Hugging Face
Models
Datasets
Spaces
Posts
Docs
Solutions
Pricing
Log In
Sign Up
DUAL-GPO
/
phi-2-irepo-chatml-i0
like
0
PEFT
TensorBoard
Safetensors
HuggingFaceH4/ultrafeedback_binarized
phi
alignment-handbook
Generated from Trainer
trl
dpo
custom_code
License:
apache-2.0
Model card
Files
Files and versions
Metrics
Training metrics
Community
Use this model
7825290
phi-2-irepo-chatml-i0
/
train_results.json
lole25
Model save
7825290
verified
3 months ago
raw
Copy download link
history
blame
No virus
196 Bytes
{
"epoch"
:
1.0
,
"train_loss"
:
0.04083743233644686
,
"train_runtime"
:
25203.2602
,
"train_samples"
:
61135
,
"train_samples_per_second"
:
2.426
,
"train_steps_per_second"
:
0.101
}