sanjay920/mistral-9.5-fc-yaml-v1

rubra-9.5b-yaml_v1

This model is a fine-tuned version of models/rubra-9.5b-base on the yaml-simple, the yaml-multiple, the yaml-parallel, the yaml-parallel_multiple, the yaml-relevance, the yaml-sql, the yaml-rest, the yaml-gptscript-x8 and the yaml-chain_of_function datasets.

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: 2e-05
• train_batch_size: 2
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 16
• total_train_batch_size: 32
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: cosine
• num_epochs: 9.0
• mixed_precision_training: Native AMP

Training results

Framework versions

• Transformers 4.39.3
• Pytorch 2.2.0+cu121
• Datasets 2.18.0
• Tokenizers 0.15.2