L3-8B-MegaSerpentine-imat-GGUFs

This model was converted to GGUF format from v000000/L3-8B-MegaSerpentine Refer to the original model card for more details on the model.

Various GGUF format quants including FP16 and pre-generated imatrix data for v000000/L3-8B-MegaSerpentine

List of quants in repo:

• Q8_0 imatrix ~8.3GB
• Q6_K imatrix ~6.4GB
• Q5_K_S imatrix ~5.4GB
• IQ4_XS imatrix ~4.3 GB
• FP16

,=e -. _,-' ,=e -. _,-' ,=e `-. _,-'

models:
  - model: Nitral-AI/Poppy_Porpoise-0.72-L3-8B+Blackroot/Llama-3-8B-Abomination-LORA
  - model: Hastagaras/Jamet-8B-L3-MK.V-Blackroot
  - model: cgato/L3-TheSpice-8b-v0.8.3
  - model: Vdr1/L3-8B-Sunfall-v0.3-Stheno-v3.2
  - model: jondurbin/bagel-8b-v1.0
  - model: lodrick-the-lafted/Fuselage-8B
  - model: HPAI-BSC/Llama3-Aloe-8B-Alpha
  - model: maldv/llama-3-fantasy-writer-8b+ToastyPigeon/Llama-3-8B-Instruct-SpringDragon-V2-QLoRA
  - model: lodrick-the-lafted/Limon-8B
  - model: Nitral-AI/Hathor_Stable-v0.2-L3-8B
  - model: princeton-nlp/Llama-3-Instruct-8B-SimPO
  - model: jeiku/Orthocopter_8B+mpasila/Llama-3-LimaRP-Instruct-LoRA-8B
  - model: jondurbin/bagel-8b-v1.0
  - model: v000000/sauce
base_model: princeton-nlp/Llama-3-Instruct-8B-SimPO
merge_method: model_stock
dtype: bfloat16

Prompt Template:

<|begin_of_text|><|start_header_id|>system<|end_header_id|>

{system_prompt}<|eot_id|><|start_header_id|>user<|end_header_id|>

{input}<|eot_id|><|start_header_id|>assistant<|end_header_id|>

{output}<|eot_id|>

./llama-quantize --imatrix ./imatrix.dat ./L3-8B-MegaSerpentine-Tria.fp16.gguf name quantsize