models/config-user.yaml

LLamandementFineTuneAvecNotation16Q.gguf$: {}
My-LLamandement-fine-tuned$:
  loader: Transformers
  cpu_memory: 0
  auto_devices: false
  disk: false
  cpu: false
  bf16: false
  load_in_8bit: false
  trust_remote_code: false
  no_use_fast: false
  use_flash_attention_2: false
  load_in_4bit: false
  compute_dtype: float16
  quant_type: nf4
  use_double_quant: false
  disable_exllama: false
  disable_exllamav2: false
  compress_pos_emb: 1
  alpha_value: 1
  rope_freq_base: 1000000
  gpu_memory_0: 0
LLamandementFineTuneSansNotation16Q.gguf$:
  instruction_template: Llama-v2
  loader: llama.cpp
  cpu: false
  threads: 0
  threads_batch: 0
  n_batch: 512
  no_mmap: false
  mlock: false
  no_mul_mat_q: false
  n_gpu_layers: 0
  tensor_split: ''
  n_ctx: 2048
  compress_pos_emb: 1
  alpha_value: 1
  rope_freq_base: 0
  numa: false
  no_offload_kqv: false
  row_split: false
  tensorcores: false
  streaming_llm: false
  attention_sink_size: 5