config.json

{
  "_name_or_path": "/home/models/helix-mRNA/helix_mrna_8l_256h_5.1883Mio_12288_2024-11-30_18-51-46/checkpoint-23000",
  "architectures": [
    "HelixmRNAForCausalLM"
  ],
  "attention_dropout": 0.0,
  "attn_implementation": "flash_attention_2",
  "bos_token_id": 0,
  "chunk_size": 256,
  "conv_kernel": 4,
  "eos_token_id": 2,
  "expand": 2,
  "head_dim": 32,
  "hidden_act": "silu",
  "hidden_size": 256,
  "initializer_range": 0.1,
  "intermediate_size": 512,
  "layer_norm_epsilon": 1e-05,
  "layers_block_type_string": "M+M*M+M+",
  "model_type": "mamba2",
  "n_groups": 1,
  "num_attention_heads": 32,
  "num_heads": 16,
  "num_hidden_layers": 8,
  "num_key_value_heads": 8,
  "pad_token_id": 1,
  "rescale_prenorm_residual": false,
  "residual_in_fp32": true,
  "rms_norm": true,
  "state_size": 128,
  "tie_word_embeddings": false,
  "time_step_floor": 0.0001,
  "time_step_limit": [
    0.0,
    Infinity
  ],
  "time_step_max": 0.1,
  "time_step_min": 0.001,
  "time_step_rank": 16,
  "torch_dtype": "float32",
  "transformers_version": "4.45.1",
  "use_bias": false,
  "use_cache": false,
  "use_conv_bias": true,
  "vocab_size": 14
}