v000000/L3-8B-Poppy-Moonfall-C-imat-GGUFs 🌙

This model was converted to GGUF format from v000000/L3-8B-Poppy-Moonfall-C using llama.cpp. Refer to the original model card for more details on the model

Various GGUF format quants and pre-generated imatrix data for v000000/Llama-3-8B-Poppy-Moonfall-C

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

Update/Notice:

This has a tendency for endless generations. It likes a bit of penalty parameters if this occurs.

0.95 temp
80 top k
0.95 top p
0.1 min p
1.1 rep pen

slices:
  - sources:
      - model: v000000/L3-8B-Poppy-Sunspice-experiment-c+Blackroot/Llama-3-8B-Abomination-LORA
        layer_range: [0, 32]
      - model: v000000/L3-8B-Poppy-Sunspice-experiment-c+ResplendentAI/BlueMoon_Llama3
        layer_range: [0, 32]
merge_method: slerp
base_model: v000000/L3-8B-Poppy-Sunspice-experiment-c+Blackroot/Llama-3-8B-Abomination-LORA
parameters:
  t:
    - filter: self_attn
      value: [0, 0.5, 0.3, 0.7, 1]
    - filter: mlp
      value: [1, 0.5, 0.7, 0.3, 0]
    - value: 0.5
dtype: bfloat16
random_seed: 0

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|>