config.json

{
  "architectures": [
    "HypeNetForCausalLM"
  ],
  "attention_bias": false,
  "attention_dropout": 0.0,
  "attn_logits_scaling": "hype 2000",
  "attn_use_output_gate": true,
  "attn_use_rope": false,
  "auto_map": {
    "AutoConfig": "configuration_hypenet.HypeNetConfig",
    "AutoModelForCausalLM": "modeling_hypenet.HypeNetForCausalLM"
  },
  "bos_token_id": 151643,
  "eos_token_id": 151645,
  "expand_kv_proj": true,
  "fused_ce_loss": true,
  "gdn_activation": null,
  "gdn_attn_mode": "chunk",
  "gdn_expand_v": 1,
  "gdn_fuse_cross_entropy": false,
  "gdn_nh": 32,
  "gdn_nkv": 8,
  "gdn_use_gate": false,
  "gdn_use_qk_norm": false,
  "gdn_use_rope": false,
  "gdn_use_short_conv": false,
  "head_dim": 128,
  "hidden_act": "silu",
  "hidden_size": 2560,
  "initializer_range": 0.02,
  "intermediate_size": 9728,
  "kda_head_dim": 128,
  "kda_num_heads": 32,
  "kda_use_conv": false,
  "kda_use_qk_norm": true,
  "kda_use_rope": false,
  "lightning_conv_size": 4,
  "lightning_head_dim": 128,
  "lightning_nh": 32,
  "lightning_nkv": 32,
  "lightning_rope_scaling": null,
  "lightning_scale": "1/sqrt(d)",
  "lightning_use_output_gate": true,
  "lightning_use_output_norm": true,
  "lightning_use_qk_norm": true,
  "lightning_use_rope": true,
  "lightning_use_short_conv": false,
  "loss_fn": "kl_div",
  "max_position_embeddings": 1048576,
  "mixer_types": [
    "attn",
    "attn",
    "lightning-attn",
    "lightning-attn",
    "lightning-attn",
    "lightning-attn",
    "lightning-attn",
    "attn",
    "lightning-attn",
    "lightning-attn",
    "lightning-attn",
    "lightning-attn",
    "lightning-attn",
    "lightning-attn",
    "attn",
    "attn",
    "lightning-attn",
    "lightning-attn",
    "lightning-attn",
    "lightning-attn",
    "lightning-attn",
    "lightning-attn",
    "attn",
    "lightning-attn",
    "attn",
    "lightning-attn",
    "lightning-attn",
    "lightning-attn",
    "lightning-attn",
    "lightning-attn",
    "lightning-attn",
    "attn",
    "lightning-attn",
    "attn",
    "lightning-attn",
    "lightning-attn"
  ],
  "model_type": "hypenet",
  "num_attention_heads": 32,
  "num_hidden_layers": 36,
  "num_key_value_heads": 8,
  "rand_init": false,
  "rms_norm_eps": 1e-06,
  "rope_parameters": {},
  "rope_scaling": null,
  "rope_theta": 1000000,
  "shift_labels": true,
  "sliding_window": null,
  "tie_word_embeddings": true,
  "torch_dtype": "bfloat16",
  "transformers_version": "4.55.0",
  "use_cache": true,
  "use_sliding_window": false,
  "vocab_size": 151936
}