mistral-7b-detox

This model is a fine-tuned version of mistralai/Mistral-7B-v0.3 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.6476

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 5e-05
• train_batch_size: 4
• eval_batch_size: 4
• seed: 42
• distributed_type: multi-GPU
• num_devices: 4
• gradient_accumulation_steps: 3
• total_train_batch_size: 48
• total_eval_batch_size: 16
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: cosine
• lr_scheduler_warmup_steps: 30
• training_steps: 700

Training results

Training Loss	Epoch	Step	Validation Loss
0.9783	0.0714	50	0.8158
0.3603	0.1429	100	0.5411
0.3077	0.2143	150	0.4969
0.2214	0.2857	200	0.5404
0.1777	0.3571	250	0.5685
0.1138	0.4286	300	0.5604
0.1029	0.5	350	0.6015
0.0766	0.5714	400	0.5587
0.0631	0.6429	450	0.5737
0.0362	0.7143	500	0.5656
0.0374	0.7857	550	0.6343
0.033	0.8571	600	0.6397
0.0338	0.9286	650	0.6391
0.0323	1.0	700	0.6476

Framework versions

• Transformers 4.44.0.dev0
• Pytorch 2.1.0+cu121
• Datasets 2.20.0
• Tokenizers 0.19.1