Model save
Browse files- README.md +61 -0
- adapter_model.safetensors +1 -1
- all_results.json +8 -0
- emissions.csv +2 -0
- runs/Sep29_02-35-31_gpu4-119-5/events.out.tfevents.1727541437.gpu4-119-5.264437.0 +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-20-40-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-20-40-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 sha256:3cbfb0d70515210c121275a81887ae400d039b242feb4219ee180aaeabdf9a42
|
| 3 |
size 335579632
|
all_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"train_loss": 0.4757139442443848,
|
| 4 |
+
"train_runtime": 7766.6535,
|
| 5 |
+
"train_samples": 20000,
|
| 6 |
+
"train_samples_per_second": 2.575,
|
| 7 |
+
"train_steps_per_second": 0.08
|
| 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 |
+
2024-09-29T04:46:44,codecarbon,7545b572-ce62-4ef5-9e04-2540499cfc29,7766.672092437744,0.004515805584888482,5.814337892912246e-07,42.5,619.946,188.74309015274048,0.09168878622502102,1.4024227659387525,0.4055057152282073,1.899617267391984,Canada,CAN,quebec,,,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,-71.2,46.8,503.3149070739746,machine,N,1.0
|
runs/Sep29_02-35-31_gpu4-119-5/events.out.tfevents.1727541437.gpu4-119-5.264437.0
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 sha256:35a2f00fecca465f886fca69ebc744102be6132fb3d7818f98d41ae0e3c9f033
|
| 3 |
+
size 45043
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"train_loss": 0.4757139442443848,
|
| 4 |
+
"train_runtime": 7766.6535,
|
| 5 |
+
"train_samples": 20000,
|
| 6 |
+
"train_samples_per_second": 2.575,
|
| 7 |
+
"train_steps_per_second": 0.08
|
| 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": 0.3703172504901886,
|
| 15 |
+
"logits/rejected": 0.17183463275432587,
|
| 16 |
+
"logps/chosen": -1537.315673828125,
|
| 17 |
+
"logps/rejected": -1665.161865234375,
|
| 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": 0.2896106243133545,
|
| 29 |
+
"logits/rejected": 0.18362945318222046,
|
| 30 |
+
"logps/chosen": -1102.260986328125,
|
| 31 |
+
"logps/rejected": -2063.15380859375,
|
| 32 |
+
"loss": 0.5003,
|
| 33 |
+
"rewards/accuracies": 0.3263888955116272,
|
| 34 |
+
"rewards/chosen": -0.0019549066200852394,
|
| 35 |
+
"rewards/margins": -0.0033436818048357964,
|
| 36 |
+
"rewards/rejected": 0.0013887754175812006,
|
| 37 |
+
"step": 10
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.03,
|
| 41 |
+
"learning_rate": 1.5873015873015873e-06,
|
| 42 |
+
"logits/chosen": 0.31699010729789734,
|
| 43 |
+
"logits/rejected": 0.2042236030101776,
|
| 44 |
+
"logps/chosen": -1399.4488525390625,
|
| 45 |
+
"logps/rejected": -2060.23193359375,
|
| 46 |
+
"loss": 0.4997,
|
| 47 |
+
"rewards/accuracies": 0.4625000059604645,
|
| 48 |
+
"rewards/chosen": -0.016661832109093666,
|
| 49 |
+
"rewards/margins": 0.002577673178166151,
|
| 50 |
+
"rewards/rejected": -0.019239503890275955,
|
| 51 |
+
"step": 20
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.05,
|
| 55 |
+
"learning_rate": 2.380952380952381e-06,
|
| 56 |
+
"logits/chosen": 0.27783340215682983,
|
| 57 |
+
"logits/rejected": 0.18502432107925415,
|
| 58 |
+
"logps/chosen": -1305.1578369140625,
|
| 59 |
+
"logps/rejected": -2624.82470703125,
|
| 60 |
+
"loss": 0.4965,
|
| 61 |
+
"rewards/accuracies": 0.543749988079071,
|
| 62 |
+
"rewards/chosen": -0.060660868883132935,
|
| 63 |
+
"rewards/margins": 0.04234201833605766,
|
| 64 |
+
"rewards/rejected": -0.1030028909444809,
|
| 65 |
+
"step": 30
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.06,
|
| 69 |
+
"learning_rate": 3.1746031746031746e-06,
|
| 70 |
+
"logits/chosen": 0.3361937701702118,
|
| 71 |
+
"logits/rejected": 0.21628756821155548,
|
| 72 |
+
"logps/chosen": -1122.9556884765625,
|
| 73 |
+
"logps/rejected": -1928.48046875,
|
| 74 |
+
"loss": 0.4885,
|
| 75 |
+
"rewards/accuracies": 0.5625,
|
| 76 |
+
"rewards/chosen": -0.13624535501003265,
|
| 77 |
+
"rewards/margins": 0.11144338548183441,
|
| 78 |
+
"rewards/rejected": -0.24768872559070587,
|
| 79 |
+
"step": 40
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.08,
|
| 83 |
+
"learning_rate": 3.968253968253968e-06,
|
| 84 |
+
"logits/chosen": 0.19992919266223907,
|
| 85 |
+
"logits/rejected": 0.05517758056521416,
|
| 86 |
+
"logps/chosen": -1331.64990234375,
|
| 87 |
+
"logps/rejected": -2720.01611328125,
|
| 88 |
+
"loss": 0.4868,
|
| 89 |
+
"rewards/accuracies": 0.5687500238418579,
|
| 90 |
+
"rewards/chosen": -0.41040802001953125,
|
| 91 |
+
"rewards/margins": 0.1472548544406891,
|
| 92 |
+
"rewards/rejected": -0.557662844657898,
|
| 93 |
+
"step": 50
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.1,
|
| 97 |
+
"learning_rate": 4.761904761904762e-06,
|
| 98 |
+
"logits/chosen": 0.13223984837532043,
|
| 99 |
+
"logits/rejected": -0.04643384367227554,
|
| 100 |
+
"logps/chosen": -1749.2845458984375,
|
| 101 |
+
"logps/rejected": -3090.0,
|
| 102 |
+
"loss": 0.4836,
|
| 103 |
+
"rewards/accuracies": 0.5874999761581421,
|
| 104 |
+
"rewards/chosen": -0.7042622566223145,
|
| 105 |
+
"rewards/margins": 0.42035597562789917,
|
| 106 |
+
"rewards/rejected": -1.1246182918548584,
|
| 107 |
+
"step": 60
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.11,
|
| 111 |
+
"learning_rate": 4.998086282661188e-06,
|
| 112 |
+
"logits/chosen": 0.12587395310401917,
|
| 113 |
+
"logits/rejected": 0.018394339829683304,
|
| 114 |
+
"logps/chosen": -2005.3883056640625,
|
| 115 |
+
"logps/rejected": -3198.51171875,
|
| 116 |
+
"loss": 0.4777,
|
| 117 |
+
"rewards/accuracies": 0.5,
|
| 118 |
+
"rewards/chosen": -0.8340966105461121,
|
| 119 |
+
"rewards/margins": 0.4107601046562195,
|
| 120 |
+
"rewards/rejected": -1.2448567152023315,
|
| 121 |
+
"step": 70
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.13,
|
| 125 |
+
"learning_rate": 4.988720025682995e-06,
|
| 126 |
+
"logits/chosen": 0.10518720000982285,
|
| 127 |
+
"logits/rejected": -0.007509663701057434,
|
| 128 |
+
"logps/chosen": -2260.054443359375,
|
| 129 |
+
"logps/rejected": -3305.970703125,
|
| 130 |
+
"loss": 0.4906,
|
| 131 |
+
"rewards/accuracies": 0.48750001192092896,
|
| 132 |
+
"rewards/chosen": -1.0553181171417236,
|
| 133 |
+
"rewards/margins": 0.1571974903345108,
|
| 134 |
+
"rewards/rejected": -1.2125155925750732,
|
| 135 |
+
"step": 80
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.14,
|
| 139 |
+
"learning_rate": 4.9715789537359126e-06,
|
| 140 |
+
"logits/chosen": 0.10318700969219208,
|
| 141 |
+
"logits/rejected": -0.06382742524147034,
|
| 142 |
+
"logps/chosen": -1909.2587890625,
|
| 143 |
+
"logps/rejected": -3141.23974609375,
|
| 144 |
+
"loss": 0.4808,
|
| 145 |
+
"rewards/accuracies": 0.48124998807907104,
|
| 146 |
+
"rewards/chosen": -0.9382199048995972,
|
| 147 |
+
"rewards/margins": 0.18296480178833008,
|
| 148 |
+
"rewards/rejected": -1.1211848258972168,
|
| 149 |
+
"step": 90
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.16,
|
| 153 |
+
"learning_rate": 4.946716615897932e-06,
|
| 154 |
+
"logits/chosen": 0.09833164513111115,
|
| 155 |
+
"logits/rejected": -0.012232898734509945,
|
| 156 |
+
"logps/chosen": -1993.842529296875,
|
| 157 |
+
"logps/rejected": -2496.283935546875,
|
| 158 |
+
"loss": 0.4818,
|
| 159 |
+
"rewards/accuracies": 0.512499988079071,
|
| 160 |
+
"rewards/chosen": -0.8898054957389832,
|
| 161 |
+
"rewards/margins": 0.23708383738994598,
|
| 162 |
+
"rewards/rejected": -1.1268893480300903,
|
| 163 |
+
"step": 100
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.18,
|
| 167 |
+
"learning_rate": 4.9142106826480114e-06,
|
| 168 |
+
"logits/chosen": 0.08790755271911621,
|
| 169 |
+
"logits/rejected": -0.04514267295598984,
|
| 170 |
+
"logps/chosen": -2460.919677734375,
|
| 171 |
+
"logps/rejected": -3302.043701171875,
|
| 172 |
+
"loss": 0.4858,
|
| 173 |
+
"rewards/accuracies": 0.4937500059604645,
|
| 174 |
+
"rewards/chosen": -1.124772310256958,
|
| 175 |
+
"rewards/margins": 0.3031695783138275,
|
| 176 |
+
"rewards/rejected": -1.4279417991638184,
|
| 177 |
+
"step": 110
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.19,
|
| 181 |
+
"learning_rate": 4.874162703221823e-06,
|
| 182 |
+
"logits/chosen": 0.14668218791484833,
|
| 183 |
+
"logits/rejected": 0.010530241765081882,
|
| 184 |
+
"logps/chosen": -2258.22119140625,
|
| 185 |
+
"logps/rejected": -3071.97802734375,
|
| 186 |
+
"loss": 0.4816,
|
| 187 |
+
"rewards/accuracies": 0.543749988079071,
|
| 188 |
+
"rewards/chosen": -1.0068477392196655,
|
| 189 |
+
"rewards/margins": 0.29694581031799316,
|
| 190 |
+
"rewards/rejected": -1.3037936687469482,
|
| 191 |
+
"step": 120
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.21,
|
| 195 |
+
"learning_rate": 4.826697788369752e-06,
|
| 196 |
+
"logits/chosen": 0.15852120518684387,
|
| 197 |
+
"logits/rejected": -0.011427855119109154,
|
| 198 |
+
"logps/chosen": -1801.1246337890625,
|
| 199 |
+
"logps/rejected": -3141.499267578125,
|
| 200 |
+
"loss": 0.4792,
|
| 201 |
+
"rewards/accuracies": 0.5249999761581421,
|
| 202 |
+
"rewards/chosen": -0.7917305827140808,
|
| 203 |
+
"rewards/margins": 0.3859596848487854,
|
| 204 |
+
"rewards/rejected": -1.1776901483535767,
|
| 205 |
+
"step": 130
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.22,
|
| 209 |
+
"learning_rate": 4.7719642195082224e-06,
|
| 210 |
+
"logits/chosen": 0.07224290072917938,
|
| 211 |
+
"logits/rejected": -0.05159113556146622,
|
| 212 |
+
"logps/chosen": -2250.46484375,
|
| 213 |
+
"logps/rejected": -2875.58056640625,
|
| 214 |
+
"loss": 0.4904,
|
| 215 |
+
"rewards/accuracies": 0.46875,
|
| 216 |
+
"rewards/chosen": -1.009852647781372,
|
| 217 |
+
"rewards/margins": 0.02716471627354622,
|
| 218 |
+
"rewards/rejected": -1.0370174646377563,
|
| 219 |
+
"step": 140
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.24,
|
| 223 |
+
"learning_rate": 4.710132985485355e-06,
|
| 224 |
+
"logits/chosen": 0.1444946527481079,
|
| 225 |
+
"logits/rejected": -0.08803781121969223,
|
| 226 |
+
"logps/chosen": -2070.7314453125,
|
| 227 |
+
"logps/rejected": -3544.486328125,
|
| 228 |
+
"loss": 0.4657,
|
| 229 |
+
"rewards/accuracies": 0.574999988079071,
|
| 230 |
+
"rewards/chosen": -1.019829511642456,
|
| 231 |
+
"rewards/margins": 0.6308184862136841,
|
| 232 |
+
"rewards/rejected": -1.6506481170654297,
|
| 233 |
+
"step": 150
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.26,
|
| 237 |
+
"learning_rate": 4.641397248408122e-06,
|
| 238 |
+
"logits/chosen": -0.055302537977695465,
|
| 239 |
+
"logits/rejected": -0.22802452743053436,
|
| 240 |
+
"logps/chosen": -2362.42822265625,
|
| 241 |
+
"logps/rejected": -3568.122314453125,
|
| 242 |
+
"loss": 0.4857,
|
| 243 |
+
"rewards/accuracies": 0.5375000238418579,
|
| 244 |
+
"rewards/chosen": -1.2188419103622437,
|
| 245 |
+
"rewards/margins": 0.24281807243824005,
|
| 246 |
+
"rewards/rejected": -1.4616600275039673,
|
| 247 |
+
"step": 160
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.27,
|
| 251 |
+
"learning_rate": 4.5659717401997655e-06,
|
| 252 |
+
"logits/chosen": -0.04191827401518822,
|
| 253 |
+
"logits/rejected": -0.17326447367668152,
|
| 254 |
+
"logps/chosen": -2562.07080078125,
|
| 255 |
+
"logps/rejected": -3827.65673828125,
|
| 256 |
+
"loss": 0.4818,
|
| 257 |
+
"rewards/accuracies": 0.5,
|
| 258 |
+
"rewards/chosen": -1.30030357837677,
|
| 259 |
+
"rewards/margins": 0.4086272120475769,
|
| 260 |
+
"rewards/rejected": -1.7089307308197021,
|
| 261 |
+
"step": 170
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.29,
|
| 265 |
+
"learning_rate": 4.4840920917726425e-06,
|
| 266 |
+
"logits/chosen": 0.05106606334447861,
|
| 267 |
+
"logits/rejected": -0.09808970242738724,
|
| 268 |
+
"logps/chosen": -2531.837890625,
|
| 269 |
+
"logps/rejected": -3445.98681640625,
|
| 270 |
+
"loss": 0.4688,
|
| 271 |
+
"rewards/accuracies": 0.5687500238418579,
|
| 272 |
+
"rewards/chosen": -1.0362106561660767,
|
| 273 |
+
"rewards/margins": 0.5165188908576965,
|
| 274 |
+
"rewards/rejected": -1.5527294874191284,
|
| 275 |
+
"step": 180
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.3,
|
| 279 |
+
"learning_rate": 4.396014096912182e-06,
|
| 280 |
+
"logits/chosen": -0.020658662542700768,
|
| 281 |
+
"logits/rejected": -0.20878557860851288,
|
| 282 |
+
"logps/chosen": -2704.795166015625,
|
| 283 |
+
"logps/rejected": -3539.75341796875,
|
| 284 |
+
"loss": 0.4847,
|
| 285 |
+
"rewards/accuracies": 0.48750001192092896,
|
| 286 |
+
"rewards/chosen": -1.280983805656433,
|
| 287 |
+
"rewards/margins": 0.08954890817403793,
|
| 288 |
+
"rewards/rejected": -1.370532751083374,
|
| 289 |
+
"step": 190
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.32,
|
| 293 |
+
"learning_rate": 4.302012913171584e-06,
|
| 294 |
+
"logits/chosen": 0.13929739594459534,
|
| 295 |
+
"logits/rejected": 0.010215060785412788,
|
| 296 |
+
"logps/chosen": -2425.408935546875,
|
| 297 |
+
"logps/rejected": -3690.294189453125,
|
| 298 |
+
"loss": 0.4679,
|
| 299 |
+
"rewards/accuracies": 0.543749988079071,
|
| 300 |
+
"rewards/chosen": -0.9245489239692688,
|
| 301 |
+
"rewards/margins": 0.5686434507369995,
|
| 302 |
+
"rewards/rejected": -1.493192434310913,
|
| 303 |
+
"step": 200
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.34,
|
| 307 |
+
"learning_rate": 4.202382202273702e-06,
|
| 308 |
+
"logits/chosen": 0.02957342565059662,
|
| 309 |
+
"logits/rejected": -0.1483859121799469,
|
| 310 |
+
"logps/chosen": -2492.020263671875,
|
| 311 |
+
"logps/rejected": -4000.793701171875,
|
| 312 |
+
"loss": 0.4764,
|
| 313 |
+
"rewards/accuracies": 0.5687500238418579,
|
| 314 |
+
"rewards/chosen": -1.1149667501449585,
|
| 315 |
+
"rewards/margins": 0.5438352823257446,
|
| 316 |
+
"rewards/rejected": -1.6588022708892822,
|
| 317 |
+
"step": 210
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.35,
|
| 321 |
+
"learning_rate": 4.097433212705492e-06,
|
| 322 |
+
"logits/chosen": -0.10198907554149628,
|
| 323 |
+
"logits/rejected": -0.25369173288345337,
|
| 324 |
+
"logps/chosen": -2015.391845703125,
|
| 325 |
+
"logps/rejected": -3833.490966796875,
|
| 326 |
+
"loss": 0.4626,
|
| 327 |
+
"rewards/accuracies": 0.5687500238418579,
|
| 328 |
+
"rewards/chosen": -0.9975125193595886,
|
| 329 |
+
"rewards/margins": 0.9066341519355774,
|
| 330 |
+
"rewards/rejected": -1.9041467905044556,
|
| 331 |
+
"step": 220
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.37,
|
| 335 |
+
"learning_rate": 3.987493807371033e-06,
|
| 336 |
+
"logits/chosen": -0.04106748104095459,
|
| 337 |
+
"logits/rejected": -0.21337568759918213,
|
| 338 |
+
"logps/chosen": -2646.331787109375,
|
| 339 |
+
"logps/rejected": -3494.35302734375,
|
| 340 |
+
"loss": 0.4828,
|
| 341 |
+
"rewards/accuracies": 0.550000011920929,
|
| 342 |
+
"rewards/chosen": -1.4516589641571045,
|
| 343 |
+
"rewards/margins": 0.18022239208221436,
|
| 344 |
+
"rewards/rejected": -1.6318813562393188,
|
| 345 |
+
"step": 230
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.38,
|
| 349 |
+
"learning_rate": 3.872907439340758e-06,
|
| 350 |
+
"logits/chosen": -0.13159896433353424,
|
| 351 |
+
"logits/rejected": -0.2913353145122528,
|
| 352 |
+
"logps/chosen": -3422.151611328125,
|
| 353 |
+
"logps/rejected": -4263.8818359375,
|
| 354 |
+
"loss": 0.4781,
|
| 355 |
+
"rewards/accuracies": 0.53125,
|
| 356 |
+
"rewards/chosen": -1.7441692352294922,
|
| 357 |
+
"rewards/margins": 0.3363359272480011,
|
| 358 |
+
"rewards/rejected": -2.080505132675171,
|
| 359 |
+
"step": 240
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.4,
|
| 363 |
+
"learning_rate": 3.75403207889666e-06,
|
| 364 |
+
"logits/chosen": -0.04592425376176834,
|
| 365 |
+
"logits/rejected": -0.269358366727829,
|
| 366 |
+
"logps/chosen": -1998.029541015625,
|
| 367 |
+
"logps/rejected": -3464.591796875,
|
| 368 |
+
"loss": 0.4739,
|
| 369 |
+
"rewards/accuracies": 0.574999988079071,
|
| 370 |
+
"rewards/chosen": -0.8778263926506042,
|
| 371 |
+
"rewards/margins": 0.5411949157714844,
|
| 372 |
+
"rewards/rejected": -1.4190213680267334,
|
| 373 |
+
"step": 250
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.42,
|
| 377 |
+
"learning_rate": 3.631239095225417e-06,
|
| 378 |
+
"logits/chosen": 0.02863461896777153,
|
| 379 |
+
"logits/rejected": -0.086650051176548,
|
| 380 |
+
"logps/chosen": -1820.9495849609375,
|
| 381 |
+
"logps/rejected": -3405.172607421875,
|
| 382 |
+
"loss": 0.474,
|
| 383 |
+
"rewards/accuracies": 0.5375000238418579,
|
| 384 |
+
"rewards/chosen": -0.7010027170181274,
|
| 385 |
+
"rewards/margins": 0.5200620889663696,
|
| 386 |
+
"rewards/rejected": -1.2210649251937866,
|
| 387 |
+
"step": 260
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.43,
|
| 391 |
+
"learning_rate": 3.5049120962530608e-06,
|
| 392 |
+
"logits/chosen": 0.05603921413421631,
|
| 393 |
+
"logits/rejected": -0.19317272305488586,
|
| 394 |
+
"logps/chosen": -1709.980224609375,
|
| 395 |
+
"logps/rejected": -3665.258544921875,
|
| 396 |
+
"loss": 0.4608,
|
| 397 |
+
"rewards/accuracies": 0.612500011920929,
|
| 398 |
+
"rewards/chosen": -0.6061512231826782,
|
| 399 |
+
"rewards/margins": 0.5793715715408325,
|
| 400 |
+
"rewards/rejected": -1.1855227947235107,
|
| 401 |
+
"step": 270
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.45,
|
| 405 |
+
"learning_rate": 3.3754457302455464e-06,
|
| 406 |
+
"logits/chosen": 0.01207334827631712,
|
| 407 |
+
"logits/rejected": -0.21013681590557098,
|
| 408 |
+
"logps/chosen": -1814.6568603515625,
|
| 409 |
+
"logps/rejected": -3202.33154296875,
|
| 410 |
+
"loss": 0.4768,
|
| 411 |
+
"rewards/accuracies": 0.5687500238418579,
|
| 412 |
+
"rewards/chosen": -0.7476822137832642,
|
| 413 |
+
"rewards/margins": 0.5972086191177368,
|
| 414 |
+
"rewards/rejected": -1.344890832901001,
|
| 415 |
+
"step": 280
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.46,
|
| 419 |
+
"learning_rate": 3.2432444529190714e-06,
|
| 420 |
+
"logits/chosen": -0.016334302723407745,
|
| 421 |
+
"logits/rejected": -0.20580680668354034,
|
| 422 |
+
"logps/chosen": -2087.526123046875,
|
| 423 |
+
"logps/rejected": -2992.07861328125,
|
| 424 |
+
"loss": 0.4821,
|
| 425 |
+
"rewards/accuracies": 0.518750011920929,
|
| 426 |
+
"rewards/chosen": -0.8422229886054993,
|
| 427 |
+
"rewards/margins": 0.2464129477739334,
|
| 428 |
+
"rewards/rejected": -1.0886359214782715,
|
| 429 |
+
"step": 290
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.48,
|
| 433 |
+
"learning_rate": 3.1087212639117057e-06,
|
| 434 |
+
"logits/chosen": -0.05513416603207588,
|
| 435 |
+
"logits/rejected": -0.2065565139055252,
|
| 436 |
+
"logps/chosen": -1855.8892822265625,
|
| 437 |
+
"logps/rejected": -3107.00634765625,
|
| 438 |
+
"loss": 0.4737,
|
| 439 |
+
"rewards/accuracies": 0.581250011920929,
|
| 440 |
+
"rewards/chosen": -0.7160204648971558,
|
| 441 |
+
"rewards/margins": 0.4110482335090637,
|
| 442 |
+
"rewards/rejected": -1.1270686388015747,
|
| 443 |
+
"step": 300
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.5,
|
| 447 |
+
"learning_rate": 2.9722964165636263e-06,
|
| 448 |
+
"logits/chosen": -0.10767300426959991,
|
| 449 |
+
"logits/rejected": -0.1805376410484314,
|
| 450 |
+
"logps/chosen": -2242.067138671875,
|
| 451 |
+
"logps/rejected": -3101.77978515625,
|
| 452 |
+
"loss": 0.4736,
|
| 453 |
+
"rewards/accuracies": 0.512499988079071,
|
| 454 |
+
"rewards/chosen": -0.8544082641601562,
|
| 455 |
+
"rewards/margins": 0.3649452030658722,
|
| 456 |
+
"rewards/rejected": -1.219353437423706,
|
| 457 |
+
"step": 310
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.51,
|
| 461 |
+
"learning_rate": 2.8343961050366275e-06,
|
| 462 |
+
"logits/chosen": -0.0001668736367719248,
|
| 463 |
+
"logits/rejected": -0.181897833943367,
|
| 464 |
+
"logps/chosen": -1954.7802734375,
|
| 465 |
+
"logps/rejected": -3072.43994140625,
|
| 466 |
+
"loss": 0.4668,
|
| 467 |
+
"rewards/accuracies": 0.59375,
|
| 468 |
+
"rewards/chosen": -0.8285297155380249,
|
| 469 |
+
"rewards/margins": 0.40796583890914917,
|
| 470 |
+
"rewards/rejected": -1.2364954948425293,
|
| 471 |
+
"step": 320
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.53,
|
| 475 |
+
"learning_rate": 2.695451132874385e-06,
|
| 476 |
+
"logits/chosen": -0.18868280947208405,
|
| 477 |
+
"logits/rejected": -0.33176952600479126,
|
| 478 |
+
"logps/chosen": -2518.55810546875,
|
| 479 |
+
"logps/rejected": -3725.364501953125,
|
| 480 |
+
"loss": 0.4799,
|
| 481 |
+
"rewards/accuracies": 0.518750011920929,
|
| 482 |
+
"rewards/chosen": -1.2756831645965576,
|
| 483 |
+
"rewards/margins": 0.33692947030067444,
|
| 484 |
+
"rewards/rejected": -1.6126124858856201,
|
| 485 |
+
"step": 330
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.54,
|
| 489 |
+
"learning_rate": 2.5558955671628964e-06,
|
| 490 |
+
"logits/chosen": -0.18861006200313568,
|
| 491 |
+
"logits/rejected": -0.3129253387451172,
|
| 492 |
+
"logps/chosen": -2410.373046875,
|
| 493 |
+
"logps/rejected": -3373.586669921875,
|
| 494 |
+
"loss": 0.4786,
|
| 495 |
+
"rewards/accuracies": 0.48750001192092896,
|
| 496 |
+
"rewards/chosen": -1.1353819370269775,
|
| 497 |
+
"rewards/margins": 0.3919072449207306,
|
| 498 |
+
"rewards/rejected": -1.5272890329360962,
|
| 499 |
+
"step": 340
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.56,
|
| 503 |
+
"learning_rate": 2.4161653824955654e-06,
|
| 504 |
+
"logits/chosen": -0.11331228911876678,
|
| 505 |
+
"logits/rejected": -0.3467509150505066,
|
| 506 |
+
"logps/chosen": -2082.77734375,
|
| 507 |
+
"logps/rejected": -3978.74560546875,
|
| 508 |
+
"loss": 0.4579,
|
| 509 |
+
"rewards/accuracies": 0.581250011920929,
|
| 510 |
+
"rewards/chosen": -1.0137884616851807,
|
| 511 |
+
"rewards/margins": 0.7807275056838989,
|
| 512 |
+
"rewards/rejected": -1.79451584815979,
|
| 513 |
+
"step": 350
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.58,
|
| 517 |
+
"learning_rate": 2.2766970989791697e-06,
|
| 518 |
+
"logits/chosen": -0.06516402959823608,
|
| 519 |
+
"logits/rejected": -0.3027499318122864,
|
| 520 |
+
"logps/chosen": -1849.7210693359375,
|
| 521 |
+
"logps/rejected": -3917.53369140625,
|
| 522 |
+
"loss": 0.4655,
|
| 523 |
+
"rewards/accuracies": 0.612500011920929,
|
| 524 |
+
"rewards/chosen": -0.8109678030014038,
|
| 525 |
+
"rewards/margins": 0.7725037336349487,
|
| 526 |
+
"rewards/rejected": -1.583471655845642,
|
| 527 |
+
"step": 360
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.59,
|
| 531 |
+
"learning_rate": 2.1379264185356545e-06,
|
| 532 |
+
"logits/chosen": -0.0026438175700604916,
|
| 533 |
+
"logits/rejected": -0.0653844028711319,
|
| 534 |
+
"logps/chosen": -2357.537841796875,
|
| 535 |
+
"logps/rejected": -3069.9169921875,
|
| 536 |
+
"loss": 0.4782,
|
| 537 |
+
"rewards/accuracies": 0.53125,
|
| 538 |
+
"rewards/chosen": -0.9414833188056946,
|
| 539 |
+
"rewards/margins": 0.37831395864486694,
|
| 540 |
+
"rewards/rejected": -1.3197972774505615,
|
| 541 |
+
"step": 370
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.61,
|
| 545 |
+
"learning_rate": 2.000286863759934e-06,
|
| 546 |
+
"logits/chosen": -0.06546775251626968,
|
| 547 |
+
"logits/rejected": -0.2656891942024231,
|
| 548 |
+
"logps/chosen": -2187.612060546875,
|
| 549 |
+
"logps/rejected": -4072.555908203125,
|
| 550 |
+
"loss": 0.4689,
|
| 551 |
+
"rewards/accuracies": 0.518750011920929,
|
| 552 |
+
"rewards/chosen": -0.9218450784683228,
|
| 553 |
+
"rewards/margins": 0.6684916019439697,
|
| 554 |
+
"rewards/rejected": -1.5903366804122925,
|
| 555 |
+
"step": 380
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.62,
|
| 559 |
+
"learning_rate": 1.8642084235859764e-06,
|
| 560 |
+
"logits/chosen": 0.09177455306053162,
|
| 561 |
+
"logits/rejected": -0.10396850109100342,
|
| 562 |
+
"logps/chosen": -1575.6951904296875,
|
| 563 |
+
"logps/rejected": -3058.1181640625,
|
| 564 |
+
"loss": 0.4711,
|
| 565 |
+
"rewards/accuracies": 0.53125,
|
| 566 |
+
"rewards/chosen": -0.597963809967041,
|
| 567 |
+
"rewards/margins": 0.5352895855903625,
|
| 568 |
+
"rewards/rejected": -1.1332534551620483,
|
| 569 |
+
"step": 390
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.64,
|
| 573 |
+
"learning_rate": 1.7301162099921013e-06,
|
| 574 |
+
"logits/chosen": 0.021384917199611664,
|
| 575 |
+
"logits/rejected": -0.25088971853256226,
|
| 576 |
+
"logps/chosen": -2101.90771484375,
|
| 577 |
+
"logps/rejected": -4023.61328125,
|
| 578 |
+
"loss": 0.4703,
|
| 579 |
+
"rewards/accuracies": 0.5375000238418579,
|
| 580 |
+
"rewards/chosen": -0.9023422002792358,
|
| 581 |
+
"rewards/margins": 0.7357300519943237,
|
| 582 |
+
"rewards/rejected": -1.6380722522735596,
|
| 583 |
+
"step": 400
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.66,
|
| 587 |
+
"learning_rate": 1.5984291299420117e-06,
|
| 588 |
+
"logits/chosen": -0.030091216787695885,
|
| 589 |
+
"logits/rejected": -0.3482876420021057,
|
| 590 |
+
"logps/chosen": -2085.20166015625,
|
| 591 |
+
"logps/rejected": -3580.270751953125,
|
| 592 |
+
"loss": 0.4731,
|
| 593 |
+
"rewards/accuracies": 0.6312500238418579,
|
| 594 |
+
"rewards/chosen": -0.902575671672821,
|
| 595 |
+
"rewards/margins": 0.486042320728302,
|
| 596 |
+
"rewards/rejected": -1.388617992401123,
|
| 597 |
+
"step": 410
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.67,
|
| 601 |
+
"learning_rate": 1.4695585767104092e-06,
|
| 602 |
+
"logits/chosen": -0.009017439559102058,
|
| 603 |
+
"logits/rejected": -0.24702921509742737,
|
| 604 |
+
"logps/chosen": -2445.38037109375,
|
| 605 |
+
"logps/rejected": -3752.704345703125,
|
| 606 |
+
"loss": 0.4784,
|
| 607 |
+
"rewards/accuracies": 0.543749988079071,
|
| 608 |
+
"rewards/chosen": -1.0142749547958374,
|
| 609 |
+
"rewards/margins": 0.40337714552879333,
|
| 610 |
+
"rewards/rejected": -1.4176521301269531,
|
| 611 |
+
"step": 420
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.69,
|
| 615 |
+
"learning_rate": 1.3439071446815452e-06,
|
| 616 |
+
"logits/chosen": 0.01383951585739851,
|
| 617 |
+
"logits/rejected": -0.25569045543670654,
|
| 618 |
+
"logps/chosen": -1831.669189453125,
|
| 619 |
+
"logps/rejected": -3227.48046875,
|
| 620 |
+
"loss": 0.4616,
|
| 621 |
+
"rewards/accuracies": 0.6312500238418579,
|
| 622 |
+
"rewards/chosen": -0.6138505339622498,
|
| 623 |
+
"rewards/margins": 0.5082579851150513,
|
| 624 |
+
"rewards/rejected": -1.1221085786819458,
|
| 625 |
+
"step": 430
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.7,
|
| 629 |
+
"learning_rate": 1.2218673716356919e-06,
|
| 630 |
+
"logits/chosen": -0.13071729242801666,
|
| 631 |
+
"logits/rejected": -0.2272064983844757,
|
| 632 |
+
"logps/chosen": -2428.37451171875,
|
| 633 |
+
"logps/rejected": -2997.77197265625,
|
| 634 |
+
"loss": 0.4825,
|
| 635 |
+
"rewards/accuracies": 0.4937500059604645,
|
| 636 |
+
"rewards/chosen": -0.9212499856948853,
|
| 637 |
+
"rewards/margins": 0.19097386300563812,
|
| 638 |
+
"rewards/rejected": -1.1122238636016846,
|
| 639 |
+
"step": 440
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.72,
|
| 643 |
+
"learning_rate": 1.103820512452661e-06,
|
| 644 |
+
"logits/chosen": -0.07564810663461685,
|
| 645 |
+
"logits/rejected": -0.3487449586391449,
|
| 646 |
+
"logps/chosen": -2048.64697265625,
|
| 647 |
+
"logps/rejected": -3798.810546875,
|
| 648 |
+
"loss": 0.4649,
|
| 649 |
+
"rewards/accuracies": 0.5687500238418579,
|
| 650 |
+
"rewards/chosen": -0.8849633932113647,
|
| 651 |
+
"rewards/margins": 0.6834510564804077,
|
| 652 |
+
"rewards/rejected": -1.5684144496917725,
|
| 653 |
+
"step": 450
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.74,
|
| 657 |
+
"learning_rate": 9.901353480633468e-07,
|
| 658 |
+
"logits/chosen": -0.10444183647632599,
|
| 659 |
+
"logits/rejected": -0.23747725784778595,
|
| 660 |
+
"logps/chosen": -2240.57470703125,
|
| 661 |
+
"logps/rejected": -3905.287109375,
|
| 662 |
+
"loss": 0.4685,
|
| 663 |
+
"rewards/accuracies": 0.550000011920929,
|
| 664 |
+
"rewards/chosen": -0.976617157459259,
|
| 665 |
+
"rewards/margins": 0.6172038912773132,
|
| 666 |
+
"rewards/rejected": -1.5938211679458618,
|
| 667 |
+
"step": 460
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.75,
|
| 671 |
+
"learning_rate": 8.811670333701544e-07,
|
| 672 |
+
"logits/chosen": -0.13727447390556335,
|
| 673 |
+
"logits/rejected": -0.23075255751609802,
|
| 674 |
+
"logps/chosen": -2049.42041015625,
|
| 675 |
+
"logps/rejected": -3127.796630859375,
|
| 676 |
+
"loss": 0.4709,
|
| 677 |
+
"rewards/accuracies": 0.5562499761581421,
|
| 678 |
+
"rewards/chosen": -0.912371039390564,
|
| 679 |
+
"rewards/margins": 0.5058678984642029,
|
| 680 |
+
"rewards/rejected": -1.418238878250122,
|
| 681 |
+
"step": 470
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.77,
|
| 685 |
+
"learning_rate": 7.772559877354341e-07,
|
| 686 |
+
"logits/chosen": -0.17451250553131104,
|
| 687 |
+
"logits/rejected": -0.405754029750824,
|
| 688 |
+
"logps/chosen": -2040.5706787109375,
|
| 689 |
+
"logps/rejected": -4039.03125,
|
| 690 |
+
"loss": 0.4696,
|
| 691 |
+
"rewards/accuracies": 0.5874999761581421,
|
| 692 |
+
"rewards/chosen": -0.9431853294372559,
|
| 693 |
+
"rewards/margins": 0.766620397567749,
|
| 694 |
+
"rewards/rejected": -1.7098054885864258,
|
| 695 |
+
"step": 480
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.78,
|
| 699 |
+
"learning_rate": 6.787268315040604e-07,
|
| 700 |
+
"logits/chosen": -0.10103428363800049,
|
| 701 |
+
"logits/rejected": -0.24396558105945587,
|
| 702 |
+
"logps/chosen": -2031.975341796875,
|
| 703 |
+
"logps/rejected": -3454.368408203125,
|
| 704 |
+
"loss": 0.4703,
|
| 705 |
+
"rewards/accuracies": 0.5375000238418579,
|
| 706 |
+
"rewards/chosen": -0.9141359329223633,
|
| 707 |
+
"rewards/margins": 0.6935009956359863,
|
| 708 |
+
"rewards/rejected": -1.6076370477676392,
|
| 709 |
+
"step": 490
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.8,
|
| 713 |
+
"learning_rate": 5.858873718824829e-07,
|
| 714 |
+
"logits/chosen": -0.23661558330059052,
|
| 715 |
+
"logits/rejected": -0.30403372645378113,
|
| 716 |
+
"logps/chosen": -2469.70458984375,
|
| 717 |
+
"logps/rejected": -3364.81005859375,
|
| 718 |
+
"loss": 0.4772,
|
| 719 |
+
"rewards/accuracies": 0.512499988079071,
|
| 720 |
+
"rewards/chosen": -1.2715297937393188,
|
| 721 |
+
"rewards/margins": 0.312258243560791,
|
| 722 |
+
"rewards/rejected": -1.5837881565093994,
|
| 723 |
+
"step": 500
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.82,
|
| 727 |
+
"learning_rate": 4.990276413423817e-07,
|
| 728 |
+
"logits/chosen": -0.09825234115123749,
|
| 729 |
+
"logits/rejected": -0.26388177275657654,
|
| 730 |
+
"logps/chosen": -1970.1832275390625,
|
| 731 |
+
"logps/rejected": -3387.107421875,
|
| 732 |
+
"loss": 0.4706,
|
| 733 |
+
"rewards/accuracies": 0.5249999761581421,
|
| 734 |
+
"rewards/chosen": -0.8808355331420898,
|
| 735 |
+
"rewards/margins": 0.6419728994369507,
|
| 736 |
+
"rewards/rejected": -1.5228084325790405,
|
| 737 |
+
"step": 510
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.83,
|
| 741 |
+
"learning_rate": 4.184189915529796e-07,
|
| 742 |
+
"logits/chosen": -0.019504955038428307,
|
| 743 |
+
"logits/rejected": -0.22602617740631104,
|
| 744 |
+
"logps/chosen": -1955.7008056640625,
|
| 745 |
+
"logps/rejected": -3776.561279296875,
|
| 746 |
+
"loss": 0.4598,
|
| 747 |
+
"rewards/accuracies": 0.606249988079071,
|
| 748 |
+
"rewards/chosen": -0.8692920804023743,
|
| 749 |
+
"rewards/margins": 0.9374381303787231,
|
| 750 |
+
"rewards/rejected": -1.8067299127578735,
|
| 751 |
+
"step": 520
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.85,
|
| 755 |
+
"learning_rate": 3.4431324567258176e-07,
|
| 756 |
+
"logits/chosen": -0.1649870127439499,
|
| 757 |
+
"logits/rejected": -0.33468708395957947,
|
| 758 |
+
"logps/chosen": -2207.761474609375,
|
| 759 |
+
"logps/rejected": -4173.55810546875,
|
| 760 |
+
"loss": 0.4613,
|
| 761 |
+
"rewards/accuracies": 0.5562499761581421,
|
| 762 |
+
"rewards/chosen": -0.9728679656982422,
|
| 763 |
+
"rewards/margins": 0.8519523739814758,
|
| 764 |
+
"rewards/rejected": -1.8248205184936523,
|
| 765 |
+
"step": 530
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.86,
|
| 769 |
+
"learning_rate": 2.769419116476052e-07,
|
| 770 |
+
"logits/chosen": -0.2527296841144562,
|
| 771 |
+
"logits/rejected": -0.37009841203689575,
|
| 772 |
+
"logps/chosen": -2005.4554443359375,
|
| 773 |
+
"logps/rejected": -3320.11083984375,
|
| 774 |
+
"loss": 0.4659,
|
| 775 |
+
"rewards/accuracies": 0.606249988079071,
|
| 776 |
+
"rewards/chosen": -0.983248233795166,
|
| 777 |
+
"rewards/margins": 0.48160916566848755,
|
| 778 |
+
"rewards/rejected": -1.4648573398590088,
|
| 779 |
+
"step": 540
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.88,
|
| 783 |
+
"learning_rate": 2.1651545897676512e-07,
|
| 784 |
+
"logits/chosen": -0.1346069872379303,
|
| 785 |
+
"logits/rejected": -0.30417710542678833,
|
| 786 |
+
"logps/chosen": -2149.47998046875,
|
| 787 |
+
"logps/rejected": -3438.53271484375,
|
| 788 |
+
"loss": 0.4659,
|
| 789 |
+
"rewards/accuracies": 0.574999988079071,
|
| 790 |
+
"rewards/chosen": -0.995121955871582,
|
| 791 |
+
"rewards/margins": 0.6242730021476746,
|
| 792 |
+
"rewards/rejected": -1.6193952560424805,
|
| 793 |
+
"step": 550
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.9,
|
| 797 |
+
"learning_rate": 1.6322266119983222e-07,
|
| 798 |
+
"logits/chosen": -0.09836331009864807,
|
| 799 |
+
"logits/rejected": -0.3026123046875,
|
| 800 |
+
"logps/chosen": -2651.5302734375,
|
| 801 |
+
"logps/rejected": -3409.03955078125,
|
| 802 |
+
"loss": 0.4739,
|
| 803 |
+
"rewards/accuracies": 0.53125,
|
| 804 |
+
"rewards/chosen": -1.2515802383422852,
|
| 805 |
+
"rewards/margins": 0.16534823179244995,
|
| 806 |
+
"rewards/rejected": -1.4169285297393799,
|
| 807 |
+
"step": 560
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 0.91,
|
| 811 |
+
"learning_rate": 1.1723000616502167e-07,
|
| 812 |
+
"logits/chosen": -0.08475066721439362,
|
| 813 |
+
"logits/rejected": -0.33103296160697937,
|
| 814 |
+
"logps/chosen": -2303.97900390625,
|
| 815 |
+
"logps/rejected": -3867.08251953125,
|
| 816 |
+
"loss": 0.4667,
|
| 817 |
+
"rewards/accuracies": 0.543749988079071,
|
| 818 |
+
"rewards/chosen": -1.1717913150787354,
|
| 819 |
+
"rewards/margins": 0.6337391138076782,
|
| 820 |
+
"rewards/rejected": -1.805530309677124,
|
| 821 |
+
"step": 570
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 0.93,
|
| 825 |
+
"learning_rate": 7.868117591737585e-08,
|
| 826 |
+
"logits/chosen": -0.1410863846540451,
|
| 827 |
+
"logits/rejected": -0.3231101930141449,
|
| 828 |
+
"logps/chosen": -2047.534423828125,
|
| 829 |
+
"logps/rejected": -3277.69921875,
|
| 830 |
+
"loss": 0.475,
|
| 831 |
+
"rewards/accuracies": 0.5,
|
| 832 |
+
"rewards/chosen": -0.9290479421615601,
|
| 833 |
+
"rewards/margins": 0.6801496744155884,
|
| 834 |
+
"rewards/rejected": -1.6091976165771484,
|
| 835 |
+
"step": 580
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 0.94,
|
| 839 |
+
"learning_rate": 4.769659783295383e-08,
|
| 840 |
+
"logits/chosen": -0.1917143017053604,
|
| 841 |
+
"logits/rejected": -0.3931889832019806,
|
| 842 |
+
"logps/chosen": -2393.9306640625,
|
| 843 |
+
"logps/rejected": -3955.93359375,
|
| 844 |
+
"loss": 0.4784,
|
| 845 |
+
"rewards/accuracies": 0.5062500238418579,
|
| 846 |
+
"rewards/chosen": -1.1227269172668457,
|
| 847 |
+
"rewards/margins": 0.465735524892807,
|
| 848 |
+
"rewards/rejected": -1.5884625911712646,
|
| 849 |
+
"step": 590
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.96,
|
| 853 |
+
"learning_rate": 2.4373068401120358e-08,
|
| 854 |
+
"logits/chosen": -0.10545693337917328,
|
| 855 |
+
"logits/rejected": -0.24310024082660675,
|
| 856 |
+
"logps/chosen": -2470.99560546875,
|
| 857 |
+
"logps/rejected": -3619.756591796875,
|
| 858 |
+
"loss": 0.4742,
|
| 859 |
+
"rewards/accuracies": 0.5375000238418579,
|
| 860 |
+
"rewards/chosen": -1.1268969774246216,
|
| 861 |
+
"rewards/margins": 0.5375250577926636,
|
| 862 |
+
"rewards/rejected": -1.664421796798706,
|
| 863 |
+
"step": 600
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.98,
|
| 867 |
+
"learning_rate": 8.78345083022425e-09,
|
| 868 |
+
"logits/chosen": -0.136544868350029,
|
| 869 |
+
"logits/rejected": -0.3129388391971588,
|
| 870 |
+
"logps/chosen": -2081.79248046875,
|
| 871 |
+
"logps/rejected": -3515.130126953125,
|
| 872 |
+
"loss": 0.4743,
|
| 873 |
+
"rewards/accuracies": 0.5562499761581421,
|
| 874 |
+
"rewards/chosen": -0.9320171475410461,
|
| 875 |
+
"rewards/margins": 0.5496521592140198,
|
| 876 |
+
"rewards/rejected": -1.4816694259643555,
|
| 877 |
+
"step": 610
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 0.99,
|
| 881 |
+
"learning_rate": 9.764474213677654e-10,
|
| 882 |
+
"logits/chosen": -0.1356605440378189,
|
| 883 |
+
"logits/rejected": -0.29532909393310547,
|
| 884 |
+
"logps/chosen": -2376.76953125,
|
| 885 |
+
"logps/rejected": -3284.452392578125,
|
| 886 |
+
"loss": 0.4779,
|
| 887 |
+
"rewards/accuracies": 0.5249999761581421,
|
| 888 |
+
"rewards/chosen": -1.0924612283706665,
|
| 889 |
+
"rewards/margins": 0.41884762048721313,
|
| 890 |
+
"rewards/rejected": -1.5113089084625244,
|
| 891 |
+
"step": 620
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 1.0,
|
| 895 |
+
"step": 625,
|
| 896 |
+
"total_flos": 0.0,
|
| 897 |
+
"train_loss": 0.4757139442443848,
|
| 898 |
+
"train_runtime": 7766.6535,
|
| 899 |
+
"train_samples_per_second": 2.575,
|
| 900 |
+
"train_steps_per_second": 0.08
|
| 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 |
+
}
|