adapter_config.json

{
  "alora_invocation_tokens": null,
  "alpha_pattern": {
    "model.decoder.layers.0.encoder_attn.k_proj": 128,
    "model.decoder.layers.0.encoder_attn.out_proj": 128,
    "model.decoder.layers.0.encoder_attn.q_proj": 128,
    "model.decoder.layers.0.encoder_attn.v_proj": 128,
    "model.decoder.layers.0.self_attn.k_proj": 64,
    "model.decoder.layers.0.self_attn.out_proj": 64,
    "model.decoder.layers.0.self_attn.q_proj": 64,
    "model.decoder.layers.0.self_attn.v_proj": 64,
    "model.decoder.layers.1.encoder_attn.k_proj": 128,
    "model.decoder.layers.1.encoder_attn.out_proj": 128,
    "model.decoder.layers.1.encoder_attn.q_proj": 128,
    "model.decoder.layers.1.encoder_attn.v_proj": 128,
    "model.decoder.layers.1.self_attn.k_proj": 64,
    "model.decoder.layers.1.self_attn.out_proj": 64,
    "model.decoder.layers.1.self_attn.q_proj": 64,
    "model.decoder.layers.1.self_attn.v_proj": 64,
    "model.decoder.layers.10.encoder_attn.k_proj": 128,
    "model.decoder.layers.10.encoder_attn.out_proj": 128,
    "model.decoder.layers.10.encoder_attn.q_proj": 128,
    "model.decoder.layers.10.encoder_attn.v_proj": 128,
    "model.decoder.layers.10.self_attn.k_proj": 64,
    "model.decoder.layers.10.self_attn.out_proj": 64,
    "model.decoder.layers.10.self_attn.q_proj": 64,
    "model.decoder.layers.10.self_attn.v_proj": 64,
    "model.decoder.layers.11.encoder_attn.k_proj": 128,
    "model.decoder.layers.11.encoder_attn.out_proj": 128,
    "model.decoder.layers.11.encoder_attn.q_proj": 128,
    "model.decoder.layers.11.encoder_attn.v_proj": 128,
    "model.decoder.layers.11.self_attn.k_proj": 64,
    "model.decoder.layers.11.self_attn.out_proj": 64,
    "model.decoder.layers.11.self_attn.q_proj": 64,
    "model.decoder.layers.11.self_attn.v_proj": 64,
    "model.decoder.layers.2.encoder_attn.k_proj": 128,
    "model.decoder.layers.2.encoder_attn.out_proj": 128,
    "model.decoder.layers.2.encoder_attn.q_proj": 128,
    "model.decoder.layers.2.encoder_attn.v_proj": 128,
    "model.decoder.layers.2.self_attn.k_proj": 64,
    "model.decoder.layers.2.self_attn.out_proj": 64,
    "model.decoder.layers.2.self_attn.q_proj": 64,
    "model.decoder.layers.2.self_attn.v_proj": 64,
    "model.decoder.layers.3.encoder_attn.k_proj": 128,
    "model.decoder.layers.3.encoder_attn.out_proj": 128,
    "model.decoder.layers.3.encoder_attn.q_proj": 128,
    "model.decoder.layers.3.encoder_attn.v_proj": 128,
    "model.decoder.layers.3.self_attn.k_proj": 64,
    "model.decoder.layers.3.self_attn.out_proj": 64,
    "model.decoder.layers.3.self_attn.q_proj": 64,
    "model.decoder.layers.3.self_attn.v_proj": 64,
    "model.decoder.layers.4.encoder_attn.k_proj": 128,
    "model.decoder.layers.4.encoder_attn.out_proj": 128,
    "model.decoder.layers.4.encoder_attn.q_proj": 128,
    "model.decoder.layers.4.encoder_attn.v_proj": 128,
    "model.decoder.layers.4.self_attn.k_proj": 64,
    "model.decoder.layers.4.self_attn.out_proj": 64,
    "model.decoder.layers.4.self_attn.q_proj": 64,
    "model.decoder.layers.4.self_attn.v_proj": 64,
    "model.decoder.layers.5.encoder_attn.k_proj": 128,
    "model.decoder.layers.5.encoder_attn.out_proj": 128,
    "model.decoder.layers.5.encoder_attn.q_proj": 128,
    "model.decoder.layers.5.encoder_attn.v_proj": 128,
    "model.decoder.layers.5.self_attn.k_proj": 64,
    "model.decoder.layers.5.self_attn.out_proj": 64,
    "model.decoder.layers.5.self_attn.q_proj": 64,
    "model.decoder.layers.5.self_attn.v_proj": 64,
    "model.decoder.layers.6.encoder_attn.k_proj": 128,
    "model.decoder.layers.6.encoder_attn.out_proj": 128,
    "model.decoder.layers.6.encoder_attn.q_proj": 128,
    "model.decoder.layers.6.encoder_attn.v_proj": 128,
    "model.decoder.layers.6.self_attn.k_proj": 64,
    "model.decoder.layers.6.self_attn.out_proj": 64,
    "model.decoder.layers.6.self_attn.q_proj": 64,
    "model.decoder.layers.6.self_attn.v_proj": 64,
    "model.decoder.layers.7.encoder_attn.k_proj": 128,
    "model.decoder.layers.7.encoder_attn.out_proj": 128,
    "model.decoder.layers.7.encoder_attn.q_proj": 128,
    "model.decoder.layers.7.encoder_attn.v_proj": 128,
    "model.decoder.layers.7.self_attn.k_proj": 64,
    "model.decoder.layers.7.self_attn.out_proj": 64,
    "model.decoder.layers.7.self_attn.q_proj": 64,
    "model.decoder.layers.7.self_attn.v_proj": 64,
    "model.decoder.layers.8.encoder_attn.k_proj": 128,
    "model.decoder.layers.8.encoder_attn.out_proj": 128,
    "model.decoder.layers.8.encoder_attn.q_proj": 128,
    "model.decoder.layers.8.encoder_attn.v_proj": 128,
    "model.decoder.layers.8.self_attn.k_proj": 64,
    "model.decoder.layers.8.self_attn.out_proj": 64,
    "model.decoder.layers.8.self_attn.q_proj": 64,
    "model.decoder.layers.8.self_attn.v_proj": 64,
    "model.decoder.layers.9.encoder_attn.k_proj": 128,
    "model.decoder.layers.9.encoder_attn.out_proj": 128,
    "model.decoder.layers.9.encoder_attn.q_proj": 128,
    "model.decoder.layers.9.encoder_attn.v_proj": 128,
    "model.decoder.layers.9.self_attn.k_proj": 64,
    "model.decoder.layers.9.self_attn.out_proj": 64,
    "model.decoder.layers.9.self_attn.q_proj": 64,
    "model.decoder.layers.9.self_attn.v_proj": 64
  },
  "arrow_config": null,
  "auto_mapping": null,
  "base_model_name_or_path": "facebook/nllb-200-distilled-600M",
  "bias": "none",
  "corda_config": null,
  "ensure_weight_tying": false,
  "eva_config": null,
  "exclude_modules": null,
  "fan_in_fan_out": false,
  "inference_mode": true,
  "init_lora_weights": true,
  "layer_replication": null,
  "layers_pattern": null,
  "layers_to_transform": null,
  "loftq_config": {},
  "lora_alpha": 16,
  "lora_bias": false,
  "lora_dropout": 0.1,
  "megatron_config": null,
  "megatron_core": "megatron.core",
  "modules_to_save": null,
  "peft_type": "LORA",
  "peft_version": "0.18.1",
  "qalora_group_size": 16,
  "r": 8,
  "rank_pattern": {
    "model.decoder.layers.0.encoder_attn.k_proj": 64,
    "model.decoder.layers.0.encoder_attn.out_proj": 64,
    "model.decoder.layers.0.encoder_attn.q_proj": 64,
    "model.decoder.layers.0.encoder_attn.v_proj": 64,
    "model.decoder.layers.0.self_attn.k_proj": 32,
    "model.decoder.layers.0.self_attn.out_proj": 32,
    "model.decoder.layers.0.self_attn.q_proj": 32,
    "model.decoder.layers.0.self_attn.v_proj": 32,
    "model.decoder.layers.1.encoder_attn.k_proj": 64,
    "model.decoder.layers.1.encoder_attn.out_proj": 64,
    "model.decoder.layers.1.encoder_attn.q_proj": 64,
    "model.decoder.layers.1.encoder_attn.v_proj": 64,
    "model.decoder.layers.1.self_attn.k_proj": 32,
    "model.decoder.layers.1.self_attn.out_proj": 32,
    "model.decoder.layers.1.self_attn.q_proj": 32,
    "model.decoder.layers.1.self_attn.v_proj": 32,
    "model.decoder.layers.10.encoder_attn.k_proj": 64,
    "model.decoder.layers.10.encoder_attn.out_proj": 64,
    "model.decoder.layers.10.encoder_attn.q_proj": 64,
    "model.decoder.layers.10.encoder_attn.v_proj": 64,
    "model.decoder.layers.10.self_attn.k_proj": 32,
    "model.decoder.layers.10.self_attn.out_proj": 32,
    "model.decoder.layers.10.self_attn.q_proj": 32,
    "model.decoder.layers.10.self_attn.v_proj": 32,
    "model.decoder.layers.11.encoder_attn.k_proj": 64,
    "model.decoder.layers.11.encoder_attn.out_proj": 64,
    "model.decoder.layers.11.encoder_attn.q_proj": 64,
    "model.decoder.layers.11.encoder_attn.v_proj": 64,
    "model.decoder.layers.11.self_attn.k_proj": 32,
    "model.decoder.layers.11.self_attn.out_proj": 32,
    "model.decoder.layers.11.self_attn.q_proj": 32,
    "model.decoder.layers.11.self_attn.v_proj": 32,
    "model.decoder.layers.2.encoder_attn.k_proj": 64,
    "model.decoder.layers.2.encoder_attn.out_proj": 64,
    "model.decoder.layers.2.encoder_attn.q_proj": 64,
    "model.decoder.layers.2.encoder_attn.v_proj": 64,
    "model.decoder.layers.2.self_attn.k_proj": 32,
    "model.decoder.layers.2.self_attn.out_proj": 32,
    "model.decoder.layers.2.self_attn.q_proj": 32,
    "model.decoder.layers.2.self_attn.v_proj": 32,
    "model.decoder.layers.3.encoder_attn.k_proj": 64,
    "model.decoder.layers.3.encoder_attn.out_proj": 64,
    "model.decoder.layers.3.encoder_attn.q_proj": 64,
    "model.decoder.layers.3.encoder_attn.v_proj": 64,
    "model.decoder.layers.3.self_attn.k_proj": 32,
    "model.decoder.layers.3.self_attn.out_proj": 32,
    "model.decoder.layers.3.self_attn.q_proj": 32,
    "model.decoder.layers.3.self_attn.v_proj": 32,
    "model.decoder.layers.4.encoder_attn.k_proj": 64,
    "model.decoder.layers.4.encoder_attn.out_proj": 64,
    "model.decoder.layers.4.encoder_attn.q_proj": 64,
    "model.decoder.layers.4.encoder_attn.v_proj": 64,
    "model.decoder.layers.4.self_attn.k_proj": 32,
    "model.decoder.layers.4.self_attn.out_proj": 32,
    "model.decoder.layers.4.self_attn.q_proj": 32,
    "model.decoder.layers.4.self_attn.v_proj": 32,
    "model.decoder.layers.5.encoder_attn.k_proj": 64,
    "model.decoder.layers.5.encoder_attn.out_proj": 64,
    "model.decoder.layers.5.encoder_attn.q_proj": 64,
    "model.decoder.layers.5.encoder_attn.v_proj": 64,
    "model.decoder.layers.5.self_attn.k_proj": 32,
    "model.decoder.layers.5.self_attn.out_proj": 32,
    "model.decoder.layers.5.self_attn.q_proj": 32,
    "model.decoder.layers.5.self_attn.v_proj": 32,
    "model.decoder.layers.6.encoder_attn.k_proj": 64,
    "model.decoder.layers.6.encoder_attn.out_proj": 64,
    "model.decoder.layers.6.encoder_attn.q_proj": 64,
    "model.decoder.layers.6.encoder_attn.v_proj": 64,
    "model.decoder.layers.6.self_attn.k_proj": 32,
    "model.decoder.layers.6.self_attn.out_proj": 32,
    "model.decoder.layers.6.self_attn.q_proj": 32,
    "model.decoder.layers.6.self_attn.v_proj": 32,
    "model.decoder.layers.7.encoder_attn.k_proj": 64,
    "model.decoder.layers.7.encoder_attn.out_proj": 64,
    "model.decoder.layers.7.encoder_attn.q_proj": 64,
    "model.decoder.layers.7.encoder_attn.v_proj": 64,
    "model.decoder.layers.7.self_attn.k_proj": 32,
    "model.decoder.layers.7.self_attn.out_proj": 32,
    "model.decoder.layers.7.self_attn.q_proj": 32,
    "model.decoder.layers.7.self_attn.v_proj": 32,
    "model.decoder.layers.8.encoder_attn.k_proj": 64,
    "model.decoder.layers.8.encoder_attn.out_proj": 64,
    "model.decoder.layers.8.encoder_attn.q_proj": 64,
    "model.decoder.layers.8.encoder_attn.v_proj": 64,
    "model.decoder.layers.8.self_attn.k_proj": 32,
    "model.decoder.layers.8.self_attn.out_proj": 32,
    "model.decoder.layers.8.self_attn.q_proj": 32,
    "model.decoder.layers.8.self_attn.v_proj": 32,
    "model.decoder.layers.9.encoder_attn.k_proj": 64,
    "model.decoder.layers.9.encoder_attn.out_proj": 64,
    "model.decoder.layers.9.encoder_attn.q_proj": 64,
    "model.decoder.layers.9.encoder_attn.v_proj": 64,
    "model.decoder.layers.9.self_attn.k_proj": 32,
    "model.decoder.layers.9.self_attn.out_proj": 32,
    "model.decoder.layers.9.self_attn.q_proj": 32,
    "model.decoder.layers.9.self_attn.v_proj": 32
  },
  "revision": null,
  "target_modules": [
    "q_proj",
    "v_proj",
    "out_proj",
    "k_proj"
  ],
  "target_parameters": null,
  "task_type": "SEQ_2_SEQ_LM",
  "trainable_token_indices": null,
  "use_dora": false,
  "use_qalora": false,
  "use_rslora": false
}