| from transformers import PretrainedConfig | |
| class TinyIndianLMConfig(PretrainedConfig): | |
| model_type = "tinyindianlm" | |
| def __init__( | |
| self, | |
| vocab_size=16000, | |
| d_model=512, | |
| n_layers=4, | |
| n_heads=8, | |
| n_kv_heads=4, | |
| max_seq_len=512, | |
| d_ffn=1408, | |
| dropout=0.0, | |
| rope_theta=10000.0, | |
| pad_token_id=0, | |
| unk_token_id=1, | |
| bos_token_id=2, | |
| eos_token_id=3, | |
| **kwargs, | |
| ): | |
| self.vocab_size = vocab_size | |
| self.d_model = d_model | |
| self.n_layers = n_layers | |
| self.n_heads = n_heads | |
| self.n_kv_heads = n_kv_heads | |
| self.max_seq_len = max_seq_len | |
| self.d_ffn = d_ffn | |
| self.dropout = dropout | |
| self.rope_theta = rope_theta | |
| super().__init__( | |
| pad_token_id=pad_token_id, | |
| unk_token_id=unk_token_id, | |
| bos_token_id=bos_token_id, | |
| eos_token_id=eos_token_id, | |
| **kwargs, | |
| ) | |