Hugging Face
Models
Datasets
Spaces
Posts
Docs
Enterprise
Pricing
Log In
Sign Up
lole25
/
phi-2-gpo-test-iter-2
like
0
PEFT
Safetensors
HuggingFaceH4/ultrafeedback_binarized
phi
alignment-handbook
Generated from Trainer
trl
dpo
custom_code
License:
mit
Model card
Files
Files and versions
Community
Use this model
3a1c508
phi-2-gpo-test-iter-2
/
all_results.json
lole25
Model save
3a1c508
verified
11 months ago
raw
Copy download link
history
blame
Safe
194 Bytes
{
"epoch"
:
2.0
,
"train_loss"
:
0.00022506012464873493
,
"train_runtime"
:
14.2725
,
"train_samples"
:
30567
,
"train_samples_per_second"
:
0.28
,
"train_steps_per_second"
:
0.14
}