Hugging Face
Models
Datasets
Spaces
Posts
Docs
Solutions
Pricing
Log In
Sign Up
wyan1
/
zephyr-7b-dpo-full
like
0
Text Generation
Transformers
TensorBoard
Safetensors
HuggingFaceH4/ultrafeedback_binarized
mistral
alignment-handbook
trl
dpo
Generated from Trainer
conversational
text-generation-inference
Inference Endpoints
License:
apache-2.0
Model card
Files
Files and versions
Metrics
Training metrics
Community
Train
Deploy
Use this model
8cfd104
zephyr-7b-dpo-full
/
all_results.json
wyan1
Model save
8cfd104
verified
about 2 months ago
raw
Copy download link
history
blame
No virus
217 Bytes
{
"epoch"
:
1.0
,
"total_flos"
:
0.0
,
"train_loss"
:
0.5389037092360492
,
"train_runtime"
:
7287.4677
,
"train_samples"
:
61134
,
"train_samples_per_second"
:
8.389
,
"train_steps_per_second"
:
0.066
}