meta-llama/Meta-Llama-3-8B-Instruct - W8A8_FP8 Compression

This is a compressed model using llmcompressor.

Compression Configuration

  • Base Model: meta-llama/Meta-Llama-3-8B-Instruct
  • Compression Scheme: W8A8_FP8
  • Dataset: HuggingFaceH4/ultrachat_200k
  • Dataset Split: train_sft
  • Number of Samples: 512
  • Preprocessor: chat
  • Maximum Sequence Length: 8192

Sample Output

Prompt:

<|begin_of_text|><|start_header_id|>user<|end_header_id|>

Who is Alan Turing?<|eot_id|>

Output:

<|begin_of_text|><|begin_of_text|><|start_header_id|>user<|end_header_id|>

Who is Alan Turing?<|eot_id|><|start_header_id|>assistant<|end_header_id|>

Alan Turing (1912-1954) was a British mathematician, computer scientist, logician, and philosopher who made significant contributions to the development of computer science, artificial intelligence, and cryptography. He is widely considered one of the most important figures in the history of computer science and is often referred to as the "father of computer science" or the "father of artificial intelligence."

Turing was born in London, England, and was a brilliant student at King's College, Cambridge, where he studied mathematics. During World War II, he worked at the Government Code and Cypher School (GC&CS) at

Evaluation

Downloads last month
27
Safetensors
Model size
8.03B params
Tensor type
BF16
·
F8_E4M3
·
Inference Examples
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social visibility and check back later, or deploy to Inference Endpoints (dedicated) instead.

Model tree for espressor/meta-llama.Meta-Llama-3-8B-Instruct_W8A8_FP8

Quantized
(188)
this model

Dataset used to train espressor/meta-llama.Meta-Llama-3-8B-Instruct_W8A8_FP8