Hugging Face
Models
Datasets
Spaces
Posts
Docs
Solutions
Pricing
Log In
Sign Up
jikaixuan
/
zephyr-7b
like
0
PEFT
TensorBoard
Safetensors
HuggingFaceH4/ultrafeedback_binarized
mistral
alignment-handbook
trl
dpo
Generated from Trainer
4-bit precision
bitsandbytes
License:
apache-2.0
Model card
Files
Files and versions
Metrics
Training metrics
Community
Train
Use this model
cd97752
zephyr-7b
/
train_results.json
jikaixuan
Model save
6b1b603
verified
5 months ago
raw
Copy download link
history
blame
No virus
193 Bytes
{
"epoch"
:
1.0
,
"train_loss"
:
0.6357159084743924
,
"train_runtime"
:
9601.7268
,
"train_samples"
:
61135
,
"train_samples_per_second"
:
6.367
,
"train_steps_per_second"
:
0.05
}