Quantization made by Richard Erkhov.

Llama-3-13B-Instruct - GGUF

Model creator: https://huggingface.co/elinas/
Original model: https://huggingface.co/elinas/Llama-3-13B-Instruct/

Name	Quant method	Size
Llama-3-13B-Instruct.Q2_K.gguf	Q2_K	4.68GB
Llama-3-13B-Instruct.IQ3_XS.gguf	IQ3_XS	5.18GB
Llama-3-13B-Instruct.IQ3_S.gguf	IQ3_S	5.45GB
Llama-3-13B-Instruct.Q3_K_S.gguf	Q3_K_S	5.42GB
Llama-3-13B-Instruct.IQ3_M.gguf	IQ3_M	5.61GB
Llama-3-13B-Instruct.Q3_K.gguf	Q3_K	5.98GB
Llama-3-13B-Instruct.Q3_K_M.gguf	Q3_K_M	5.98GB
Llama-3-13B-Instruct.Q3_K_L.gguf	Q3_K_L	6.47GB
Llama-3-13B-Instruct.IQ4_XS.gguf	IQ4_XS	6.69GB
Llama-3-13B-Instruct.Q4_0.gguf	Q4_0	6.97GB
Llama-3-13B-Instruct.IQ4_NL.gguf	IQ4_NL	7.04GB
Llama-3-13B-Instruct.Q4_K_S.gguf	Q4_K_S	7.01GB
Llama-3-13B-Instruct.Q4_K.gguf	Q4_K	7.38GB
Llama-3-13B-Instruct.Q4_K_M.gguf	Q4_K_M	7.38GB
Llama-3-13B-Instruct.Q4_1.gguf	Q4_1	7.7GB
Llama-3-13B-Instruct.Q5_0.gguf	Q5_0	8.43GB
Llama-3-13B-Instruct.Q5_K_S.gguf	Q5_K_S	8.43GB
Llama-3-13B-Instruct.Q5_K.gguf	Q5_K	8.64GB
Llama-3-13B-Instruct.Q5_K_M.gguf	Q5_K_M	8.64GB
Llama-3-13B-Instruct.Q5_1.gguf	Q5_1	9.16GB
Llama-3-13B-Instruct.Q6_K.gguf	Q6_K	9.98GB
Llama-3-13B-Instruct.Q8_0.gguf	Q8_0	12.92GB

Original model description:

base_model: - meta-llama/Meta-Llama-3-8B-Instruct library_name: transformers tags: - mergekit - merge license: llama3

Llama-3-13B-Instruct

This is a merge of pre-trained language models created using mergekit.

The goal was to create a Llama 3 13B model to have a "mid" sized model which Meta has released in the past, but I would consider this a base model to be further finetuned on. Surprisingly, it is usable for chat and storywriting with Llama 3 Instruct template, though it does occasionally have some grammatical quirks like L3-120B.

Logical ability (programming, math, science, etc.) has been deteriorated by the merge process.

Use no repetition penalty or <1.05 or it might go a bit haywire, other than that, it is suitable for writing use. I have not tested it against L3 8B in that regard.

Finetuned Version

A finetuned version of this model can be found at elinas/Llama-3-13B-Instruct-ft which seems to improve performance.

Merge Details

Merge Method

This model was merged using the passthrough merge method.

Models Merged

The following models were included in the merge:

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

Configuration

The following YAML configuration was used to produce this model:

dtype: bfloat16
merge_method: passthrough
slices:
- sources:
  - layer_range: [0, 10]
    model: meta-llama/Meta-Llama-3-8B-Instruct
- sources:
  - layer_range: [5, 15]
    model: meta-llama/Meta-Llama-3-8B-Instruct
- sources:
  - layer_range: [10, 20]
    model: meta-llama/Meta-Llama-3-8B-Instruct
- sources:
  - layer_range: [15, 25]
    model: meta-llama/Meta-Llama-3-8B-Instruct
- sources:
  - layer_range: [20, 25]
    model: meta-llama/Meta-Llama-3-8B-Instruct
- sources:
  - layer_range: [22, 32]
    model: meta-llama/Meta-Llama-3-8B-Instruct

Model Evaluation

TBD - submitted