Functions — github.com/OpenMOSS/rope_pp

CodeHub MCPcopy Create free account

hub / github.com/OpenMOSS/rope_pp / functions

Functions305 in github.com/OpenMOSS/rope_pp

⨍ Functions 305◇ Types & classes 79 ↳ Endpoints 5

rope_pp/modeling_llama_alibi.py:1093

rope_pp/modeling_llama_alibi.py:1171

LlamaRMSNorm is equivalent to T5LayerNorm

rope_pp/modeling_llama_fope.py:56

(self, config: LlamaConfig, device=None)

rope_pp/modeling_llama_fope.py:159

( self, config: LlamaConfig, cache: int = None, dim: int = None, n_

rope_pp/modeling_llama_fope.py:202

(self, config, *args, **kwargs)

rope_pp/modeling_llama_fope.py:568

(self, config: LlamaConfig, layer_idx: int)

rope_pp/modeling_llama_fope.py:907

(self, config: LlamaConfig, layer_idx: int)

rope_pp/modeling_llama_fope.py:1071

(self, config: LlamaConfig)

rope_pp/modeling_llama_fope.py:1250

rope_pp/modeling_llama_fope.py:1519

rope_pp/modeling_llama_fope.py:1651

rope_pp/modeling_llama_fope.py:1747

rope_pp/modeling_llama_fope.py:1825

( self, vocab_size=32000, hidden_size=4096, intermediate_size=11008, n

rope_pp/configuration_llama.py:141

LlamaRMSNorm is equivalent to T5LayerNorm

rope_pp/modeling_llama_mask.py:55

(self, config: LlamaConfig, device=None)

rope_pp/modeling_llama_mask.py:78

(self, config: LlamaConfig, layer_idx: int)

rope_pp/modeling_llama_mask.py:187

(self, config: LlamaConfig, layer_idx: int)

rope_pp/modeling_llama_mask.py:331

(self, config: LlamaConfig)

rope_pp/modeling_llama_mask.py:510

rope_pp/modeling_llama_mask.py:778

rope_pp/modeling_llama_mask.py:910

rope_pp/modeling_llama_mask.py:1006

rope_pp/modeling_llama_mask.py:1084

(self, stop_words: List[str], tokenizer, batch_size: int)

rope_pp/mask_wrapper.py:27

(self, path: str, model_kwargs: dict = dict(), tokenizer_pa

rope_pp/mask_wrapper.py:100

(self, stop_words: List[str], tokenizer, batch_size: int)

rope_pp/rope_pp_wrapper.py:27

(self, path: str, model_kwargs: dict = dict(), tokenizer_pa

rope_pp/rope_pp_wrapper.py:100

LlamaRMSNorm is equivalent to T5LayerNorm

rope_pp/modeling_llama_rope_pp.py:55

(self, config: LlamaConfig, device=None)

rope_pp/modeling_llama_rope_pp.py:78

(self, config: LlamaConfig, layer_idx: int)

rope_pp/modeling_llama_rope_pp.py:195

(self, config: LlamaConfig, layer_idx: int)

rope_pp/modeling_llama_rope_pp.py:332

(self, config: LlamaConfig)

rope_pp/modeling_llama_rope_pp.py:511

rope_pp/modeling_llama_rope_pp.py:779

rope_pp/modeling_llama_rope_pp.py:911

rope_pp/modeling_llama_rope_pp.py:1007

rope_pp/modeling_llama_rope_pp.py:1085

(self, stop_words: List[str], tokenizer, batch_size: int)

rope_pp/alibi_wrapper.py:27

(self, path: str, model_kwargs: dict = dict(), tokenizer_pa

rope_pp/alibi_wrapper.py:100

(self, stop_words: List[str], tokenizer, batch_size: int)

rope_pp/fope_wrapper.py:27

(self, path: str, model_kwargs: dict = dict(), tokenizer_pa

rope_pp/fope_wrapper.py:100

utils/dataset_utils.py:62

utils/dataset_utils.py:149

utils/dataset_utils.py:202

rope_pp/modeling_llama_fope.py:847

Function_get_possible_max_seq_len

(max_seq_len, path)

rope_pp/pythia_wrapper.py:50

Function_get_possible_max_seq_len

(max_seq_len, path)

rope_pp/mask_wrapper.py:50

Function_get_possible_max_seq_len

(max_seq_len, path)

rope_pp/rope_pp_wrapper.py:50

Function_get_possible_max_seq_len

(max_seq_len, path)

rope_pp/alibi_wrapper.py:50

Function_get_possible_max_seq_len

(max_seq_len, path)

rope_pp/fope_wrapper.py:50

Method_init_weights

rope_pp/modeling_llama_pythia.py:387

Method_init_weights

rope_pp/modeling_llama_alibi.py:504

Method_init_weights

rope_pp/modeling_llama_fope.py:1159

Method_init_weights

rope_pp/modeling_llama_mask.py:419

Method_init_weights

rope_pp/modeling_llama_rope_pp.py:420

Method_save_checkpoint

(self, model, trial)

utils/trainer_utils.py:16

Functionapply_rotary_pos_emb

Applies Rotary Position Embedding to the query and key tensors. Args: q (`torch.Tensor`): The query tensor. k (`torch.Tensor`): T

rope_pp/modeling_llama_alibi.py:135

Methodapply_rotary_pos_emb

(self, pos_sin, pos_cos, t, inverse=False)

rope_pp/modeling_llama_fope.py:674

Functioneager_attention_forward

( module: nn.Module, query: torch.Tensor, key: torch.Tensor, value: torch.Tensor, attentio

rope_pp/modeling_llama_pythia.py:196

Functioneager_attention_forward

( module: nn.Module, query: torch.Tensor, key: torch.Tensor, value: torch.Tensor, attentio

rope_pp/modeling_llama_alibi.py:190

Functioneager_attention_forward

( module: nn.Module, query: torch.Tensor, key: torch.Tensor, value: torch.Tensor, attentio

rope_pp/modeling_llama_fope.py:878

Functioneager_attention_forward

( module: nn.Module, query: torch.Tensor, key: torch.Tensor, value: torch.Tensor, attentio

rope_pp/modeling_llama_mask.py:158

Functioneager_attention_forward

( module: nn.Module, query: torch.Tensor, key: torch.Tensor, value: torch.Tensor, attentio

rope_pp/modeling_llama_rope_pp.py:166

Methodextra_repr

rope_pp/modeling_llama_pythia.py:85

Methodextra_repr

rope_pp/modeling_llama_alibi.py:87

Methodextra_repr

rope_pp/modeling_llama_fope.py:71

Methodextra_repr

rope_pp/modeling_llama_mask.py:70

Methodextra_repr

rope_pp/modeling_llama_rope_pp.py:70

Methodfind_correction_dim

(num_rotations, dim, base=self.config.rope_theta, orig_max_position_embeddings=self.config.len_extra_orig_leng

rope_pp/modeling_llama_fope.py:302

Methodfind_correction_range

(low_rot, high_rot, dim, base=self.config.rope_theta, orig_max_position_embeddings=self.config.len_extra_orig_

rope_pp/modeling_llama_fope.py:305

(self, hidden_states)

rope_pp/modeling_llama_pythia.py:78

(self, x, position_ids)

rope_pp/modeling_llama_pythia.py:112

rope_pp/modeling_llama_pythia.py:179

( self, hidden_states: torch.Tensor, position_embeddings: Tuple[torch.Tensor, torch.Te

rope_pp/modeling_llama_pythia.py:248

( self, hidden_states: torch.Tensor, attention_mask: Optional[torch.Tensor] = None,

rope_pp/modeling_llama_pythia.py:309

( self, input_ids: Optional[torch.LongTensor] = None, attention_mask: Optional[torch.T

rope_pp/modeling_llama_pythia.py:502

r""" labels (`torch.LongTensor` of shape `(batch_size, sequence_length)`, *optional*): Labels for computing the masked lan

rope_pp/modeling_llama_pythia.py:779

r""" labels (`torch.LongTensor` of shape `(batch_size,)`, *optional*): Labels for computing the sequence classification/regression

rope_pp/modeling_llama_pythia.py:897

r""" start_positions (`torch.LongTensor` of shape `(batch_size,)`, *optional*): Labels for position (index) of the start of the la

rope_pp/modeling_llama_pythia.py:992

r""" labels (`torch.LongTensor` of shape `(batch_size,)`, *optional*): Labels for computing the sequence classification/regression

rope_pp/modeling_llama_pythia.py:1083

(self, hidden_states)

rope_pp/modeling_llama_alibi.py:80

(self, x, position_ids)

rope_pp/modeling_llama_alibi.py:114

rope_pp/modeling_llama_alibi.py:173

( self, hidden_states: torch.Tensor, position_embeddings: Optional[Tuple[torch.Tensor,

rope_pp/modeling_llama_alibi.py:353

( self, hidden_states: torch.Tensor, attention_mask: Optional[torch.Tensor] = None,

rope_pp/modeling_llama_alibi.py:426

( self, input_ids: Optional[torch.LongTensor] = None, attention_mask: Optional[torch.T

rope_pp/modeling_llama_alibi.py:619

r""" labels (`torch.LongTensor` of shape `(batch_size, sequence_length)`, *optional*): Labels for computing the masked lan

rope_pp/modeling_llama_alibi.py:896

r""" labels (`torch.LongTensor` of shape `(batch_size,)`, *optional*): Labels for computing the sequence classification/regression

rope_pp/modeling_llama_alibi.py:1014

r""" start_positions (`torch.LongTensor` of shape `(batch_size,)`, *optional*): Labels for position (index) of the start of the la

rope_pp/modeling_llama_alibi.py:1109

r""" labels (`torch.LongTensor` of shape `(batch_size,)`, *optional*): Labels for computing the sequence classification/regression

rope_pp/modeling_llama_alibi.py:1200

(self, hidden_states)

rope_pp/modeling_llama_fope.py:64

(self, x, position_ids)

rope_pp/modeling_llama_fope.py:178

( self, x: torch.Tensor, all_len: int, layer_idx: Optional[int] = None,

rope_pp/modeling_llama_fope.py:528

rope_pp/modeling_llama_fope.py:861

( self, hidden_states: torch.Tensor, position_embeddings: Tuple[torch.Tensor, torch.Te

rope_pp/modeling_llama_fope.py:1009

( self, hidden_states: torch.Tensor, attention_mask: Optional[torch.Tensor] = None,

rope_pp/modeling_llama_fope.py:1081

( self, input_ids: Optional[torch.LongTensor] = None, attention_mask: Optional[torch.T

rope_pp/modeling_llama_fope.py:1274

r""" labels (`torch.LongTensor` of shape `(batch_size, sequence_length)`, *optional*): Labels for computing the masked lan

rope_pp/modeling_llama_fope.py:1551

r""" labels (`torch.LongTensor` of shape `(batch_size,)`, *optional*): Labels for computing the sequence classification/regression

rope_pp/modeling_llama_fope.py:1668

r""" start_positions (`torch.LongTensor` of shape `(batch_size,)`, *optional*): Labels for position (index) of the start of the la

rope_pp/modeling_llama_fope.py:1763

r""" labels (`torch.LongTensor` of shape `(batch_size,)`, *optional*): Labels for computing the sequence classification/regression

rope_pp/modeling_llama_fope.py:1854

← previous next →101–200 of 305, ranked by callers