Hugging Face
Models
Datasets
Spaces
Posts
Docs
Solutions
Pricing
Log In
Sign Up
just1nseo
/
zephyr-7b-dpop-uf6k-qlora-5e-7-epoch3
like
0
PEFT
TensorBoard
Safetensors
generation/UF6konly
mistral
alignment-handbook
trl
dpo
Generated from Trainer
4-bit precision
bitsandbytes
License:
apache-2.0
Model card
Files
Files and versions
Metrics
Training metrics
Community
Train
Use this model
376e9eb
zephyr-7b-dpop-uf6k-qlora-5e-7-epoch3
/
all_results.json
just1nseo
Model save
376e9eb
verified
2 months ago
raw
Copy download link
history
blame
No virus
191 Bytes
{
"epoch"
:
3.0
,
"train_loss"
:
0.684076373458754
,
"train_runtime"
:
8695.656
,
"train_samples"
:
5263
,
"train_samples_per_second"
:
1.816
,
"train_steps_per_second"
:
0.114
}