Crystalcareai
commited on
Update modeling_quiet.py
Browse files- modeling_quiet.py +2 -2
modeling_quiet.py
CHANGED
@@ -1175,8 +1175,8 @@ class QuietForCausalLM(QuietPreTrainedModel):
|
|
1175 |
self.vocab_size = config.vocab_size
|
1176 |
self.lm_head = nn.Linear(config.hidden_size, config.vocab_size, bias=False)
|
1177 |
# self.router_aux_loss_coef = config.router_aux_loss_coef
|
1178 |
-
self.num_experts = config.num_experts
|
1179 |
-
self.num_experts_per_tok = config.num_experts_per_tok
|
1180 |
self.max_thoughts = config.max_thoughts
|
1181 |
self.merged_lm_and_talk_heads = config.merged_lm_and_talk_heads
|
1182 |
self.use_concat_talk_head = config.use_concat_talk_head
|
|
|
1175 |
self.vocab_size = config.vocab_size
|
1176 |
self.lm_head = nn.Linear(config.hidden_size, config.vocab_size, bias=False)
|
1177 |
# self.router_aux_loss_coef = config.router_aux_loss_coef
|
1178 |
+
# self.num_experts = config.num_experts
|
1179 |
+
# self.num_experts_per_tok = config.num_experts_per_tok
|
1180 |
self.max_thoughts = config.max_thoughts
|
1181 |
self.merged_lm_and_talk_heads = config.merged_lm_and_talk_heads
|
1182 |
self.use_concat_talk_head = config.use_concat_talk_head
|