ruadapt_llama2_7b_v0.1

ruadapt_llama2_7b_v0.1

This model is a fine-tuned (embeddings, lm head) version of TheBloke/Llama-2-7B-fp16 on the Russian dataset (33GB). It achieves the following results on the evaluation set:

• Loss: 2.7569
• Accuracy: 0.4617

Instruct version: https://huggingface.co/rccmsu/ruadapt_saiga2_7b_v0.1

Model description

Russian adaptation of LLaMa-2-7B by replacing the tokenizer. Paper: Tikhomirov M., Chernyshev D. Impact of Tokenization on LLaMa Russian Adaptation //arXiv preprint arXiv:2312.02598. – 2023.

Intended uses & limitations

LLAMA 2 COMMUNITY LICENSE AGREEMENT

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 2e-05
• train_batch_size: 6
• eval_batch_size: 6
• seed: 42
• distributed_type: multi-GPU
• num_devices: 16
• gradient_accumulation_steps: 2
• total_train_batch_size: 192
• total_eval_batch_size: 96
• optimizer: Adam with betas=(0.9,0.95) and epsilon=1e-05
• lr_scheduler_type: linear
• num_epochs: 2.0

Framework versions

• Transformers 4.34.0
• Pytorch 2.0.1+cu118
• Datasets 2.14.5
• Tokenizers 0.14.1