Sparse Foundational Llama 2 Models - a neuralmagic Collection

neuralmagic 's Collections

Vision Language Models Quantization

FP8 LLMs for vLLM

Llama-3.2 Quantization

Llama-3.1 Quantization

INT8 LLMs for vLLM

INT4 LLMs for vLLM

Sparse Foundational Llama 2 Models

Compression Papers

DeepSparse Sparse LLMs

Sparse Finetuning MPT

Compressed LLMs from the Community

Sparse Foundational Llama 2 Models

updated Sep 26

Sparse pre-trained and fine-tuned Llama models made by Neural Magic + Cerebras

Enabling High-Sparsity Foundational Llama Models with Efficient Pretraining and Deployment

Paper • 2405.03594 • Published May 6 • 7
Sleeping

🏃

Llama 2 Sparse Transfer Chat Deepsparse
neuralmagic/Llama-2-7b-pruned50-retrained

Text Generation • Updated May 7 • 613

Note Our sparse Llama 2 7B base that was pruned to 50% sparsity and retrained on 50B tokens.
neuralmagic/Llama-2-7b-pruned70-retrained

Text Generation • Updated May 7 • 312

Note Our sparse Llama 2 7B base that was pruned to 70% sparsity and retrained on 150B tokens.
neuralmagic/Llama-2-7b-ultrachat200k-pruned_50

Text Generation • Updated May 15 • 30
neuralmagic/Llama-2-7b-ultrachat200k-pruned_70

Text Generation • Updated May 15 • 15
neuralmagic/Llama-2-7b-evol-code-alpaca-pruned_50

Text Generation • Updated May 15 • 19
neuralmagic/Llama-2-7b-evol-code-alpaca-pruned_70

Text Generation • Updated May 15 • 19
neuralmagic/Llama-2-7b-dolphin-open_platypus-pruned_50

Text Generation • Updated May 15 • 23
neuralmagic/Llama-2-7b-dolphin-open_platypus-pruned_70

Text Generation • Updated May 15 • 20
neuralmagic/Llama-2-7b-ultrachat200k-pruned_50-quantized-deepsparse

Text Generation • Updated May 7 • 13
neuralmagic/Llama-2-7b-ultrachat200k-pruned_70-quantized-deepsparse

Text Generation • Updated May 15 • 21
neuralmagic/Llama-2-7b-evol-code-alpaca-pruned_50-quantized-deepsparse

Text Generation • Updated May 15 • 15
neuralmagic/Llama-2-7b-evol-code-alpaca-pruned_70-quantized-deepsparse

Text Generation • Updated May 15 • 28
neuralmagic/Llama-2-7b-dolphin-open_platypus-pruned_50-quantized-deepsparse

Text Generation • Updated May 16 • 19
neuralmagic/Llama-2-7b-dolphin-open_platypus-pruned_70-quantized-deepsparse

Text Generation • Updated May 16 • 17 • 1
neuralmagic/Llama-2-7b-ultrachat200k

Text Generation • Updated May 7 • 994
neuralmagic/Llama-2-7b-evolcodealpaca

Text Generation • Updated May 7 • 31 • 1
neuralmagic/Llama-2-7b-dolphin-open_platypus

Text Generation • Updated May 15 • 15
neuralmagic/Llama-2-7b-gsm8k

Text Generation • Updated Jun 20 • 492 • 1
neuralmagic/Llama-2-7b-gsm8k-pruned_50

Text Generation • Updated Jun 20 • 17 • 1
neuralmagic/Llama-2-7b-gsm8k-pruned_70

Text Generation • Updated Jun 20 • 17
neuralmagic/Llama-2-7b-cnn-daily-mail

Text Generation • Updated May 15 • 123
neuralmagic/Llama-2-7b-cnn-daily-mail-pruned_50

Text Generation • Updated May 15 • 4
neuralmagic/Llama-2-7b-cnn-daily-mail-pruned_70

Text Generation • Updated May 15 • 7
neuralmagic/Llama-2-7b-cnn-daily-mail-pruned_50-quantized-deepsparse

Text Generation • Updated May 17 • 8
neuralmagic/Llama-2-7b-cnn-daily-mail-pruned_70-quantized-deepsparse

Text Generation • Updated May 17 • 4