hf-transformers-bot's picture
Update tiny models for Cohere2VisionModel
cb4b336 verified
Raw
History Blame Contribute Delete
1.44 kB
{
"alignment_intermediate_size": 32,
"architectures": [
"Cohere2VisionModel"
],
"downsample_factor": 2,
"dtype": "float32",
"image_token_id": 2,
"model_type": "cohere2_vision",
"text_config": {
"attention_bias": false,
"attention_dropout": 0.0,
"bos_token_id": 5,
"eos_token_id": 255001,
"head_dim": 32,
"hidden_act": "silu",
"hidden_size": 128,
"initializer_range": 0.02,
"intermediate_size": 37,
"layer_norm_eps": 1e-05,
"layer_types": [
"sliding_attention",
"sliding_attention"
],
"logit_scale": 0.0625,
"max_position_embeddings": 512,
"model_type": "cohere2",
"num_attention_heads": 4,
"num_hidden_layers": 2,
"num_key_value_heads": 4,
"output_channels": 64,
"pad_token_id": 0,
"rope_parameters": {
"rope_theta": 10000.0,
"rope_type": "default"
},
"sliding_window": 4096,
"tie_word_embeddings": true,
"use_cache": true,
"vocab_size": 255037
},
"tie_word_embeddings": true,
"transformers_version": "5.16.0.dev0",
"vision_config": {
"attention_dropout": 0.0,
"hidden_act": "gelu_pytorch_tanh",
"hidden_size": 32,
"image_size": 64,
"intermediate_size": 128,
"layer_norm_eps": 1e-06,
"model_type": "siglip_vision_model",
"num_attention_heads": 4,
"num_channels": 3,
"num_hidden_layers": 2,
"patch_size": 8,
"vision_use_head": false
}
}