vlite3.5-nano-RURU-63M / config.json
aixk's picture
Training Complete
2e9b4fc verified
raw
history blame contribute delete
329 Bytes
{
"architectures": [
"VLite3_5ForCausalLM"
],
"dtype": "float32",
"hidden_size": 512,
"intermediate_size": 1024,
"max_position_embeddings": 2048,
"model_type": "vlite3.5",
"num_hidden_layers": 11,
"tie_word_embeddings": true,
"transformers_version": "5.8.1",
"use_cache": false,
"vocab_size": 40963
}