Edit model card

Refer to the original models for best usage.


Mergekit Recipe

# Mistral Merging
merge_method: linear
dtype: float16
models:
  - model: KoboldAI/Mistral-7B-Holodeck-1 # Sci-Fi Novel Base (45%)
    parameters:
      weight: 0.45
  - model: KoboldAI/Mistral-7B-Erebus-v3 # NSFW Novel Base (45%)
    parameters:
      weight: 0.45
  - model: fblgit/UNA-TheBeagle-7b-v1 # Instruct (2.5%)
    parameters:
      weight: 0.025
  - model: teknium/airoboros-mistral2.2-7b # Instruct (2.5%)
    parameters:
      weight: 0.025
  - model: senseable/WestLake-7B-v2 # RP (2.5%)
    parameters:
      weight: 0.025
  - model: chargoddard/loyal-piano-m7 # RP (1.25%)
    parameters:
      weight: 0.0125
  - model: NeverSleep/Noromaid-7b-v0.2 # RP (1.25%)
    parameters:
      weight: 0.0125
name: EmptiMM3-7B
---

# Mistral Stacking
merge_method: passthrough
dtype: float16
slices:
  - sources:
    - model: EmptiMM3-7B
      layer_range: [0, 24]
  - sources:
    - model: EmptiMM3-7B
      layer_range: [8, 32]
name: EmptiMM3-Stacked-Unhealed-10.7B
---

# Mistral Stack Healing
merge_method: task_arithmetic
dtype: float16
base_model: Undi95/Mistral-11B-v0.1
models:
  - model: Sao10K/Fimbulvetr-11B-v2 # RP
    parameters:
      weight: 1.0
  - model: EmptiMM3-Stacked-Unhealed-10.7B
    parameters:
      weight: 1.0
name: EmptiSolar-4-UnfixedConfig-10.7B
---

# Healed Stack Config Fixing
merge_method: task_arithmetic
dtype: float16
base_model: upstage/SOLAR-10.7B-v1.0
models:
  - model: EmptiSolar-4-UnfixedConfig-10.7B
    parameters:
      weight: 1.0
Downloads last month
6
GGUF
Model size
10.7B params
Architecture
llama

4-bit

Inference API
Unable to determine this model's library. Check the docs .

Model tree for PJMixers/EmptiSolar-4-10.7B-GGUF