Hugging Face
Models
Datasets
Spaces
Posts
Docs
Solutions
Pricing
Log In
Sign Up
dball
/
zephyr-7b-sft-qlora
like
0
PEFT
TensorBoard
Safetensors
HuggingFaceH4/ultrachat_200k
mistral
alignment-handbook
Generated from Trainer
trl
sft
4-bit precision
bitsandbytes
License:
apache-2.0
Model card
Files
Files and versions
Metrics
Training metrics
Community
2
Train
Use this model
1dda6bb
zephyr-7b-sft-qlora
/
eval_results.json
dball
Model save
1dda6bb
verified
10 months ago
raw
Copy download link
history
blame
Safe
184 Bytes
{
"epoch"
:
1.0
,
"eval_loss"
:
0.952344536781311
,
"eval_runtime"
:
8574.5952
,
"eval_samples"
:
23110
,
"eval_samples_per_second"
:
1.8
,
"eval_steps_per_second"
:
0.9
}