trl-internal-testing/tiny-Gemma3ForConditionalGeneration

1{2  "architectures": [3    "Gemma3ForConditionalGeneration"4  ],5  "boi_token_index": 255999,6  "dtype": "bfloat16",7  "eoi_token_index": 256000,8  "eos_token_id": [9    1,10    10611  ],12  "image_token_index": 262144,13  "initializer_range": 0.02,14  "mm_tokens_per_image": 256,15  "model_type": "gemma3",16  "text_config": {17    "_sliding_window_pattern": 6,18    "attention_bias": false,19    "attention_dropout": 0.0,20    "attn_logit_softcapping": null,21    "final_logit_softcapping": null,22    "head_dim": 256,23    "hidden_activation": "gelu_pytorch_tanh",24    "hidden_size": 16,25    "initializer_range": 0.02,26    "intermediate_size": 32,27    "layer_types": [28      "sliding_attention",29      "full_attention"30    ],31    "max_position_embeddings": 131072,32    "model_type": "gemma3_text",33    "num_attention_heads": 4,34    "num_hidden_layers": 2,35    "num_key_value_heads": 2,36    "query_pre_attn_scalar": 256,37    "rms_norm_eps": 1e-06,38    "rope_local_base_freq": 10000.0,39    "rope_scaling": {40      "factor": 8.0,41      "rope_type": "linear"42    },43    "rope_theta": 1000000.0,44    "sliding_window": 1024,45    "use_cache": true,46    "vocab_size": 26220847  },48  "transformers_version": "4.56.2",49  "vision_config": {50    "attention_dropout": 0.0,51    "embed_dim": 64,52    "hidden_act": "gelu_pytorch_tanh",53    "hidden_size": 16,54    "image_size": 224,55    "intermediate_size": 32,56    "layer_norm_eps": 1e-06,57    "model_type": "siglip_vision_model",58    "num_attention_heads": 4,59    "num_channels": 3,60    "num_hidden_layers": 2,61    "num_key_value_heads": 2,62    "patch_size": 14,63    "vision_use_head": false64  }65}66