Hugging Face
Models
Datasets
Spaces
Posts
Docs
Solutions
Pricing
Log In
Sign Up
THU-KEG
/
Mistral-Crab-DPO
like
2
Follow
Knowledge Engineer Group @ Tsinghua University
18
Text Generation
Safetensors
English
mistral
alignment-handbook
Generated from Trainer
conversational
arxiv:
2410.24175
License:
apache-2.0
Model card
Files
Files and versions
Community
Train
c1412d7
Mistral-Crab-DPO
/
all_results.json
Kikkk
commit
c1412d7
6 days ago
raw
Copy download link
history
blame
Safe
217 Bytes
{
"epoch"
:
1.0
,
"total_flos"
:
0.0
,
"train_loss"
:
0.524798985838762
,
"train_runtime"
:
12257.7363
,
"train_samples"
:
59556
,
"train_samples_per_second"
:
4.859
,
"train_steps_per_second"
:
0.076
}