robust-rlhf/Llama-3.3-70B-Instruct_ftjob-1e99f7048485-merged

Hugging Face
TEXT GENERATIONConcurrency Cost:4Model Size:70BQuant:FP8Ctx Length:32kLicense:apache-2.0Architecture:Transformer0.0K Open Weights Warm

The robust-rlhf/Llama-3.3-70B-Instruct_ftjob-1e99f7048485-merged model is a 70 billion parameter instruction-tuned Llama 3.3 variant developed by robust-rlhf. This model was fine-tuned using Unsloth and Huggingface's TRL library, achieving 2x faster training. It is designed for instruction-following tasks, leveraging its large parameter count and optimized training process for enhanced performance.

Loading preview...

Model Overview

This model, robust-rlhf/Llama-3.3-70B-Instruct_ftjob-1e99f7048485-merged, is a 70 billion parameter instruction-tuned variant of the Llama 3.3 architecture. Developed by robust-rlhf, it was fine-tuned from unsloth/Llama-3.3-70B-Instruct-bnb-4bit.

Key Characteristics

  • Architecture: Llama 3.3, instruction-tuned.
  • Parameter Count: 70 billion parameters.
  • Training Optimization: Fine-tuned using Unsloth and Huggingface's TRL library, resulting in a 2x speedup during the training process.
  • Context Length: Supports a context length of 32768 tokens.
  • License: Distributed under the Apache-2.0 license.

Use Cases

This model is suitable for a wide range of instruction-following applications, benefiting from its large parameter count and optimized fine-tuning. Its enhanced training efficiency suggests a robust and well-optimized instruction-following capability.

Popular Sampler Settings

Top 3 parameter combinations used by Featherless users for this model. Click a tab to see each config.

temperature
top_p
top_k
frequency_penalty
presence_penalty
repetition_penalty
min_p