AtAndDev_-_ShortKing-3b-v0.2-gguf – AI Model by RichardErkhov | AlphaNeural AI

You can deploy this model and start earning money today!

RichardErkhov/
AtAndDev_-_ShortKing-3b-v0.2-gguf

like

0

endpoints_compatible

Views

No views yet

Quantization made by Richard Erkhov.

Request more models

ShortKing-3b-v0.2 - GGUF

Model creator: https://huggingface.co/AtAndDev/
Original model: https://huggingface.co/AtAndDev/ShortKing-3b-v0.2/

Name	Quant method	Size
ShortKing-3b-v0.2.Q2_K.gguf	Q2_K	1.84GB
ShortKing-3b-v0.2.IQ3_XS.gguf	IQ3_XS	1.84GB
ShortKing-3b-v0.2.IQ3_S.gguf	IQ3_S	1.84GB
ShortKing-3b-v0.2.Q3_K_S.gguf	Q3_K_S	1.84GB
ShortKing-3b-v0.2.IQ3_M.gguf	IQ3_M	1.92GB
ShortKing-3b-v0.2.Q3_K.gguf	Q3_K	1.99GB
ShortKing-3b-v0.2.Q3_K_M.gguf	Q3_K_M	1.99GB
ShortKing-3b-v0.2.Q3_K_L.gguf	Q3_K_L	2.06GB
ShortKing-3b-v0.2.IQ4_XS.gguf	IQ4_XS	1.86GB
ShortKing-3b-v0.2.Q4_0.gguf	Q4_0	1.84GB
ShortKing-3b-v0.2.IQ4_NL.gguf	IQ4_NL	1.86GB
ShortKing-3b-v0.2.Q4_K_S.gguf	Q4_K_S	2.24GB
ShortKing-3b-v0.2.Q4_K.gguf	Q4_K	2.4GB
ShortKing-3b-v0.2.Q4_K_M.gguf	Q4_K_M	2.4GB
ShortKing-3b-v0.2.Q4_1.gguf	Q4_1	2.04GB
ShortKing-3b-v0.2.Q5_0.gguf	Q5_0	2.23GB
ShortKing-3b-v0.2.Q5_K_S.gguf	Q5_K_S	2.42GB
ShortKing-3b-v0.2.Q5_K.gguf	Q5_K	2.57GB
ShortKing-3b-v0.2.Q5_K_M.gguf	Q5_K_M	2.57GB
ShortKing-3b-v0.2.Q5_1.gguf	Q5_1	2.42GB
ShortKing-3b-v0.2.Q6_K.gguf	Q6_K	3.39GB
ShortKing-3b-v0.2.Q8_0.gguf	Q8_0	3.39GB

Original model description:

license: cc-by-4.0 datasets:

Photolens/alpaca-cleaned-airoboros-2.1-no-code-oasst1-en-merged language:
en

Model overview

This model is finetuned on a merged dataset of: oasst1-en, alpaca-cleaned and airoboros-2.1-no-code on a base model: Marx-3b-V2

License: "Creative-Commons-Attribution-4.0"
Language: "en"
Size: "3.43b params"

Prompt template

Prompt template:

### SYSTEM:
<system_prompt_here>

### HUMAN:
<prompter_message_here>

### INPUT:
<input_text_here>

### RESPONSE:
<leave_a_blank_line_here>

Note: If you dont have a system or input text, do not include the tokens in the prompt.

Training Details

This model took 2:40:54 to train in LoRA on a single A100 40gb GPU.

epochs: 1
train batch size: 8
eval batch size: 8
gradient accumulation steps: 1
maximum gradient normal: 0.3
learning rate: 2e-4
weight decay: 0.001
optimizer: paged_adamw_32bit
learning rate schedule: cosine
warmup ratio (linear): 0.03