Hugging Face
Models
Datasets
Spaces
Posts
Docs
Solutions
Pricing
Log In
Sign Up
imelnyk
/
slm-2-dpo-full
like
0
Text Generation
Transformers
TensorBoard
Safetensors
HuggingFaceH4/ultrafeedback_binarized
stablelm_epoch
alignment-handbook
Generated from Trainer
trl
dpo
conversational
custom_code
License:
other
Model card
Files
Files and versions
Metrics
Training metrics
Community
Train
Use this model
152b38d
slm-2-dpo-full
/
train_results.json
imelnyk
Model save
152b38d
verified
9 months ago
raw
Copy download link
history
blame
Safe
195 Bytes
{
"epoch"
:
1.0
,
"train_loss"
:
46.613421885577296
,
"train_runtime"
:
4597.6924
,
"train_samples"
:
61135
,
"train_samples_per_second"
:
13.297
,
"train_steps_per_second"
:
0.166
}