lora_delta_product_r_m0.5_constant/config.json

{
  "architectures": [
    "TpttModel"
  ],
  "attention_bias": false,
  "attention_dropout": 0.0,
  "auto_map": {
    "AutoConfig": "configuration_tptt.TpttConfig",
    "AutoModelForCausalLM": "modeling_tptt.TpttModel"
  },
  "base_model_name": "meta-llama/Llama-3.2-1B",
  "base_model_subfolder": null,
  "base_scale_attn": null,
  "bidirectional": false,
  "cross_gate": false,
  "head_dim": 64,
  "hidden_act": "silu",
  "hidden_size": 2048,
  "initializer_range": 0.02,
  "intermediate_size": 8192,
  "linear_precision": "bfloat16",
  "lora_config": {
    "alpha_pattern": {},
    "auto_mapping": null,
    "base_model_name_or_path": null,
    "bias": "none",
    "corda_config": null,
    "eva_config": null,
    "exclude_modules": null,
    "fan_in_fan_out": false,
    "inference_mode": false,
    "init_lora_weights": true,
    "layer_replication": null,
    "layers_pattern": null,
    "layers_to_transform": null,
    "loftq_config": {},
    "lora_alpha": 16,
    "lora_bias": false,
    "lora_dropout": 0.05,
    "megatron_config": null,
    "megatron_core": "megatron.core",
    "modules_to_save": null,
    "peft_type": "LORA",
    "r": 8,
    "rank_pattern": {},
    "revision": null,
    "target_modules": [
      "k_proj",
      "v_proj",
      "o_proj",
      "q_proj"
    ],
    "task_type": "CAUSAL_LM",
    "trainable_token_indices": null,
    "use_dora": false,
    "use_rslora": false
  },
  "mag_weight": 0.5,
  "max_chunk_size": 64,
  "max_position_embeddings": 131072,
  "max_self_attn_length": null,
  "mlp_bias": false,
  "model_type": "tptt",
  "num_attention_heads": 32,
  "num_hidden_layers": 16,
  "num_key_value_heads": 8,
  "operator_mode": "delta_product_r",
  "padding_side": "right",
  "pooling_config": null,
  "pretraining_tp": 1,
  "recurrent_config": {
    "gate_type": "k",
    "linear": true,
    "order": 2,
    "trick": "rotative"
  },
  "rms_norm_eps": 1e-05,
  "rope_scaling": {
    "factor": 32.0,
    "high_freq_factor": 4.0,
    "low_freq_factor": 1.0,
    "original_max_position_embeddings": 8192,
    "rope_type": "llama3"
  },
  "rope_theta": 500000.0,
  "target_modules_names": [
    "attn",
    "self_attn",
    "attention"
  ],
  "torch_dtype": "bfloat16",
  "transformers_version": "4.49.0",
  "use_cache": true,
  "use_linear_checkpoint": false,
  "vocab_size": 128256
}