Hugging Face
Models
Datasets
Spaces
Posts
Docs
Solutions
Pricing
Log In
Sign Up
HuggingFaceH4
/
zephyr-orpo-141b-A35b-v0.1
like
260
Follow
Hugging Face H4
643
Text Generation
Transformers
TensorBoard
Safetensors
argilla/distilabel-capybara-dpo-7k-binarized
mixtral
trl
orpo
Generated from Trainer
conversational
text-generation-inference
Inference Endpoints
arxiv:
2403.07691
arxiv:
2311.07911
License:
apache-2.0
Model card
Files
Files and versions
Metrics
Training metrics
Community
15
Train
Deploy
Use this model
b08d449
zephyr-orpo-141b-A35b-v0.1
/
train_results.json
lewtun
HF staff
Model save
b08d449
verified
7 months ago
raw
Copy download link
history
blame
192 Bytes
{
"epoch"
:
3.0
,
"train_loss"
:
1.0000138937419827
,
"train_runtime"
:
4770.0198
,
"train_samples"
:
6932
,
"train_samples_per_second"
:
4.36
,
"train_steps_per_second"
:
0.136
}