README.md

metadata

library_name: transformers
language:
  - en
license: unknown
tags:
  - exl2

miqu-1-70b-sf - EXL2 2.25bpw

This is a 2.25bpw EXL2 quant of 152334H/miqu-1-70b-sf

Details about the model can be found at the above model page.

EXL2 Version

These quants were made with exllamav2 version 0.0.18. Quants made on this version of EXL2 may not work on older versions of the exllamav2 library.

If you have problems loading these models, please update Text Generation WebUI to the latest version.

Perplexity Scoring

Below are the perplexity scores for the EXL2 models. A lower score is better.

Quant Level	Perplexity Score
5.0	4.2637
4.5	4.2876
4.0	4.3097
3.5	4.4459
3.0	4.6504
2.75	5.1638
2.5	5.1715
2.25	6.0848

EQ Bench

Here are the EQ Bench scores for the EXL2 quants using Alpaca, ChatML, Mistral, Vicuna-v1.1 and Vicuna-v0 prompt templates. A higher score is better.

Quant Size	Instruct Template	Score
5.0	ChatML	79.91
5.0	Alpaca	81.45
5.0	Mistral	81.11
5.0	Vicuna-v1.1	78.37
5.0	Vicuna-v0	76.64
4.5	ChatML	80.64
4.5	Alpaca	80.9
4.5	Mistral	81.65
4.5	Vicuna-v1.1	77.04
4.5	Vicuna-v0	74.6
4.0	ChatML	80.78
4.0	Alpaca	79.53
4.0	Mistral	82.78
4.0	Vicuna-v1.1	79.17
4.0	Vicuna-v0	76.41
3.5	ChatML	81.11
3.5	Alpaca	82.42
3.5	Mistral	82.34
3.5	Vicuna-v1.1	81.04
3.5	Vicuna-v0	78.09
3.0	ChatML	79.13
3.0	Alpaca	77.74
3.0	Mistral	80.11
3.0	Vicuna-v1.1	79.38
3.0	Vicuna-v0	77.25
2.75	ChatML	79.6
2.75	Alpaca	77.85
2.75	Mistral	79.71
2.75	Vicuna-v1.1	76.93
2.75	Vicuna-v0	75.91
2.5	ChatML	77.45
2.5	Alpaca	77.0
2.5	Mistral	78.4
2.5	Vicuna-v1.1	75.86
2.5	Vicuna-v0	75.25
2.25	ChatML	77.18
2.25	Alpaca	74.06
2.25	Mistral	76.75
2.25	Vicuna-v1.1	75.56
2.25	Vicuna-v0	74.28

Perplexity Script

This was the script used for perplexity testing.

#!/bin/bash

# Activate the conda environment
source ~/miniconda3/etc/profile.d/conda.sh
conda activate exllamav2

# Set the model name and bit size
MODEL_NAME="miqu-1-70b-sf"
BIT_PRECISIONS=(8.0 7.0 6.0 5.5 5.0 4.5 4.0 3.5 3.0 2.75 2.5 2.25)

# Print the markdown table header
echo "| Quant Level | Perplexity Score |"
echo "|-------------|------------------|"

for BIT_PRECISION in "${BIT_PRECISIONS[@]}"
do
  MODEL_DIR="models/${MODEL_NAME}_exl2_${BIT_PRECISION}bpw"
  if [ -d "$MODEL_DIR" ]; then
    output=$(python test_inference.py -m "$MODEL_DIR" -gs 22,24 -ed data/wikitext/wikitext-2-v1.parquet)
    score=$(echo "$output" | grep -oP 'Evaluation perplexity: \K[\d.]+')
    echo "| $BIT_PRECISION | $score |"
  fi
done

Quant Details

This is the script used for quantization.

#!/bin/bash

# Activate the conda environment
source ~/miniconda3/etc/profile.d/conda.sh
conda activate exllamav2

# Set the model name and bit size
MODEL_NAME="miqu-1-70b-sf"

# Define variables
MODEL_DIR="models/152334H_miqu-1-70b-sf"
OUTPUT_DIR="exl2_$MODEL_NAME"
MEASUREMENT_FILE="measurements/$MODEL_NAME.json"

# Create the measurement file if needed
if [ ! -f "$MEASUREMENT_FILE" ]; then
    echo "Creating $MEASUREMENT_FILE"
    # Create directories
    if [ -d "$OUTPUT_DIR" ]; then
        rm -r "$OUTPUT_DIR"
    fi
    mkdir "$OUTPUT_DIR"

    python convert.py -i $MODEL_DIR -o $OUTPUT_DIR -nr -om $MEASUREMENT_FILE
fi

# Choose one of the below. Either create a single quant for testing or a batch of them.
# BIT_PRECISIONS=(5.0)
BIT_PRECISIONS=(8.0 7.0 6.0 5.5 5.0 4.5 4.0 3.5 3.0 2.75 2.5 2.25)

for BIT_PRECISION in "${BIT_PRECISIONS[@]}"
do
    CONVERTED_FOLDER="models/${MODEL_NAME}_exl2_${BIT_PRECISION}bpw"

    # If it doesn't already exist, make the quant
    if [ ! -d "$CONVERTED_FOLDER" ]; then

        echo "Creating $CONVERTED_FOLDER"

        # Create directories
        if [ -d "$OUTPUT_DIR" ]; then
            rm -r "$OUTPUT_DIR"
        fi
        mkdir "$OUTPUT_DIR"
        mkdir "$CONVERTED_FOLDER"
        
        # Run conversion commands  
        python convert.py -i $MODEL_DIR -o $OUTPUT_DIR -nr -m $MEASUREMENT_FILE -b $BIT_PRECISION -cf $CONVERTED_FOLDER

    fi
done