Hugging Face
Models
Datasets
Spaces
Posts
Docs
Solutions
Pricing
Log In
Sign Up
bartowski
/
zephyr-7b-dpo-full-exl2
like
1
Text Generation
HuggingFaceH4/ultrafeedback_binarized
alignment-handbook
Generated from Trainer
trl
dpo
License:
apache-2.0
Model card
Files
Files and versions
Community
3_5
zephyr-7b-dpo-full-exl2
/
train_results.json
bartowski
Quant for 3.5
abaff34
verified
9 months ago
raw
Copy download link
history
blame
No virus
195 Bytes
{
"epoch"
:
1.0
,
"train_loss"
:
0.5379065808890754
,
"train_runtime"
:
5396.8094
,
"train_samples"
:
61135
,
"train_samples_per_second"
:
11.328
,
"train_steps_per_second"
:
0.089
}