| 672 |
+
"logits/chosen": -0.06362856924533844,
|
| 673 |
+
"logits/rejected": -0.2936163544654846,
|
| 674 |
+
"logps/chosen": -1988.303955078125,
|
| 675 |
+
"logps/rejected": -4044.038330078125,
|
| 676 |
+
"loss": 0.4641,
|
| 677 |
+
"rewards/accuracies": 0.5062500238418579,
|
| 678 |
+
"rewards/chosen": -0.8915748596191406,
|
| 679 |
+
"rewards/margins": 0.8164836168289185,
|
| 680 |
+
"rewards/rejected": -1.7080585956573486,
|
| 681 |
+
"step": 470
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.77,
|
| 685 |
+
"learning_rate": 7.772559877354341e-07,
|
| 686 |
+
"logits/chosen": -0.030277037993073463,
|
| 687 |
+
"logits/rejected": -0.21407613158226013,
|
| 688 |
+
"logps/chosen": -2318.84912109375,
|
| 689 |
+
"logps/rejected": -4174.81298828125,
|
| 690 |
+
"loss": 0.4745,
|
| 691 |
+
"rewards/accuracies": 0.543749988079071,
|
| 692 |
+
"rewards/chosen": -0.9398199319839478,
|
| 693 |
+
"rewards/margins": 0.7878143787384033,
|
| 694 |
+
"rewards/rejected": -1.7276341915130615,
|
| 695 |
+
"step": 480
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.78,
|
| 699 |
+
"learning_rate": 6.787268315040604e-07,
|
| 700 |
+
"logits/chosen": -0.1001216396689415,
|
| 701 |
+
"logits/rejected": -0.24970397353172302,
|
| 702 |
+
"logps/chosen": -2327.34375,
|
| 703 |
+
"logps/rejected": -3589.403564453125,
|
| 704 |
+
"loss": 0.4775,
|
| 705 |
+
"rewards/accuracies": 0.5062500238418579,
|
| 706 |
+
"rewards/chosen": -0.9565193057060242,
|
| 707 |
+
"rewards/margins": 0.5348533987998962,
|
| 708 |
+
"rewards/rejected": -1.4913727045059204,
|
| 709 |
+
"step": 490
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.8,
|
| 713 |
+
"learning_rate": 5.858873718824829e-07,
|
| 714 |
+
"logits/chosen": -0.08827606588602066,
|
| 715 |
+
"logits/rejected": -0.2271912395954132,
|
| 716 |
+
"logps/chosen": -2329.033935546875,
|
| 717 |
+
"logps/rejected": -3762.891357421875,
|
| 718 |
+
"loss": 0.4701,
|
| 719 |
+
"rewards/accuracies": 0.543749988079071,
|
| 720 |
+
"rewards/chosen": -0.9019953608512878,
|
| 721 |
+
"rewards/margins": 0.6341419219970703,
|
| 722 |
+
"rewards/rejected": -1.536137342453003,
|
| 723 |
+
"step": 500
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.82,
|
| 727 |
+
"learning_rate": 4.990276413423817e-07,
|
| 728 |
+
"logits/chosen": 0.006130737718194723,
|
| 729 |
+
"logits/rejected": -0.2663702070713043,
|
| 730 |
+
"logps/chosen": -1954.1243896484375,
|
| 731 |
+
"logps/rejected": -4118.31689453125,
|
| 732 |
+
"loss": 0.4608,
|
| 733 |
+
"rewards/accuracies": 0.59375,
|
| 734 |
+
"rewards/chosen": -0.7334184050559998,
|
| 735 |
+
"rewards/margins": 0.8962429761886597,
|
| 736 |
+
"rewards/rejected": -1.6296613216400146,
|
| 737 |
+
"step": 510
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.83,
|
| 741 |
+
"learning_rate": 4.184189915529796e-07,
|
| 742 |
+
"logits/chosen": -0.022313248366117477,
|
| 743 |
+
"logits/rejected": -0.2796049416065216,
|
| 744 |
+
"logps/chosen": -2069.07861328125,
|
| 745 |
+
"logps/rejected": -4346.296875,
|
| 746 |
+
"loss": 0.4631,
|
| 747 |
+
"rewards/accuracies": 0.6187499761581421,
|
| 748 |
+
"rewards/chosen": -0.8130847215652466,
|
| 749 |
+
"rewards/margins": 0.9093513488769531,
|
| 750 |
+
"rewards/rejected": -1.7224359512329102,
|
| 751 |
+
"step": 520
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.85,
|
| 755 |
+
"learning_rate": 3.4431324567258176e-07,
|
| 756 |
+
"logits/chosen": -0.006228646729141474,
|
| 757 |
+
"logits/rejected": -0.18093259632587433,
|
| 758 |
+
"logps/chosen": -1853.189453125,
|
| 759 |
+
"logps/rejected": -3369.416748046875,
|
| 760 |
+
"loss": 0.4685,
|
| 761 |
+
"rewards/accuracies": 0.550000011920929,
|
| 762 |
+
"rewards/chosen": -0.7921710014343262,
|
| 763 |
+
"rewards/margins": 0.5718700885772705,
|
| 764 |
+
"rewards/rejected": -1.3640410900115967,
|
| 765 |
+
"step": 530
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.86,
|
| 769 |
+
"learning_rate": 2.769419116476052e-07,
|
| 770 |
+
"logits/chosen": -0.0418720506131649,
|
| 771 |
+
"logits/rejected": -0.30269795656204224,
|
| 772 |
+
"logps/chosen": -2341.14501953125,
|
| 773 |
+
"logps/rejected": -4083.86083984375,
|
| 774 |
+
"loss": 0.4729,
|
| 775 |
+
"rewards/accuracies": 0.5687500238418579,
|
| 776 |
+
"rewards/chosen": -0.9471173286437988,
|
| 777 |
+
"rewards/margins": 0.6616966128349304,
|
| 778 |
+
"rewards/rejected": -1.6088138818740845,
|
| 779 |
+
"step": 540
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.88,
|
| 783 |
+
"learning_rate": 2.1651545897676512e-07,
|
| 784 |
+
"logits/chosen": -0.08359251916408539,
|
| 785 |
+
"logits/rejected": -0.31541210412979126,
|
| 786 |
+
"logps/chosen": -1795.650390625,
|
| 787 |
+
"logps/rejected": -3922.41259765625,
|
| 788 |
+
"loss": 0.4713,
|
| 789 |
+
"rewards/accuracies": 0.581250011920929,
|
| 790 |
+
"rewards/chosen": -0.6783340573310852,
|
| 791 |
+
"rewards/margins": 0.7878134846687317,
|
| 792 |
+
"rewards/rejected": -1.4661474227905273,
|
| 793 |
+
"step": 550
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.9,
|
| 797 |
+
"learning_rate": 1.6322266119983222e-07,
|
| 798 |
+
"logits/chosen": -0.08543523401021957,
|
| 799 |
+
"logits/rejected": -0.22058483958244324,
|
| 800 |
+
"logps/chosen": -2356.09619140625,
|
| 801 |
+
"logps/rejected": -2975.605712890625,
|
| 802 |
+
"loss": 0.4693,
|
| 803 |
+
"rewards/accuracies": 0.581250011920929,
|
| 804 |
+
"rewards/chosen": -0.9343336224555969,
|
| 805 |
+
"rewards/margins": 0.29671743512153625,
|
| 806 |
+
"rewards/rejected": -1.231050968170166,
|
| 807 |
+
"step": 560
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 0.91,
|
| 811 |
+
"learning_rate": 1.1723000616502167e-07,
|
| 812 |
+
"logits/chosen": -0.07000622898340225,
|
| 813 |
+
"logits/rejected": -0.2409258633852005,
|
| 814 |
+
"logps/chosen": -2480.648681640625,
|
| 815 |
+
"logps/rejected": -3360.589111328125,
|
| 816 |
+
"loss": 0.4772,
|
| 817 |
+
"rewards/accuracies": 0.5062500238418579,
|
| 818 |
+
"rewards/chosen": -1.0692129135131836,
|
| 819 |
+
"rewards/margins": 0.2927553355693817,
|
| 820 |
+
"rewards/rejected": -1.3619682788848877,
|
| 821 |
+
"step": 570
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 0.93,
|
| 825 |
+
"learning_rate": 7.868117591737585e-08,
|
| 826 |
+
"logits/chosen": -0.02028440684080124,
|
| 827 |
+
"logits/rejected": -0.17536571621894836,
|
| 828 |
+
"logps/chosen": -2015.087646484375,
|
| 829 |
+
"logps/rejected": -3354.944580078125,
|
| 830 |
+
"loss": 0.4748,
|
| 831 |
+
"rewards/accuracies": 0.5874999761581421,
|
| 832 |
+
"rewards/chosen": -0.7683865427970886,
|
| 833 |
+
"rewards/margins": 0.602486252784729,
|
| 834 |
+
"rewards/rejected": -1.3708727359771729,
|
| 835 |
+
"step": 580
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 0.94,
|
| 839 |
+
"learning_rate": 4.769659783295383e-08,
|
| 840 |
+
"logits/chosen": -0.12467147409915924,
|
| 841 |
+
"logits/rejected": -0.2483750879764557,
|
| 842 |
+
"logps/chosen": -2129.997314453125,
|
| 843 |
+
"logps/rejected": -3120.9951171875,
|
| 844 |
+
"loss": 0.4782,
|
| 845 |
+
"rewards/accuracies": 0.5,
|
| 846 |
+
"rewards/chosen": -0.9002014398574829,
|
| 847 |
+
"rewards/margins": 0.3127484619617462,
|
| 848 |
+
"rewards/rejected": -1.2129498720169067,
|
| 849 |
+
"step": 590
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.96,
|
| 853 |
+
"learning_rate": 2.4373068401120358e-08,
|
| 854 |
+
"logits/chosen": -0.02018025331199169,
|
| 855 |
+
"logits/rejected": -0.2830565869808197,
|
| 856 |
+
"logps/chosen": -1880.4671630859375,
|
| 857 |
+
"logps/rejected": -3650.389404296875,
|
| 858 |
+
"loss": 0.4589,
|
| 859 |
+
"rewards/accuracies": 0.574999988079071,
|
| 860 |
+
"rewards/chosen": -0.7423778772354126,
|
| 861 |
+
"rewards/margins": 0.7750624418258667,
|
| 862 |
+
"rewards/rejected": -1.5174401998519897,
|
| 863 |
+
"step": 600
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.98,
|
| 867 |
+
"learning_rate": 8.78345083022425e-09,
|
| 868 |
+
"logits/chosen": -0.017909971997141838,
|
| 869 |
+
"logits/rejected": -0.2379368245601654,
|
| 870 |
+
"logps/chosen": -2240.98193359375,
|
| 871 |
+
"logps/rejected": -3461.29345703125,
|
| 872 |
+
"loss": 0.467,
|
| 873 |
+
"rewards/accuracies": 0.518750011920929,
|
| 874 |
+
"rewards/chosen": -1.001007080078125,
|
| 875 |
+
"rewards/margins": 0.4783032536506653,
|
| 876 |
+
"rewards/rejected": -1.479310154914856,
|
| 877 |
+
"step": 610
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 0.99,
|
| 881 |
+
"learning_rate": 9.764474213677654e-10,
|
| 882 |
+
"logits/chosen": -0.011101929470896721,
|
| 883 |
+
"logits/rejected": -0.18822148442268372,
|
| 884 |
+
"logps/chosen": -1725.9417724609375,
|
| 885 |
+
"logps/rejected": -3190.5283203125,
|
| 886 |
+
"loss": 0.4695,
|
| 887 |
+
"rewards/accuracies": 0.5,
|
| 888 |
+
"rewards/chosen": -0.661989688873291,
|
| 889 |
+
"rewards/margins": 0.6126888394355774,
|
| 890 |
+
"rewards/rejected": -1.2746784687042236,
|
| 891 |
+
"step": 620
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 1.0,
|
| 895 |
+
"step": 625,
|
| 896 |
+
"total_flos": 0.0,
|
| 897 |
+
"train_loss": 0.4739649757385254,
|
| 898 |
+
"train_runtime": 8016.6369,
|
| 899 |
+
"train_samples_per_second": 2.495,
|
| 900 |
+
"train_steps_per_second": 0.078
|
| 901 |
+
}
|
| 902 |
+
],
|
| 903 |
+
"logging_steps": 10,
|
| 904 |
+
"max_steps": 625,
|
| 905 |
+
"num_input_tokens_seen": 0,
|
| 906 |
+
"num_train_epochs": 1,
|
| 907 |
+
"save_steps": 20,
|
| 908 |
+
"total_flos": 0.0,
|
| 909 |
+
"train_batch_size": 4,
|
| 910 |
+
"trial_name": null,
|
| 911 |
+
"trial_params": null
|
| 912 |
+
}
|