config.json

{
  "_name_or_path": "huggingface_repos/opt-125m-asvd90",
  "_remove_final_layer_norm": false,
  "activation_dropout": 0.0,
  "activation_function": "relu",
  "architectures": [
    "ASVDOPTForCausalLM"
  ],
  "attention_dropout": 0.0,
  "auto_map": {
    "AutoConfig": "configuration_asvd_opt.ASVDOPTConfig",
    "AutoModelForCausalLM": "modeling_asvd_opt.ASVDOPTForCausalLM"
  },
  "bos_token_id": 2,
  "do_layer_norm_before": true,
  "dropout": 0.1,
  "enable_bias": true,
  "eos_token_id": 2,
  "ffn_dim": 3072,
  "hidden_size": 768,
  "init_std": 0.02,
  "layer_norm_elementwise_affine": true,
  "layerdrop": 0.0,
  "max_position_embeddings": 2048,
  "model_type": "opt",
  "num_attention_heads": 12,
  "num_hidden_layers": 12,
  "pad_token_id": 1,
  "prefix": "</s>",
  "torch_dtype": "float16",
  "transformers_version": "4.35.2",
  "truncation_ranks": {
    "model.decoder.layers.0.self_attn.k_proj": 230,
    "model.decoder.layers.0.self_attn.out_proj": 345,
    "model.decoder.layers.0.self_attn.q_proj": 230,
    "model.decoder.layers.0.self_attn.v_proj": 153,
    "model.decoder.layers.1.fc2": 491,
    "model.decoder.layers.1.self_attn.k_proj": 115,
    "model.decoder.layers.1.self_attn.out_proj": 192,
    "model.decoder.layers.1.self_attn.q_proj": 115,
    "model.decoder.layers.10.self_attn.k_proj": 268,
    "model.decoder.layers.10.self_attn.q_proj": 230,
    "model.decoder.layers.11.self_attn.k_proj": 268,
    "model.decoder.layers.11.self_attn.q_proj": 307,
    "model.decoder.layers.2.fc1": 307,
    "model.decoder.layers.2.fc2": 368,
    "model.decoder.layers.2.self_attn.k_proj": 307,
    "model.decoder.layers.2.self_attn.out_proj": 268,
    "model.decoder.layers.2.self_attn.q_proj": 268,
    "model.decoder.layers.2.self_attn.v_proj": 268,
    "model.decoder.layers.3.fc2": 307,
    "model.decoder.layers.3.self_attn.k_proj": 153,
    "model.decoder.layers.3.self_attn.q_proj": 230,
    "model.decoder.layers.4.fc2": 430,
    "model.decoder.layers.4.self_attn.q_proj": 230,
    "model.decoder.layers.4.self_attn.v_proj": 307,
    "model.decoder.layers.5.fc2": 491,
    "model.decoder.layers.5.self_attn.out_proj": 345,
    "model.decoder.layers.5.self_attn.q_proj": 268,
    "model.decoder.layers.6.fc2": 430,
    "model.decoder.layers.6.self_attn.out_proj": 345,
    "model.decoder.layers.6.self_attn.q_proj": 268,
    "model.decoder.layers.7.fc2": 552,
    "model.decoder.layers.7.self_attn.k_proj": 345,
    "model.decoder.layers.7.self_attn.out_proj": 230,
    "model.decoder.layers.7.self_attn.q_proj": 345,
    "model.decoder.layers.8.fc1": 430,
    "model.decoder.layers.8.self_attn.k_proj": 268,
    "model.decoder.layers.8.self_attn.out_proj": 345,
    "model.decoder.layers.8.self_attn.q_proj": 307,
    "model.decoder.layers.9.fc2": 552,
    "model.decoder.layers.9.self_attn.k_proj": 345
  },
  "use_cache": true,
  "vocab_size": 50272,
  "word_embed_proj_dim": 768
}