config.json

{
  "vocab_size": 35303,
  "hidden_size": 1792,
  "intermediate_size": 3584,
  "num_hidden_layers": 18,
  "num_attention_heads": 14,
  "num_key_value_heads": 14,
  "head_dim": 128,
  "sliding_window": 4096,
  "hidden_act": "silu",
  "attention_dropout": 0.0,
  "rms_norm_eps": 1e-05,
  "initializer_range": 0.02,
  "is_decoder": false,
  "add_cross_attention": false,
  "use_cache": false,
  "attention_bias": true,
  "num_local_experts": 1,
  "num_experts_per_tok": 1,
  "router_aux_loss_coef": 0.0,
  "output_router_logits": false,
  "layer_types": [
    "sliding_attention",
    "full_attention",
    "sliding_attention",
    "full_attention",
    "sliding_attention",
    "full_attention",
    "sliding_attention",
    "full_attention",
    "sliding_attention",
    "full_attention",
    "sliding_attention",
    "full_attention",
    "sliding_attention",
    "full_attention",
    "sliding_attention",
    "full_attention",
    "sliding_attention",
    "full_attention"
  ],
  "max_position_embeddings": 8192,
  "bos_token_id": 1,
  "eos_token_id": 2,
  "pad_token_id": 3,
  "return_dict": true,
  "output_hidden_states": false,
  "dtype": "bfloat16",
  "tie_word_embeddings": false,
  "chunk_size_feed_forward": 0,
  "is_encoder_decoder": false,
  "cross_attention_hidden_size": null,
  "tie_encoder_decoder": false,
  "architectures": [
    "EvaGptForCausalLM"
  ],
  "finetuning_task": null,
  "id2label": {
    "0": "LABEL_0",
    "1": "LABEL_1"
  },
  "label2id": {
    "LABEL_0": 0,
    "LABEL_1": 1
  },
  "task_specific_params": null,
  "problem_type": null,
  "tokenizer_class": null,
  "prefix": null,
  "sep_token_id": null,
  "decoder_start_token_id": null,
  "_name_or_path": "./training_outputs/EvaGPT-German-0.7B_2026_7_2026-01-08_17-33-14_2026-01-08_19-10-49",
  "transformers_version": "5.0.0.dev0",
  "model_type": "eva_gpt",
  "output_attentions": false,
  "rope_scaling": {
    "type": "yarn",
    "factor": 2.0,
    "original_max_position_embeddings": 4096
  }
}