Hugging Face
Models
Datasets
Spaces
Posts
Docs
Solutions
Pricing
Log In
Sign Up
DUAL-GPO
/
zephyr-7b-gpo-new-v1-i0
like
0
PEFT
TensorBoard
Safetensors
HuggingFaceH4/ultrafeedback_binarized
mistral
alignment-handbook
Generated from Trainer
trl
dpo
Model card
Files
Files and versions
Metrics
Training metrics
Community
Train
Use this model
fda141e
zephyr-7b-gpo-new-v1-i0
/
all_results.json
lole25
Model save
879a39a
verified
5 months ago
raw
Copy download link
history
blame
Safe
197 Bytes
{
"epoch"
:
1.0
,
"train_loss"
:
0.0048277108391146795
,
"train_runtime"
:
13417.0015
,
"train_samples"
:
61135
,
"train_samples_per_second"
:
4.557
,
"train_steps_per_second"
:
0.57
}