rylansed/blvflag_llama3.0-GGUF

blvflag_llama

This model is a fine-tuned version of TinyLlama/TinyLlama-1.1B-Chat-v1.0.

It was trained using the Mostly Basic Python Problems data set https://github.com/google-research/google-research/blob/master/mbpp/README.md

Intended uses & limitations

A small parameter model to be used in the BLVFLAG tool. https://github.com/rylansedlacek/blvflag

Training and evaluation data

Trained using Trainer and Pytorch.

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 0.0002
• train_batch_size: 1
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 8
• total_train_batch_size: 8
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 1
• training_steps: 700

Training results

Step Training Loss 500 1.446500

Framework versions

• Transformers 4.51.3
• Pytorch 2.6.0+cu124
• Datasets 3.6.0
• Tokenizers 0.21.1