Model save
Browse files- README.md +61 -0
- adapter_model.safetensors +1 -1
- all_results.json +8 -0
- emissions.csv +2 -0
- runs/Sep17_23-02-35_gpu4-119-5/events.out.tfevents.1726578203.gpu4-119-5.2889638.0 +2 -2
- train_results.json +8 -0
- trainer_state.json +926 -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-ipo-chatml-lora-20k-40k-i1-merged
|
| 8 |
+
model-index:
|
| 9 |
+
- name: phi-2-ipo-chatml-lora-i2
|
| 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-ipo-chatml-lora-i2
|
| 17 |
+
|
| 18 |
+
This model is a fine-tuned version of [DUAL-GPO/phi-2-ipo-chatml-lora-20k-40k-i1-merged](https://huggingface.co/DUAL-GPO/phi-2-ipo-chatml-lora-20k-40k-i1-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 sha256:82ac03ea52b11d2b00671bc5e33ec8893b5403961fc2136557122ad5766ad267
|
| 3 |
size 335579632
|
all_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"train_loss": 219463.76923076922,
|
| 4 |
+
"train_runtime": 8069.1141,
|
| 5 |
+
"train_samples": 20378,
|
| 6 |
+
"train_samples_per_second": 2.525,
|
| 7 |
+
"train_steps_per_second": 0.079
|
| 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 |
+
2024-09-18T01:17:52,codecarbon,c30909b4-6887-4050-9a4c-72115b144b87,8069.119428157806,0.006541673913533137,8.107048076033511e-07,42.5,890.919,188.74309015274048,0.09525928991585973,2.2346971822524178,0.42186188896298993,2.7518183611312668,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/Sep17_23-02-35_gpu4-119-5/events.out.tfevents.1726578203.gpu4-119-5.2889638.0
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:a35c04714af27340723365d29095c23b6d76be24860893f0ba9812ad3aaf5789
|
| 3 |
+
size 45717
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"train_loss": 219463.76923076922,
|
| 4 |
+
"train_runtime": 8069.1141,
|
| 5 |
+
"train_samples": 20378,
|
| 6 |
+
"train_samples_per_second": 2.525,
|
| 7 |
+
"train_steps_per_second": 0.079
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,926 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 1.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 637,
|
| 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": 7.8125e-08,
|
| 14 |
+
"logits/chosen": 0.14330744743347168,
|
| 15 |
+
"logits/rejected": 0.18303190171718597,
|
| 16 |
+
"logps/chosen": -714.194580078125,
|
| 17 |
+
"logps/rejected": -783.393798828125,
|
| 18 |
+
"loss": 250000.0,
|
| 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": 7.8125e-07,
|
| 28 |
+
"logits/chosen": 0.1265290379524231,
|
| 29 |
+
"logits/rejected": 0.17839111387729645,
|
| 30 |
+
"logps/chosen": -714.8074340820312,
|
| 31 |
+
"logps/rejected": -733.1469116210938,
|
| 32 |
+
"loss": 250078.1667,
|
| 33 |
+
"rewards/accuracies": 0.3611111044883728,
|
| 34 |
+
"rewards/chosen": -0.00013909974950365722,
|
| 35 |
+
"rewards/margins": -0.00010831771942321211,
|
| 36 |
+
"rewards/rejected": -3.078207009821199e-05,
|
| 37 |
+
"step": 10
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.03,
|
| 41 |
+
"learning_rate": 1.5625e-06,
|
| 42 |
+
"logits/chosen": 0.11676273494958878,
|
| 43 |
+
"logits/rejected": 0.15343983471393585,
|
| 44 |
+
"logps/chosen": -693.2666015625,
|
| 45 |
+
"logps/rejected": -712.1083374023438,
|
| 46 |
+
"loss": 249833.05,
|
| 47 |
+
"rewards/accuracies": 0.5,
|
| 48 |
+
"rewards/chosen": -0.0006469396757893264,
|
| 49 |
+
"rewards/margins": 0.00024659099290147424,
|
| 50 |
+
"rewards/rejected": -0.0008935307268984616,
|
| 51 |
+
"step": 20
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.05,
|
| 55 |
+
"learning_rate": 2.3437500000000002e-06,
|
| 56 |
+
"logits/chosen": 0.09228599816560745,
|
| 57 |
+
"logits/rejected": 0.14996477961540222,
|
| 58 |
+
"logps/chosen": -694.0119018554688,
|
| 59 |
+
"logps/rejected": -726.91796875,
|
| 60 |
+
"loss": 249133.3,
|
| 61 |
+
"rewards/accuracies": 0.5687500238418579,
|
| 62 |
+
"rewards/chosen": -0.005162687040865421,
|
| 63 |
+
"rewards/margins": 0.0009325396385975182,
|
| 64 |
+
"rewards/rejected": -0.006095226388424635,
|
| 65 |
+
"step": 30
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.06,
|
| 69 |
+
"learning_rate": 3.125e-06,
|
| 70 |
+
"logits/chosen": 0.05852314084768295,
|
| 71 |
+
"logits/rejected": 0.1720169633626938,
|
| 72 |
+
"logps/chosen": -751.1815795898438,
|
| 73 |
+
"logps/rejected": -776.6752319335938,
|
| 74 |
+
"loss": 247264.85,
|
| 75 |
+
"rewards/accuracies": 0.606249988079071,
|
| 76 |
+
"rewards/chosen": -0.01876734383404255,
|
| 77 |
+
"rewards/margins": 0.0032565854489803314,
|
| 78 |
+
"rewards/rejected": -0.02202392928302288,
|
| 79 |
+
"step": 40
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.08,
|
| 83 |
+
"learning_rate": 3.90625e-06,
|
| 84 |
+
"logits/chosen": 0.10499389469623566,
|
| 85 |
+
"logits/rejected": 0.16752108931541443,
|
| 86 |
+
"logps/chosen": -779.3970947265625,
|
| 87 |
+
"logps/rejected": -821.4215698242188,
|
| 88 |
+
"loss": 243623.9,
|
| 89 |
+
"rewards/accuracies": 0.581250011920929,
|
| 90 |
+
"rewards/chosen": -0.04817109555006027,
|
| 91 |
+
"rewards/margins": 0.007549160160124302,
|
| 92 |
+
"rewards/rejected": -0.0557202585041523,
|
| 93 |
+
"step": 50
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.09,
|
| 97 |
+
"learning_rate": 4.6875000000000004e-06,
|
| 98 |
+
"logits/chosen": 0.10056255757808685,
|
| 99 |
+
"logits/rejected": 0.1355510652065277,
|
| 100 |
+
"logps/chosen": -761.1482543945312,
|
| 101 |
+
"logps/rejected": -791.7299194335938,
|
| 102 |
+
"loss": 241589.05,
|
| 103 |
+
"rewards/accuracies": 0.512499988079071,
|
| 104 |
+
"rewards/chosen": -0.06844565272331238,
|
| 105 |
+
"rewards/margins": 0.00780609343200922,
|
| 106 |
+
"rewards/rejected": -0.07625173777341843,
|
| 107 |
+
"step": 60
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.11,
|
| 111 |
+
"learning_rate": 4.998647417232375e-06,
|
| 112 |
+
"logits/chosen": 0.08108599483966827,
|
| 113 |
+
"logits/rejected": 0.12395985424518585,
|
| 114 |
+
"logps/chosen": -819.1280517578125,
|
| 115 |
+
"logps/rejected": -899.9339599609375,
|
| 116 |
+
"loss": 236403.25,
|
| 117 |
+
"rewards/accuracies": 0.518750011920929,
|
| 118 |
+
"rewards/chosen": -0.0921686664223671,
|
| 119 |
+
"rewards/margins": 0.011640750803053379,
|
| 120 |
+
"rewards/rejected": -0.1038094162940979,
|
| 121 |
+
"step": 70
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.13,
|
| 125 |
+
"learning_rate": 4.990386933279973e-06,
|
| 126 |
+
"logits/chosen": 0.07269687950611115,
|
| 127 |
+
"logits/rejected": 0.12636110186576843,
|
| 128 |
+
"logps/chosen": -799.1640625,
|
| 129 |
+
"logps/rejected": -871.1455078125,
|
| 130 |
+
"loss": 228858.95,
|
| 131 |
+
"rewards/accuracies": 0.59375,
|
| 132 |
+
"rewards/chosen": -0.09310872852802277,
|
| 133 |
+
"rewards/margins": 0.03275894373655319,
|
| 134 |
+
"rewards/rejected": -0.12586766481399536,
|
| 135 |
+
"step": 80
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.14,
|
| 139 |
+
"learning_rate": 4.974642195009681e-06,
|
| 140 |
+
"logits/chosen": 0.059646714478731155,
|
| 141 |
+
"logits/rejected": 0.11064349114894867,
|
| 142 |
+
"logps/chosen": -906.8914184570312,
|
| 143 |
+
"logps/rejected": -967.6486206054688,
|
| 144 |
+
"loss": 231742.65,
|
| 145 |
+
"rewards/accuracies": 0.550000011920929,
|
| 146 |
+
"rewards/chosen": -0.17958079278469086,
|
| 147 |
+
"rewards/margins": 0.026337388902902603,
|
| 148 |
+
"rewards/rejected": -0.20591816306114197,
|
| 149 |
+
"step": 90
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.16,
|
| 153 |
+
"learning_rate": 4.951460519416228e-06,
|
| 154 |
+
"logits/chosen": 0.09507628530263901,
|
| 155 |
+
"logits/rejected": 0.06315924227237701,
|
| 156 |
+
"logps/chosen": -831.5269775390625,
|
| 157 |
+
"logps/rejected": -934.2014770507812,
|
| 158 |
+
"loss": 228067.35,
|
| 159 |
+
"rewards/accuracies": 0.550000011920929,
|
| 160 |
+
"rewards/chosen": -0.16508157551288605,
|
| 161 |
+
"rewards/margins": 0.02746037021279335,
|
| 162 |
+
"rewards/rejected": -0.1925419270992279,
|
| 163 |
+
"step": 100
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.17,
|
| 167 |
+
"learning_rate": 4.920911573406925e-06,
|
| 168 |
+
"logits/chosen": 0.08507482707500458,
|
| 169 |
+
"logits/rejected": 0.06426723301410675,
|
| 170 |
+
"logps/chosen": -886.0023193359375,
|
| 171 |
+
"logps/rejected": -968.9869384765625,
|
| 172 |
+
"loss": 231192.4,
|
| 173 |
+
"rewards/accuracies": 0.53125,
|
| 174 |
+
"rewards/chosen": -0.19480349123477936,
|
| 175 |
+
"rewards/margins": 0.028372664004564285,
|
| 176 |
+
"rewards/rejected": -0.22317615151405334,
|
| 177 |
+
"step": 110
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.19,
|
| 181 |
+
"learning_rate": 4.883087164434672e-06,
|
| 182 |
+
"logits/chosen": 0.12870684266090393,
|
| 183 |
+
"logits/rejected": 0.19525791704654694,
|
| 184 |
+
"logps/chosen": -918.3961181640625,
|
| 185 |
+
"logps/rejected": -981.2435302734375,
|
| 186 |
+
"loss": 228031.675,
|
| 187 |
+
"rewards/accuracies": 0.606249988079071,
|
| 188 |
+
"rewards/chosen": -0.16016361117362976,
|
| 189 |
+
"rewards/margins": 0.033955834805965424,
|
| 190 |
+
"rewards/rejected": -0.1941194236278534,
|
| 191 |
+
"step": 120
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.2,
|
| 195 |
+
"learning_rate": 4.8381009645929044e-06,
|
| 196 |
+
"logits/chosen": 0.19024774432182312,
|
| 197 |
+
"logits/rejected": 0.1789695769548416,
|
| 198 |
+
"logps/chosen": -888.7951049804688,
|
| 199 |
+
"logps/rejected": -1009.6251220703125,
|
| 200 |
+
"loss": 225728.075,
|
| 201 |
+
"rewards/accuracies": 0.625,
|
| 202 |
+
"rewards/chosen": -0.218052938580513,
|
| 203 |
+
"rewards/margins": 0.05132851004600525,
|
| 204 |
+
"rewards/rejected": -0.26938146352767944,
|
| 205 |
+
"step": 130
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.22,
|
| 209 |
+
"learning_rate": 4.786088169001671e-06,
|
| 210 |
+
"logits/chosen": 0.17952367663383484,
|
| 211 |
+
"logits/rejected": 0.15534032881259918,
|
| 212 |
+
"logps/chosen": -932.0501098632812,
|
| 213 |
+
"logps/rejected": -1063.305419921875,
|
| 214 |
+
"loss": 228587.2,
|
| 215 |
+
"rewards/accuracies": 0.606249988079071,
|
| 216 |
+
"rewards/chosen": -0.2622358500957489,
|
| 217 |
+
"rewards/margins": 0.05344052240252495,
|
| 218 |
+
"rewards/rejected": -0.31567639112472534,
|
| 219 |
+
"step": 140
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.24,
|
| 223 |
+
"learning_rate": 4.727205089511466e-06,
|
| 224 |
+
"logits/chosen": 0.1564352661371231,
|
| 225 |
+
"logits/rejected": 0.1726730763912201,
|
| 226 |
+
"logps/chosen": -859.6404418945312,
|
| 227 |
+
"logps/rejected": -931.6002807617188,
|
| 228 |
+
"loss": 237987.3,
|
| 229 |
+
"rewards/accuracies": 0.53125,
|
| 230 |
+
"rewards/chosen": -0.1820591390132904,
|
| 231 |
+
"rewards/margins": 0.03248381242156029,
|
| 232 |
+
"rewards/rejected": -0.2145429402589798,
|
| 233 |
+
"step": 150
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.25,
|
| 237 |
+
"learning_rate": 4.661628684945851e-06,
|
| 238 |
+
"logits/chosen": 0.1257745772600174,
|
| 239 |
+
"logits/rejected": 0.18414214253425598,
|
| 240 |
+
"logps/chosen": -935.7888793945312,
|
| 241 |
+
"logps/rejected": -1013.6844482421875,
|
| 242 |
+
"loss": 227312.15,
|
| 243 |
+
"rewards/accuracies": 0.59375,
|
| 244 |
+
"rewards/chosen": -0.18025702238082886,
|
| 245 |
+
"rewards/margins": 0.035888925194740295,
|
| 246 |
+
"rewards/rejected": -0.21614596247673035,
|
| 247 |
+
"step": 160
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.27,
|
| 251 |
+
"learning_rate": 4.5895560292946e-06,
|
| 252 |
+
"logits/chosen": 0.2171991616487503,
|
| 253 |
+
"logits/rejected": 0.2624339461326599,
|
| 254 |
+
"logps/chosen": -1001.1527099609375,
|
| 255 |
+
"logps/rejected": -1123.304443359375,
|
| 256 |
+
"loss": 217504.55,
|
| 257 |
+
"rewards/accuracies": 0.612500011920929,
|
| 258 |
+
"rewards/chosen": -0.2904219925403595,
|
| 259 |
+
"rewards/margins": 0.05342133715748787,
|
| 260 |
+
"rewards/rejected": -0.34384334087371826,
|
| 261 |
+
"step": 170
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.28,
|
| 265 |
+
"learning_rate": 4.511203719455588e-06,
|
| 266 |
+
"logits/chosen": 0.17157816886901855,
|
| 267 |
+
"logits/rejected": 0.3099578320980072,
|
| 268 |
+
"logps/chosen": -1003.0543212890625,
|
| 269 |
+
"logps/rejected": -1025.053955078125,
|
| 270 |
+
"loss": 231876.85,
|
| 271 |
+
"rewards/accuracies": 0.5249999761581421,
|
| 272 |
+
"rewards/chosen": -0.28897398710250854,
|
| 273 |
+
"rewards/margins": 0.03390089422464371,
|
| 274 |
+
"rewards/rejected": -0.32287487387657166,
|
| 275 |
+
"step": 180
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.3,
|
| 279 |
+
"learning_rate": 4.426807224305315e-06,
|
| 280 |
+
"logits/chosen": 0.17717203497886658,
|
| 281 |
+
"logits/rejected": 0.23975355923175812,
|
| 282 |
+
"logps/chosen": -887.27294921875,
|
| 283 |
+
"logps/rejected": -988.5928955078125,
|
| 284 |
+
"loss": 219205.3,
|
| 285 |
+
"rewards/accuracies": 0.637499988079071,
|
| 286 |
+
"rewards/chosen": -0.2065242975950241,
|
| 287 |
+
"rewards/margins": 0.056366823613643646,
|
| 288 |
+
"rewards/rejected": -0.26289111375808716,
|
| 289 |
+
"step": 190
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.31,
|
| 293 |
+
"learning_rate": 4.336620177054269e-06,
|
| 294 |
+
"logits/chosen": 0.19569167494773865,
|
| 295 |
+
"logits/rejected": 0.23822470009326935,
|
| 296 |
+
"logps/chosen": -890.8048706054688,
|
| 297 |
+
"logps/rejected": -993.693359375,
|
| 298 |
+
"loss": 222280.1,
|
| 299 |
+
"rewards/accuracies": 0.6187499761581421,
|
| 300 |
+
"rewards/chosen": -0.23872694373130798,
|
| 301 |
+
"rewards/margins": 0.05283508822321892,
|
| 302 |
+
"rewards/rejected": -0.291562020778656,
|
| 303 |
+
"step": 200
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.33,
|
| 307 |
+
"learning_rate": 4.240913613013785e-06,
|
| 308 |
+
"logits/chosen": 0.2506358325481415,
|
| 309 |
+
"logits/rejected": 0.184006929397583,
|
| 310 |
+
"logps/chosen": -1007.4181518554688,
|
| 311 |
+
"logps/rejected": -1067.2701416015625,
|
| 312 |
+
"loss": 221823.5,
|
| 313 |
+
"rewards/accuracies": 0.5625,
|
| 314 |
+
"rewards/chosen": -0.2965286672115326,
|
| 315 |
+
"rewards/margins": 0.035524263978004456,
|
| 316 |
+
"rewards/rejected": -0.33205297589302063,
|
| 317 |
+
"step": 210
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.35,
|
| 321 |
+
"learning_rate": 4.139975155065109e-06,
|
| 322 |
+
"logits/chosen": 0.2400396764278412,
|
| 323 |
+
"logits/rejected": 0.29167982935905457,
|
| 324 |
+
"logps/chosen": -1041.81103515625,
|
| 325 |
+
"logps/rejected": -1105.313720703125,
|
| 326 |
+
"loss": 219784.95,
|
| 327 |
+
"rewards/accuracies": 0.637499988079071,
|
| 328 |
+
"rewards/chosen": -0.2964867353439331,
|
| 329 |
+
"rewards/margins": 0.05159877613186836,
|
| 330 |
+
"rewards/rejected": -0.34808549284935,
|
| 331 |
+
"step": 220
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.36,
|
| 335 |
+
"learning_rate": 4.034108149278544e-06,
|
| 336 |
+
"logits/chosen": 0.24634487926959991,
|
| 337 |
+
"logits/rejected": 0.2607974112033844,
|
| 338 |
+
"logps/chosen": -1003.81884765625,
|
| 339 |
+
"logps/rejected": -1118.8782958984375,
|
| 340 |
+
"loss": 206378.925,
|
| 341 |
+
"rewards/accuracies": 0.581250011920929,
|
| 342 |
+
"rewards/chosen": -0.3140796422958374,
|
| 343 |
+
"rewards/margins": 0.06415858119726181,
|
| 344 |
+
"rewards/rejected": -0.3782382011413574,
|
| 345 |
+
"step": 230
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.38,
|
| 349 |
+
"learning_rate": 3.923630753280358e-06,
|
| 350 |
+
"logits/chosen": 0.20628812909126282,
|
| 351 |
+
"logits/rejected": 0.37754902243614197,
|
| 352 |
+
"logps/chosen": -1053.105712890625,
|
| 353 |
+
"logps/rejected": -1159.8330078125,
|
| 354 |
+
"loss": 200137.8125,
|
| 355 |
+
"rewards/accuracies": 0.637499988079071,
|
| 356 |
+
"rewards/chosen": -0.3624956011772156,
|
| 357 |
+
"rewards/margins": 0.06471217423677444,
|
| 358 |
+
"rewards/rejected": -0.4272077977657318,
|
| 359 |
+
"step": 240
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.39,
|
| 363 |
+
"learning_rate": 3.80887498010715e-06,
|
| 364 |
+
"logits/chosen": 0.29324811697006226,
|
| 365 |
+
"logits/rejected": 0.2819308936595917,
|
| 366 |
+
"logps/chosen": -1104.125244140625,
|
| 367 |
+
"logps/rejected": -1199.90185546875,
|
| 368 |
+
"loss": 227303.975,
|
| 369 |
+
"rewards/accuracies": 0.59375,
|
| 370 |
+
"rewards/chosen": -0.37163108587265015,
|
| 371 |
+
"rewards/margins": 0.04434230178594589,
|
| 372 |
+
"rewards/rejected": -0.41597336530685425,
|
| 373 |
+
"step": 250
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.41,
|
| 377 |
+
"learning_rate": 3.690185700421145e-06,
|
| 378 |
+
"logits/chosen": 0.2746821343898773,
|
| 379 |
+
"logits/rejected": 0.33129218220710754,
|
| 380 |
+
"logps/chosen": -1043.5469970703125,
|
| 381 |
+
"logps/rejected": -1157.398193359375,
|
| 382 |
+
"loss": 211889.525,
|
| 383 |
+
"rewards/accuracies": 0.6187499761581421,
|
| 384 |
+
"rewards/chosen": -0.3348512649536133,
|
| 385 |
+
"rewards/margins": 0.06582818925380707,
|
| 386 |
+
"rewards/rejected": -0.40067943930625916,
|
| 387 |
+
"step": 260
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.42,
|
| 391 |
+
"learning_rate": 3.567919606085004e-06,
|
| 392 |
+
"logits/chosen": 0.20646102726459503,
|
| 393 |
+
"logits/rejected": 0.3259996473789215,
|
| 394 |
+
"logps/chosen": -1120.896240234375,
|
| 395 |
+
"logps/rejected": -1177.1099853515625,
|
| 396 |
+
"loss": 217746.775,
|
| 397 |
+
"rewards/accuracies": 0.5562499761581421,
|
| 398 |
+
"rewards/chosen": -0.37923678755760193,
|
| 399 |
+
"rewards/margins": 0.05121999233961105,
|
| 400 |
+
"rewards/rejected": -0.4304567873477936,
|
| 401 |
+
"step": 270
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.44,
|
| 405 |
+
"learning_rate": 3.442444138210883e-06,
|
| 406 |
+
"logits/chosen": 0.23033392429351807,
|
| 407 |
+
"logits/rejected": 0.3274117708206177,
|
| 408 |
+
"logps/chosen": -1110.678466796875,
|
| 409 |
+
"logps/rejected": -1200.6253662109375,
|
| 410 |
+
"loss": 223675.5,
|
| 411 |
+
"rewards/accuracies": 0.581250011920929,
|
| 412 |
+
"rewards/chosen": -0.3978205919265747,
|
| 413 |
+
"rewards/margins": 0.051620304584503174,
|
| 414 |
+
"rewards/rejected": -0.4494408667087555,
|
| 415 |
+
"step": 280
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.46,
|
| 419 |
+
"learning_rate": 3.314136382905234e-06,
|
| 420 |
+
"logits/chosen": 0.16156849265098572,
|
| 421 |
+
"logits/rejected": 0.30493414402008057,
|
| 422 |
+
"logps/chosen": -1073.877685546875,
|
| 423 |
+
"logps/rejected": -1188.1846923828125,
|
| 424 |
+
"loss": 216146.5,
|
| 425 |
+
"rewards/accuracies": 0.606249988079071,
|
| 426 |
+
"rewards/chosen": -0.3979111611843109,
|
| 427 |
+
"rewards/margins": 0.06408871710300446,
|
| 428 |
+
"rewards/rejected": -0.46199989318847656,
|
| 429 |
+
"step": 290
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.47,
|
| 433 |
+
"learning_rate": 3.1833819380279028e-06,
|
| 434 |
+
"logits/chosen": 0.25436246395111084,
|
| 435 |
+
"logits/rejected": 0.35649871826171875,
|
| 436 |
+
"logps/chosen": -970.6337890625,
|
| 437 |
+
"logps/rejected": -1103.1796875,
|
| 438 |
+
"loss": 211613.25,
|
| 439 |
+
"rewards/accuracies": 0.5562499761581421,
|
| 440 |
+
"rewards/chosen": -0.3316889703273773,
|
| 441 |
+
"rewards/margins": 0.06474041938781738,
|
| 442 |
+
"rewards/rejected": -0.3964293897151947,
|
| 443 |
+
"step": 300
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.49,
|
| 447 |
+
"learning_rate": 3.050573754371228e-06,
|
| 448 |
+
"logits/chosen": 0.19806575775146484,
|
| 449 |
+
"logits/rejected": 0.30608507990837097,
|
| 450 |
+
"logps/chosen": -1002.451171875,
|
| 451 |
+
"logps/rejected": -1153.102783203125,
|
| 452 |
+
"loss": 199471.9375,
|
| 453 |
+
"rewards/accuracies": 0.625,
|
| 454 |
+
"rewards/chosen": -0.3274732828140259,
|
| 455 |
+
"rewards/margins": 0.08164636790752411,
|
| 456 |
+
"rewards/rejected": -0.40911969542503357,
|
| 457 |
+
"step": 310
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.5,
|
| 461 |
+
"learning_rate": 2.916110954741667e-06,
|
| 462 |
+
"logits/chosen": 0.23385831713676453,
|
| 463 |
+
"logits/rejected": 0.3278101682662964,
|
| 464 |
+
"logps/chosen": -1135.4813232421875,
|
| 465 |
+
"logps/rejected": -1250.9149169921875,
|
| 466 |
+
"loss": 214479.075,
|
| 467 |
+
"rewards/accuracies": 0.637499988079071,
|
| 468 |
+
"rewards/chosen": -0.3785611689090729,
|
| 469 |
+
"rewards/margins": 0.06773276627063751,
|
| 470 |
+
"rewards/rejected": -0.44629397988319397,
|
| 471 |
+
"step": 320
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.52,
|
| 475 |
+
"learning_rate": 2.7803976344929497e-06,
|
| 476 |
+
"logits/chosen": 0.2279421091079712,
|
| 477 |
+
"logits/rejected": 0.2835574150085449,
|
| 478 |
+
"logps/chosen": -1054.18359375,
|
| 479 |
+
"logps/rejected": -1125.3388671875,
|
| 480 |
+
"loss": 222046.5,
|
| 481 |
+
"rewards/accuracies": 0.625,
|
| 482 |
+
"rewards/chosen": -0.3506453037261963,
|
| 483 |
+
"rewards/margins": 0.05092238262295723,
|
| 484 |
+
"rewards/rejected": -0.4015676975250244,
|
| 485 |
+
"step": 330
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.53,
|
| 489 |
+
"learning_rate": 2.6438416471154277e-06,
|
| 490 |
+
"logits/chosen": 0.2912670373916626,
|
| 491 |
+
"logits/rejected": 0.41436976194381714,
|
| 492 |
+
"logps/chosen": -993.8232421875,
|
| 493 |
+
"logps/rejected": -1134.54296875,
|
| 494 |
+
"loss": 197803.8,
|
| 495 |
+
"rewards/accuracies": 0.6312500238418579,
|
| 496 |
+
"rewards/chosen": -0.34450316429138184,
|
| 497 |
+
"rewards/margins": 0.09640122950077057,
|
| 498 |
+
"rewards/rejected": -0.4409043788909912,
|
| 499 |
+
"step": 340
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.55,
|
| 503 |
+
"learning_rate": 2.5068533785312673e-06,
|
| 504 |
+
"logits/chosen": 0.259615033864975,
|
| 505 |
+
"logits/rejected": 0.3351518213748932,
|
| 506 |
+
"logps/chosen": -1080.523681640625,
|
| 507 |
+
"logps/rejected": -1217.09521484375,
|
| 508 |
+
"loss": 224075.1,
|
| 509 |
+
"rewards/accuracies": 0.5874999761581421,
|
| 510 |
+
"rewards/chosen": -0.3753746747970581,
|
| 511 |
+
"rewards/margins": 0.07089592516422272,
|
| 512 |
+
"rewards/rejected": -0.446270614862442,
|
| 513 |
+
"step": 350
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.57,
|
| 517 |
+
"learning_rate": 2.369844513779026e-06,
|
| 518 |
+
"logits/chosen": 0.16386622190475464,
|
| 519 |
+
"logits/rejected": 0.20941393077373505,
|
| 520 |
+
"logps/chosen": -1049.8115234375,
|
| 521 |
+
"logps/rejected": -1173.7218017578125,
|
| 522 |
+
"loss": 214063.75,
|
| 523 |
+
"rewards/accuracies": 0.65625,
|
| 524 |
+
"rewards/chosen": -0.30884891748428345,
|
| 525 |
+
"rewards/margins": 0.07134245336055756,
|
| 526 |
+
"rewards/rejected": -0.3801913857460022,
|
| 527 |
+
"step": 360
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.58,
|
| 531 |
+
"learning_rate": 2.2332267997940514e-06,
|
| 532 |
+
"logits/chosen": 0.1576313078403473,
|
| 533 |
+
"logits/rejected": 0.2314819097518921,
|
| 534 |
+
"logps/chosen": -989.9000244140625,
|
| 535 |
+
"logps/rejected": -1112.245849609375,
|
| 536 |
+
"loss": 212630.15,
|
| 537 |
+
"rewards/accuracies": 0.625,
|
| 538 |
+
"rewards/chosen": -0.28885313868522644,
|
| 539 |
+
"rewards/margins": 0.05920270085334778,
|
| 540 |
+
"rewards/rejected": -0.3480558395385742,
|
| 541 |
+
"step": 370
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.6,
|
| 545 |
+
"learning_rate": 2.097410808002869e-06,
|
| 546 |
+
"logits/chosen": 0.2326604425907135,
|
| 547 |
+
"logits/rejected": 0.3002316653728485,
|
| 548 |
+
"logps/chosen": -1086.225341796875,
|
| 549 |
+
"logps/rejected": -1175.258056640625,
|
| 550 |
+
"loss": 216490.55,
|
| 551 |
+
"rewards/accuracies": 0.6187499761581421,
|
| 552 |
+
"rewards/chosen": -0.3227333724498749,
|
| 553 |
+
"rewards/margins": 0.06897234916687012,
|
| 554 |
+
"rewards/rejected": -0.3917057514190674,
|
| 555 |
+
"step": 380
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.61,
|
| 559 |
+
"learning_rate": 1.962804700450265e-06,
|
| 560 |
+
"logits/chosen": 0.2371029108762741,
|
| 561 |
+
"logits/rejected": 0.3002297580242157,
|
| 562 |
+
"logps/chosen": -1030.2000732421875,
|
| 563 |
+
"logps/rejected": -1175.873046875,
|
| 564 |
+
"loss": 214079.6,
|
| 565 |
+
"rewards/accuracies": 0.6187499761581421,
|
| 566 |
+
"rewards/chosen": -0.3257220983505249,
|
| 567 |
+
"rewards/margins": 0.07455430179834366,
|
| 568 |
+
"rewards/rejected": -0.40027642250061035,
|
| 569 |
+
"step": 390
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.63,
|
| 573 |
+
"learning_rate": 1.8298130031671974e-06,
|
| 574 |
+
"logits/chosen": 0.25191420316696167,
|
| 575 |
+
"logits/rejected": 0.29652512073516846,
|
| 576 |
+
"logps/chosen": -1057.87646484375,
|
| 577 |
+
"logps/rejected": -1160.5743408203125,
|
| 578 |
+
"loss": 227584.0,
|
| 579 |
+
"rewards/accuracies": 0.5375000238418579,
|
| 580 |
+
"rewards/chosen": -0.3368803858757019,
|
| 581 |
+
"rewards/margins": 0.04710642248392105,
|
| 582 |
+
"rewards/rejected": -0.38398683071136475,
|
| 583 |
+
"step": 400
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.64,
|
| 587 |
+
"learning_rate": 1.6988353904658495e-06,
|
| 588 |
+
"logits/chosen": 0.25754934549331665,
|
| 589 |
+
"logits/rejected": 0.1926787793636322,
|
| 590 |
+
"logps/chosen": -1006.3465576171875,
|
| 591 |
+
"logps/rejected": -1107.968994140625,
|
| 592 |
+
"loss": 213971.8,
|
| 593 |
+
"rewards/accuracies": 0.59375,
|
| 594 |
+
"rewards/chosen": -0.2910354435443878,
|
| 595 |
+
"rewards/margins": 0.06115920469164848,
|
| 596 |
+
"rewards/rejected": -0.3521946668624878,
|
| 597 |
+
"step": 410
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.66,
|
| 601 |
+
"learning_rate": 1.5702654838153641e-06,
|
| 602 |
+
"logits/chosen": 0.32284024357795715,
|
| 603 |
+
"logits/rejected": 0.3785749673843384,
|
| 604 |
+
"logps/chosen": -1025.57373046875,
|
| 605 |
+
"logps/rejected": -1150.352783203125,
|
| 606 |
+
"loss": 213707.075,
|
| 607 |
+
"rewards/accuracies": 0.65625,
|
| 608 |
+
"rewards/chosen": -0.32002902030944824,
|
| 609 |
+
"rewards/margins": 0.0713513195514679,
|
| 610 |
+
"rewards/rejected": -0.3913803696632385,
|
| 611 |
+
"step": 420
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.68,
|
| 615 |
+
"learning_rate": 1.4444896689079142e-06,
|
| 616 |
+
"logits/chosen": 0.24807333946228027,
|
| 617 |
+
"logits/rejected": 0.26119035482406616,
|
| 618 |
+
"logps/chosen": -981.4894409179688,
|
| 619 |
+
"logps/rejected": -1027.531494140625,
|
| 620 |
+
"loss": 228694.3,
|
| 621 |
+
"rewards/accuracies": 0.512499988079071,
|
| 622 |
+
"rewards/chosen": -0.31893014907836914,
|
| 623 |
+
"rewards/margins": 0.03217142075300217,
|
| 624 |
+
"rewards/rejected": -0.3511015772819519,
|
| 625 |
+
"step": 430
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.69,
|
| 629 |
+
"learning_rate": 1.3218859344701634e-06,
|
| 630 |
+
"logits/chosen": 0.24505586922168732,
|
| 631 |
+
"logits/rejected": 0.23741266131401062,
|
| 632 |
+
"logps/chosen": -976.6732177734375,
|
| 633 |
+
"logps/rejected": -1118.38134765625,
|
| 634 |
+
"loss": 197599.0,
|
| 635 |
+
"rewards/accuracies": 0.637499988079071,
|
| 636 |
+
"rewards/chosen": -0.2975761890411377,
|
| 637 |
+
"rewards/margins": 0.08645204454660416,
|
| 638 |
+
"rewards/rejected": -0.38402819633483887,
|
| 639 |
+
"step": 440
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.71,
|
| 643 |
+
"learning_rate": 1.2028227363097583e-06,
|
| 644 |
+
"logits/chosen": 0.2540598213672638,
|
| 645 |
+
"logits/rejected": 0.33230239152908325,
|
| 646 |
+
"logps/chosen": -1014.8942260742188,
|
| 647 |
+
"logps/rejected": -1128.714599609375,
|
| 648 |
+
"loss": 204484.4,
|
| 649 |
+
"rewards/accuracies": 0.6499999761581421,
|
| 650 |
+
"rewards/chosen": -0.31904861330986023,
|
| 651 |
+
"rewards/margins": 0.0724792331457138,
|
| 652 |
+
"rewards/rejected": -0.39152783155441284,
|
| 653 |
+
"step": 450
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.72,
|
| 657 |
+
"learning_rate": 1.0876578900107053e-06,
|
| 658 |
+
"logits/chosen": 0.2582911550998688,
|
| 659 |
+
"logits/rejected": 0.27663809061050415,
|
| 660 |
+
"logps/chosen": -1002.5812377929688,
|
| 661 |
+
"logps/rejected": -1135.8900146484375,
|
| 662 |
+
"loss": 212971.9,
|
| 663 |
+
"rewards/accuracies": 0.625,
|
| 664 |
+
"rewards/chosen": -0.33705443143844604,
|
| 665 |
+
"rewards/margins": 0.0685950368642807,
|
| 666 |
+
"rewards/rejected": -0.40564948320388794,
|
| 667 |
+
"step": 460
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.74,
|
| 671 |
+
"learning_rate": 9.767374956053584e-07,
|
| 672 |
+
"logits/chosen": 0.21633680164813995,
|
| 673 |
+
"logits/rejected": 0.2309296429157257,
|
| 674 |
+
"logps/chosen": -1026.5072021484375,
|
| 675 |
+
"logps/rejected": -1110.014892578125,
|
| 676 |
+
"loss": 204311.2875,
|
| 677 |
+
"rewards/accuracies": 0.6000000238418579,
|
| 678 |
+
"rewards/chosen": -0.33438438177108765,
|
| 679 |
+
"rewards/margins": 0.06735749542713165,
|
| 680 |
+
"rewards/rejected": -0.4017418920993805,
|
| 681 |
+
"step": 470
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.75,
|
| 685 |
+
"learning_rate": 8.703948974546592e-07,
|
| 686 |
+
"logits/chosen": 0.17137208580970764,
|
| 687 |
+
"logits/rejected": 0.29713553190231323,
|
| 688 |
+
"logps/chosen": -1010.2471923828125,
|
| 689 |
+
"logps/rejected": -1127.9761962890625,
|
| 690 |
+
"loss": 210587.325,
|
| 691 |
+
"rewards/accuracies": 0.5687500238418579,
|
| 692 |
+
"rewards/chosen": -0.3343172073364258,
|
| 693 |
+
"rewards/margins": 0.048689451068639755,
|
| 694 |
+
"rewards/rejected": -0.3830066919326782,
|
| 695 |
+
"step": 480
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.77,
|
| 699 |
+
"learning_rate": 7.689496824624526e-07,
|
| 700 |
+
"logits/chosen": 0.24797169864177704,
|
| 701 |
+
"logits/rejected": 0.3339524269104004,
|
| 702 |
+
"logps/chosen": -1013.1183471679688,
|
| 703 |
+
"logps/rejected": -1139.7716064453125,
|
| 704 |
+
"loss": 205908.4125,
|
| 705 |
+
"rewards/accuracies": 0.6000000238418579,
|
| 706 |
+
"rewards/chosen": -0.31524255871772766,
|
| 707 |
+
"rewards/margins": 0.06370928138494492,
|
| 708 |
+
"rewards/rejected": -0.3789518177509308,
|
| 709 |
+
"step": 490
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.78,
|
| 713 |
+
"learning_rate": 6.7270671963451e-07,
|
| 714 |
+
"logits/chosen": 0.2785577178001404,
|
| 715 |
+
"logits/rejected": 0.34557706117630005,
|
| 716 |
+
"logps/chosen": -1004.1671752929688,
|
| 717 |
+
"logps/rejected": -1095.4044189453125,
|
| 718 |
+
"loss": 222017.075,
|
| 719 |
+
"rewards/accuracies": 0.5687500238418579,
|
| 720 |
+
"rewards/chosen": -0.324979692697525,
|
| 721 |
+
"rewards/margins": 0.05109068751335144,
|
| 722 |
+
"rewards/rejected": -0.3760703504085541,
|
| 723 |
+
"step": 500
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.8,
|
| 727 |
+
"learning_rate": 5.819552438686238e-07,
|
| 728 |
+
"logits/chosen": 0.22823992371559143,
|
| 729 |
+
"logits/rejected": 0.3369165360927582,
|
| 730 |
+
"logps/chosen": -1076.311767578125,
|
| 731 |
+
"logps/rejected": -1174.0753173828125,
|
| 732 |
+
"loss": 216482.85,
|
| 733 |
+
"rewards/accuracies": 0.606249988079071,
|
| 734 |
+
"rewards/chosen": -0.3366367220878601,
|
| 735 |
+
"rewards/margins": 0.06757824122905731,
|
| 736 |
+
"rewards/rejected": -0.404215008020401,
|
| 737 |
+
"step": 510
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.82,
|
| 741 |
+
"learning_rate": 4.969679867292276e-07,
|
| 742 |
+
"logits/chosen": 0.2003372609615326,
|
| 743 |
+
"logits/rejected": 0.33764129877090454,
|
| 744 |
+
"logps/chosen": -1053.8939208984375,
|
| 745 |
+
"logps/rejected": -1123.6129150390625,
|
| 746 |
+
"loss": 221411.85,
|
| 747 |
+
"rewards/accuracies": 0.5625,
|
| 748 |
+
"rewards/chosen": -0.3437155783176422,
|
| 749 |
+
"rewards/margins": 0.03899475932121277,
|
| 750 |
+
"rewards/rejected": -0.3827103078365326,
|
| 751 |
+
"step": 520
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.83,
|
| 755 |
+
"learning_rate": 4.1800035681877765e-07,
|
| 756 |
+
"logits/chosen": 0.3177185654640198,
|
| 757 |
+
"logits/rejected": 0.20578141510486603,
|
| 758 |
+
"logps/chosen": -1085.5306396484375,
|
| 759 |
+
"logps/rejected": -1226.7509765625,
|
| 760 |
+
"loss": 199054.875,
|
| 761 |
+
"rewards/accuracies": 0.643750011920929,
|
| 762 |
+
"rewards/chosen": -0.34231036901474,
|
| 763 |
+
"rewards/margins": 0.07949638366699219,
|
| 764 |
+
"rewards/rejected": -0.4218067526817322,
|
| 765 |
+
"step": 530
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.85,
|
| 769 |
+
"learning_rate": 3.4528967220911287e-07,
|
| 770 |
+
"logits/chosen": 0.26802507042884827,
|
| 771 |
+
"logits/rejected": 0.2959344983100891,
|
| 772 |
+
"logps/chosen": -1103.7427978515625,
|
| 773 |
+
"logps/rejected": -1205.734619140625,
|
| 774 |
+
"loss": 206713.1125,
|
| 775 |
+
"rewards/accuracies": 0.612500011920929,
|
| 776 |
+
"rewards/chosen": -0.34727898240089417,
|
| 777 |
+
"rewards/margins": 0.0809955969452858,
|
| 778 |
+
"rewards/rejected": -0.42827457189559937,
|
| 779 |
+
"step": 540
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.86,
|
| 783 |
+
"learning_rate": 2.7905444723949765e-07,
|
| 784 |
+
"logits/chosen": 0.24068915843963623,
|
| 785 |
+
"logits/rejected": 0.36839014291763306,
|
| 786 |
+
"logps/chosen": -1065.5458984375,
|
| 787 |
+
"logps/rejected": -1131.9764404296875,
|
| 788 |
+
"loss": 218209.65,
|
| 789 |
+
"rewards/accuracies": 0.581250011920929,
|
| 790 |
+
"rewards/chosen": -0.3374769687652588,
|
| 791 |
+
"rewards/margins": 0.04803069308400154,
|
| 792 |
+
"rewards/rejected": -0.3855076730251312,
|
| 793 |
+
"step": 550
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.88,
|
| 797 |
+
"learning_rate": 2.1949373582475065e-07,
|
| 798 |
+
"logits/chosen": 0.30096447467803955,
|
| 799 |
+
"logits/rejected": 0.34244808554649353,
|
| 800 |
+
"logps/chosen": -962.107421875,
|
| 801 |
+
"logps/rejected": -1130.420166015625,
|
| 802 |
+
"loss": 203465.5125,
|
| 803 |
+
"rewards/accuracies": 0.65625,
|
| 804 |
+
"rewards/chosen": -0.3097023069858551,
|
| 805 |
+
"rewards/margins": 0.073573537170887,
|
| 806 |
+
"rewards/rejected": -0.3832758069038391,
|
| 807 |
+
"step": 560
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 0.89,
|
| 811 |
+
"learning_rate": 1.667865332469379e-07,
|
| 812 |
+
"logits/chosen": 0.22145691514015198,
|
| 813 |
+
"logits/rejected": 0.3323192000389099,
|
| 814 |
+
"logps/chosen": -997.6228637695312,
|
| 815 |
+
"logps/rejected": -1113.1171875,
|
| 816 |
+
"loss": 211145.9,
|
| 817 |
+
"rewards/accuracies": 0.6000000238418579,
|
| 818 |
+
"rewards/chosen": -0.3235357105731964,
|
| 819 |
+
"rewards/margins": 0.06192871183156967,
|
| 820 |
+
"rewards/rejected": -0.3854644000530243,
|
| 821 |
+
"step": 570
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 0.91,
|
| 825 |
+
"learning_rate": 1.2109123822844653e-07,
|
| 826 |
+
"logits/chosen": 0.2121209353208542,
|
| 827 |
+
"logits/rejected": 0.32750505208969116,
|
| 828 |
+
"logps/chosen": -1062.212158203125,
|
| 829 |
+
"logps/rejected": -1173.4722900390625,
|
| 830 |
+
"loss": 216943.15,
|
| 831 |
+
"rewards/accuracies": 0.6187499761581421,
|
| 832 |
+
"rewards/chosen": -0.3464076817035675,
|
| 833 |
+
"rewards/margins": 0.06313388049602509,
|
| 834 |
+
"rewards/rejected": -0.409541517496109,
|
| 835 |
+
"step": 580
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 0.93,
|
| 839 |
+
"learning_rate": 8.254517690300946e-08,
|
| 840 |
+
"logits/chosen": 0.2600761651992798,
|
| 841 |
+
"logits/rejected": 0.277840793132782,
|
| 842 |
+
"logps/chosen": -1001.9027099609375,
|
| 843 |
+
"logps/rejected": -1103.707763671875,
|
| 844 |
+
"loss": 212477.1,
|
| 845 |
+
"rewards/accuracies": 0.612500011920929,
|
| 846 |
+
"rewards/chosen": -0.321969211101532,
|
| 847 |
+
"rewards/margins": 0.06838768720626831,
|
| 848 |
+
"rewards/rejected": -0.3903568983078003,
|
| 849 |
+
"step": 590
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.94,
|
| 853 |
+
"learning_rate": 5.126419011529993e-08,
|
| 854 |
+
"logits/chosen": 0.2812707722187042,
|
| 855 |
+
"logits/rejected": 0.3223172128200531,
|
| 856 |
+
"logps/chosen": -1003.4280395507812,
|
| 857 |
+
"logps/rejected": -1109.736083984375,
|
| 858 |
+
"loss": 207337.8,
|
| 859 |
+
"rewards/accuracies": 0.6312500238418579,
|
| 860 |
+
"rewards/chosen": -0.3305527865886688,
|
| 861 |
+
"rewards/margins": 0.05715988203883171,
|
| 862 |
+
"rewards/rejected": -0.38771265745162964,
|
| 863 |
+
"step": 600
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.96,
|
| 867 |
+
"learning_rate": 2.734228528934679e-08,
|
| 868 |
+
"logits/chosen": 0.25758108496665955,
|
| 869 |
+
"logits/rejected": 0.26223865151405334,
|
| 870 |
+
"logps/chosen": -1068.127197265625,
|
| 871 |
+
"logps/rejected": -1201.6923828125,
|
| 872 |
+
"loss": 207598.25,
|
| 873 |
+
"rewards/accuracies": 0.625,
|
| 874 |
+
"rewards/chosen": -0.34323567152023315,
|
| 875 |
+
"rewards/margins": 0.07567054033279419,
|
| 876 |
+
"rewards/rejected": -0.41890621185302734,
|
| 877 |
+
"step": 610
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 0.97,
|
| 881 |
+
"learning_rate": 1.0851353912008644e-08,
|
| 882 |
+
"logits/chosen": 0.24180451035499573,
|
| 883 |
+
"logits/rejected": 0.30353251099586487,
|
| 884 |
+
"logps/chosen": -1058.5758056640625,
|
| 885 |
+
"logps/rejected": -1153.9122314453125,
|
| 886 |
+
"loss": 209104.875,
|
| 887 |
+
"rewards/accuracies": 0.606249988079071,
|
| 888 |
+
"rewards/chosen": -0.3343828320503235,
|
| 889 |
+
"rewards/margins": 0.06873960793018341,
|
| 890 |
+
"rewards/rejected": -0.4031224846839905,
|
| 891 |
+
"step": 620
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 0.99,
|
| 895 |
+
"learning_rate": 1.8409554805329243e-09,
|
| 896 |
+
"logits/chosen": 0.22301486134529114,
|
| 897 |
+
"logits/rejected": 0.2677055895328522,
|
| 898 |
+
"logps/chosen": -1038.4554443359375,
|
| 899 |
+
"logps/rejected": -1139.900634765625,
|
| 900 |
+
"loss": 215125.55,
|
| 901 |
+
"rewards/accuracies": 0.6312500238418579,
|
| 902 |
+
"rewards/chosen": -0.3383429944515228,
|
| 903 |
+
"rewards/margins": 0.06525392830371857,
|
| 904 |
+
"rewards/rejected": -0.403596967458725,
|
| 905 |
+
"step": 630
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 1.0,
|
| 909 |
+
"step": 637,
|
| 910 |
+
"total_flos": 0.0,
|
| 911 |
+
"train_loss": 219463.76923076922,
|
| 912 |
+
"train_runtime": 8069.1141,
|
| 913 |
+
"train_samples_per_second": 2.525,
|
| 914 |
+
"train_steps_per_second": 0.079
|
| 915 |
+
}
|
| 916 |
+
],
|
| 917 |
+
"logging_steps": 10,
|
| 918 |
+
"max_steps": 637,
|
| 919 |
+
"num_input_tokens_seen": 0,
|
| 920 |
+
"num_train_epochs": 1,
|
| 921 |
+
"save_steps": 100,
|
| 922 |
+
"total_flos": 0.0,
|
| 923 |
+
"train_batch_size": 4,
|
| 924 |
+
"trial_name": null,
|
| 925 |
+
"trial_params": null
|
| 926 |
+
}
|