Hugging Face
Models
Datasets
Spaces
Community
Docs
Enterprise
Pricing
Log In
Sign Up
nlee-208
/
Q3-Harm-DPO
like
0
Text Generation
Transformers
TensorBoard
Safetensors
qwen3
Generated from Trainer
trl
dpo
conversational
text-generation-inference
arxiv:
2305.18290
Model card
Files
Files and versions
Metrics
Training metrics
Community
Train
Deploy
Use this model
2e99e50
Q3-Harm-DPO
/
all_results.json
nlee-208
Model save
2e99e50
verified
11 days ago
raw
Copy download link
history
blame
216 Bytes
{
"epoch"
:
1.0
,
"total_flos"
:
0.0
,
"train_loss"
:
0.377603563789741
,
"train_runtime"
:
6327.9011
,
"train_samples"
:
25473
,
"train_samples_per_second"
:
4.026
,
"train_steps_per_second"
:
0.126
}