Hugging Face
Models
Datasets
Spaces
Posts
Docs
Enterprise
Pricing
Log In
Sign Up
klcsp
/
gemma7b-gpt4o_100k_closedqa-k
like
0
Follow
klcsp
1
PEFT
TensorBoard
Safetensors
llama-duo/synth_closed_qa_dataset_dedup
gemma
alignment-handbook
trl
sft
Generated from Trainer
8-bit precision
bitsandbytes
License:
gemma
Model card
Files
Files and versions
Metrics
Training metrics
Community
Train
Use this model
9d8a7d5
gemma7b-gpt4o_100k_closedqa-k
/
eval_results.json
chansung
End of training
9d8a7d5
verified
3 months ago
raw
Copy download link
history
blame
Safe
198 Bytes
{
"epoch"
:
0.9990375360923965
,
"eval_loss"
:
2.2841131687164307
,
"eval_runtime"
:
1.3399
,
"eval_samples"
:
15
,
"eval_samples_per_second"
:
4.478
,
"eval_steps_per_second"
:
0.746
}