README.md

metadata

license: apache-2.0
datasets:
  - cerebras/SlimPajama-627B
  - bigcode/starcoderdata
language:
  - en
quantized_by: bartowski

Exllama v2 Quantizations of TinyLlama-1.1B-intermediate-step-955k-token-2T at 6.0 bits per weight

Using turboderp's ExLlamaV2 v0.0.8 for quantization.

Conversion was done using wikitext-103-raw-v1-test.parquet as calibration dataset.

Original model: https://huggingface.co/TinyLlama/TinyLlama-1.1B-intermediate-step-955k-token-2T

Download instructions

With git:

git clone --single-branch --branch 6.0 https://huggingface.co/bartowski/TinyLlama-1.1B-intermediate-step-955k-token-2T-exl2

With huggingface hub (credit to TheBloke for instructions):

pip3 install huggingface-hub

To download from a different branch, add the --revision parameter:

mkdir TinyLlama-1.1B-intermediate-step-955k-token-2T-exl2
huggingface-cli download bartowski/TinyLlama-1.1B-intermediate-step-955k-token-2T-exl2 --revision 6_0 --local-dir TinyLlama-1.1B-intermediate-step-955k-token-2T-exl2 --local-dir-use-symlinks False