CharlesLi/llama_2_rlhf_safe_llama_3_70B_default_1000_full

TEXT GENERATIONConcurrency Cost:1Model Size:7BQuant:FP8Ctx Length:4kPublished:Jan 13, 2025License:llama2Architecture:Transformer Open Weights Cold

The CharlesLi/llama_2_rlhf_safe_llama_3_70B_default_1000_full model is a 7 billion parameter language model fine-tuned from meta-llama/Llama-2-7b-chat-hf. This model was fine-tuned using a generator dataset, achieving a loss of 0.8687 on its evaluation set. It is based on the Llama 2 architecture and is intended for applications requiring a Llama 2-based model with specific fine-tuning characteristics.

Loading preview...

Model Overview

This model, llama_2_rlhf_safe_llama_3_70B_default_1000_full, is a 7 billion parameter language model derived from the meta-llama/Llama-2-7b-chat-hf base model. It has undergone a specific fine-tuning process using a generator dataset, resulting in an evaluation loss of 0.8687.

Key Training Details

  • Base Model: meta-llama/Llama-2-7b-chat-hf
  • Parameters: 7 Billion
  • Fine-tuning Objective: Generator dataset
  • Evaluation Loss: 0.8687
  • Hyperparameters:
    • Learning Rate: 2e-05
    • Optimizer: Adam with betas=(0.9, 0.999) and epsilon=1e-08
    • Epochs: 1
    • Total Train Batch Size: 32 (across 4 GPUs)

Intended Use Cases

While specific intended uses and limitations are not detailed in the provided information, this model is suitable for developers looking for a Llama 2-7B variant that has been fine-tuned on a generator dataset. Its performance metrics suggest it has learned patterns from this specific dataset, which could be beneficial for tasks aligned with its training data.