PrepAI/la-cohere-aya-expanse-8b-8k_4bit_2ep-lora

la-cohere-aya-expanse-8b-8k_4bit_r32_alpha_16_lr1e-4_3ep

This model is a fine-tuned version of CohereForAI/aya-expanse-8b on the la_chat_data_v4 dataset. It achieves the following results on the evaluation set:

• Loss: 0.3464

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: 0.0001
• train_batch_size: 1
• eval_batch_size: 1
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 4
• optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: cosine
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 2.0

Training results

Training Loss	Epoch	Step	Validation Loss
0.5011	0.6649	10000	0.4348
0.4008	1.3299	20000	0.3684
0.4035	1.9948	30000	0.3464

Framework versions

• PEFT 0.11.1
• Transformers 4.46.1
• Pytorch 2.4.1+cu121
• Datasets 2.20.0
• Tokenizers 0.20.3