hf-transformers-bot's picture
Update tiny models for MuseGlimmerTextModel
99fb634 verified
Raw
History Blame Contribute Delete
991 Bytes
{
"architectures": [
"MuseGlimmerTextModel"
],
"attention_bias": false,
"attention_dropout": 0.0,
"bos_token_id": 200000,
"dtype": "float32",
"eos_token_id": 200001,
"final_logit_softcapping": 20.0,
"head_dim": 16,
"hidden_activation": "silu",
"hidden_size": 32,
"initializer_range": 0.02,
"intermediate_size": 37,
"layer_rope_theta": [
10000.0,
0
],
"layer_types": [
"full_attention",
"sliding_attention"
],
"max_position_embeddings": 512,
"model_type": "muse_glimmer_text",
"num_attention_heads": 2,
"num_hidden_layers": 2,
"num_key_value_heads": 2,
"output_multiplier": 0.19611613513818404,
"pad_token_id": 200018,
"post_norm_eps": 1e-08,
"qk_scale_factor": 3.87,
"rms_norm_eps": 1e-05,
"rope_parameters": {
"rope_theta": 10000.0,
"rope_type": "default"
},
"sliding_window": 2048,
"tie_word_embeddings": false,
"transformers_version": "5.16.0.dev0",
"use_cache": true,
"vocab_size": 202048
}