config.json

{
  "_name_or_path": "phi2-wikitext2-0.1-param-ratio-0.95-top10_delta0.1",
  "architectures": [
    "ASVDPhiForCausalLM"
  ],
  "attention_dropout": 0.0,
  "auto_map": {
    "AutoConfig": "configuration_asvd_phi.ASVDPhiConfig",
    "AutoModelForCausalLM": "modeling_asvd_phi.ASVDPhiForCausalLM"
  },
  "bos_token_id": 50256,
  "embd_pdrop": 0.0,
  "eos_token_id": 50256,
  "hidden_act": "gelu_new",
  "hidden_size": 2560,
  "initializer_range": 0.02,
  "intermediate_size": 10240,
  "layer_norm_eps": 1e-05,
  "max_position_embeddings": 2048,
  "model_type": "phi",
  "num_attention_heads": 32,
  "num_hidden_layers": 32,
  "num_key_value_heads": 32,
  "partial_rotary_factor": 0.4,
  "qk_layernorm": false,
  "resid_pdrop": 0.1,
  "rope_scaling": null,
  "rope_theta": 10000.0,
  "tie_word_embeddings": false,
  "torch_dtype": "bfloat16",
  "transformers_version": "4.44.2",
  "truncation_ranks": {
    "model.layers.0.mlp.fc1": [
      1475,
      163
    ],
    "model.layers.0.mlp.fc2": [
      1290,
      143
    ],
    "model.layers.0.self_attn.dense": [
      922,
      102
    ],
    "model.layers.0.self_attn.k_proj": [
      922,
      102
    ],
    "model.layers.0.self_attn.q_proj": [
      807,
      89
    ],
    "model.layers.1.mlp.fc1": [
      1475,
      163
    ],
    "model.layers.1.self_attn.dense": [
      807,
      89
    ],
    "model.layers.1.self_attn.k_proj": [
      692,
      76
    ],
    "model.layers.1.self_attn.v_proj": [
      576,
      64
    ],
    "model.layers.10.self_attn.q_proj": [
      692,
      76
    ],
    "model.layers.11.self_attn.dense": [
      461,
      51
    ],
    "model.layers.11.self_attn.v_proj": [
      692,
      76
    ],
    "model.layers.12.self_attn.dense": [
      1037,
      115
    ],
    "model.layers.12.self_attn.q_proj": [
      807,
      89
    ],
    "model.layers.12.self_attn.v_proj": [
      922,
      102
    ],
    "model.layers.13.self_attn.dense": [
      807,
      89
    ],
    "model.layers.13.self_attn.q_proj": [
      576,
      64
    ],
    "model.layers.14.self_attn.dense": [
      1037,
      115
    ],
    "model.layers.14.self_attn.q_proj": [
      807,
      89
    ],
    "model.layers.15.self_attn.dense": [
      1037,
      115
    ],
    "model.layers.15.self_attn.q_proj": [
      807,
      89
    ],
    "model.layers.16.self_attn.dense": [
      922,
      102
    ],
    "model.layers.16.self_attn.k_proj": [
      922,
      102
    ],
    "model.layers.17.self_attn.dense": [
      1037,
      115
    ],
    "model.layers.17.self_attn.q_proj": [
      807,
      89
    ],
    "model.layers.18.self_attn.dense": [
      922,
      102
    ],
    "model.layers.19.self_attn.k_proj": [
      922,
      102
    ],
    "model.layers.19.self_attn.v_proj": [
      922,
      102
    ],
    "model.layers.2.mlp.fc1": [
      1475,
      163
    ],
    "model.layers.2.self_attn.dense": [
      807,
      89
    ],
    "model.layers.2.self_attn.k_proj": [
      1037,
      115
    ],
    "model.layers.2.self_attn.q_proj": [
      461,
      51
    ],
    "model.layers.20.self_attn.dense": [
      807,
      89
    ],
    "model.layers.20.self_attn.q_proj": [
      1037,
      115
    ],
    "model.layers.21.self_attn.dense": [
      692,
      76
    ],
    "model.layers.21.self_attn.q_proj": [
      1037,
      115
    ],
    "model.layers.22.self_attn.dense": [
      1037,
      115
    ],
    "model.layers.22.self_attn.q_proj": [
      807,
      89
    ],
    "model.layers.23.self_attn.dense": [
      922,
      102
    ],
    "model.layers.23.self_attn.q_proj": [
      922,
      102
    ],
    "model.layers.23.self_attn.v_proj": [
      692,
      76
    ],
    "model.layers.24.self_attn.dense": [
      461,
      51
    ],
    "model.layers.24.self_attn.q_proj": [
      922,
      102
    ],
    "model.layers.25.self_attn.k_proj": [
      576,
      64
    ],
    "model.layers.26.self_attn.k_proj": [
      922,
      102
    ],
    "model.layers.26.self_attn.q_proj": [
      1037,
      115
    ],
    "model.layers.27.self_attn.dense": [
      692,
      76
    ],
    "model.layers.27.self_attn.k_proj": [
      807,
      89
    ],
    "model.layers.28.self_attn.dense": [
      576,
      64
    ],
    "model.layers.28.self_attn.q_proj": [
      922,
      102
    ],
    "model.layers.29.self_attn.q_proj": [
      922,
      102
    ],
    "model.layers.29.self_attn.v_proj": [
      807,
      89
    ],
    "model.layers.3.self_attn.dense": [
      922,
      102
    ],
    "model.layers.3.self_attn.k_proj": [
      1037,
      115
    ],
    "model.layers.30.self_attn.dense": [
      922,
      102
    ],
    "model.layers.30.self_attn.q_proj": [
      807,
      89
    ],
    "model.layers.31.self_attn.dense": [
      1037,
      115
    ],
    "model.layers.31.self_attn.q_proj": [
      1037,
      115
    ],
    "model.layers.4.self_attn.dense": [
      807,
      89
    ],
    "model.layers.4.self_attn.k_proj": [
      1037,
      115
    ],
    "model.layers.4.self_attn.q_proj": [
      692,
      76
    ],
    "model.layers.5.self_attn.k_proj": [
      1037,
      115
    ],
    "model.layers.5.self_attn.q_proj": [
      807,
      89
    ],
    "model.layers.6.self_attn.dense": [
      807,
      89
    ],
    "model.layers.6.self_attn.q_proj": [
      922,
      102
    ],
    "model.layers.7.self_attn.q_proj": [
      692,
      76
    ],
    "model.layers.8.self_attn.dense": [
      1037,
      115
    ],
    "model.layers.8.self_attn.k_proj": [
      692,
      76
    ],
    "model.layers.8.self_attn.q_proj": [
      1037,
      115
    ],
    "model.layers.9.self_attn.dense": [
      692,
      76
    ],
    "model.layers.9.self_attn.q_proj": [
      576,
      64
    ]
  },
  "use_cache": true,
  "vocab_size": 51200
}