(self, config)
| 347 | class MossForCausalLM(Module): |
| 348 | |
| 349 | def __init__(self, config): |
| 350 | super(MossForCausalLM, self).__init__() |
| 351 | self.config = config |
| 352 | self.transformer = MossModel(config) |
| 353 | self.lm_head = nn.Linear(config.n_embd, config.vocab_size) |
| 354 | |
| 355 | # Initialize weights and apply final processing |
| 356 | self.apply(partial(_init_weights, config)) |
| 357 | |
| 358 | def execute( |
| 359 | self, |