File size: 380 Bytes
e4af6d2
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
{
    "model_type": "ternarylm",
    "vocab_size": 30522,
    "hidden_size": 768,
    "num_hidden_layers": 12,
    "num_attention_heads": 12,
    "max_position_embeddings": 512,
    "quantization": "native ternary {-1,0,+1}",
    "training_dataset": "roneneldan/TinyStories",
    "epochs": 15,
    "optimizer": "AdamW",
    "learning_rate": 0.0003,
    "scheduler": "OneCycleLR"
}