generation / config.json
itstobinwosu's picture
Add predict.py, README.md, config.json, training_args.json, model.safetensors
0bcf67b verified
Raw History Blame Contribute Delete
501 Bytes
{
"architectures": [
"CustomResearchModel"
],
"architecture": "deit",
"model_type": "deit",
"hidden_size": 384,
"num_hidden_layers": 4,
"num_attention_heads": 2,
"intermediate_size": 768,
"hidden_act": "swish",
"max_position_embeddings": 128,
"layer_norm_eps": 1e-12,
"checkpoint_status": "initialization-only",
"notes": "Untrained checkpoint for smoke tests; no benchmark claim.",
"use_cache": true,
"torch_dtype": "float32",
"vocab_size": 2048
}