hf-transformers-bot's picture
Update tiny models for Gemma3ForSequenceClassification
9dea17a verified
Raw
History Blame Contribute Delete
1.76 kB
{
"architectures": [
"Gemma3ForSequenceClassification"
],
"boi_token_index": 255999,
"dtype": "float32",
"eoi_token_index": 256000,
"image_token_index": 4,
"initializer_range": 0.02,
"mm_tokens_per_image": 2,
"model_type": "gemma3",
"pad_token_id": 0,
"text_config": {
"_sliding_window_pattern": 6,
"attention_bias": false,
"attention_dropout": 0.0,
"attn_logit_softcapping": null,
"bos_token_id": 2,
"eos_token_id": 1,
"final_logit_softcapping": null,
"head_dim": 16,
"hidden_activation": "gelu_pytorch_tanh",
"hidden_size": 32,
"initializer_range": 0.02,
"intermediate_size": 32,
"layer_types": [
"full_attention",
"sliding_attention"
],
"max_position_embeddings": 512,
"model_type": "gemma3_text",
"num_attention_heads": 2,
"num_hidden_layers": 2,
"num_key_value_heads": 1,
"pad_token_id": 0,
"query_pre_attn_scalar": 256,
"rms_norm_eps": 1e-06,
"rope_parameters": {
"full_attention": {
"rope_theta": 1000000.0,
"rope_type": "default"
},
"sliding_attention": {
"rope_theta": 10000.0,
"rope_type": "default"
}
},
"sliding_window": 4096,
"tie_word_embeddings": false,
"use_bidirectional_attention": false,
"use_cache": true,
"vocab_size": 262145
},
"tie_word_embeddings": false,
"transformers_version": "5.16.0.dev0",
"vision_config": {
"attention_dropout": 0.0,
"hidden_act": "gelu",
"hidden_size": 32,
"image_size": 20,
"intermediate_size": 32,
"layer_norm_eps": 1e-06,
"model_type": "siglip_vision_model",
"num_attention_heads": 2,
"num_channels": 3,
"num_hidden_layers": 2,
"patch_size": 5
}
}