Hugging Face
Models
Datasets
Spaces
Posts
Docs
Solutions
Pricing
Log In
Sign Up
ilperrozzi
/
zephyr-7b-dpo-qlora
like
0
PEFT
TensorBoard
Safetensors
HuggingFaceH4/ultrafeedback_binarized
mistral
alignment-handbook
Generated from Trainer
trl
dpo
4-bit precision
bitsandbytes
License:
apache-2.0
Model card
Files
Files and versions
Metrics
Training metrics
Community
Train
Use this model
main
zephyr-7b-dpo-qlora
1 contributor
History:
3 commits
ilperrozzi
End of training
fb6f820
verified
6 months ago
runs
Model save
6 months ago
.gitattributes
1.52 kB
initial commit
6 months ago
README.md
1.74 kB
End of training
6 months ago
adapter_config.json
657 Bytes
Model save
6 months ago
adapter_model.safetensors
671 MB
LFS
Model save
6 months ago
all_results.json
748 Bytes
Model save
6 months ago
config.json
1.03 kB
End of training
6 months ago
eval_results.json
575 Bytes
Model save
6 months ago
special_tokens_map.json
551 Bytes
Model save
6 months ago
tokenizer.json
1.8 MB
Model save
6 months ago
tokenizer_config.json
1.39 kB
Model save
6 months ago
train_results.json
194 Bytes
Model save
6 months ago
trainer_state.json
1.58 kB
Model save
6 months ago
training_args.bin
pickle
Detected Pickle imports (8)
"accelerate.utils.dataclasses.DistributedType"
,
"transformers.training_args.OptimizerNames"
,
"transformers.trainer_utils.HubStrategy"
,
"alignment.configs.DPOConfig"
,
"transformers.trainer_utils.IntervalStrategy"
,
"accelerate.state.PartialState"
,
"torch.device"
,
"transformers.trainer_utils.SchedulerType"
How to fix it?
4.86 kB
LFS
Model save
6 months ago