90bff7b0-4e85-410c-bc74-7b3ecdf81b9c

This model is a fine-tuned version of NousResearch/Yarn-Solar-10b-32k on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.1753

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: 0.000206
• train_batch_size: 4
• eval_batch_size: 4
• seed: 60
• gradient_accumulation_steps: 2
• total_train_batch_size: 8
• optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: cosine
• lr_scheduler_warmup_steps: 50
• training_steps: 500

Training results

Training Loss	Epoch	Step	Validation Loss
No log	0.0002	1	1.1962
0.9767	0.0095	50	0.7277
0.8686	0.0191	100	0.6216
0.638	0.0286	150	0.7299
0.7123	0.0381	200	0.4410
0.6757	0.0476	250	0.3962
0.5306	0.0572	300	0.3492
0.4661	0.0667	350	0.2493
0.3753	0.0762	400	0.2015
0.3964	0.0858	450	0.1792
0.3404	0.0953	500	0.1753

Framework versions

• PEFT 0.13.2
• Transformers 4.46.0
• Pytorch 2.5.0+cu124
• Datasets 3.0.1
• Tokenizers 0.20.1