Code
Hub
Workspaces
Following
Trending
Connect
MCP
copy
Create free account
hub
/
github.com/Tele-AI/Telechat
/ functions
Functions
680 in github.com/Tele-AI/Telechat
⨍
Functions
680
◇
Types & classes
127
↳
Endpoints
2
Method
end
Flushes any remaining cache and prints a newline to stdout.
models/12B-V2_8bit/generation_utils.py:123
Method
end
Flushes any remaining cache and prints a newline to stdout.
models/12B_8bit/generation_utils.py:123
Method
end
Flushes any remaining cache and prints a newline to stdout.
models/12B-V2/generation_utils.py:123
Method
forward
(self, input)
deepspeed-telechat/utils/module/lora.py:75
Method
forward
(self, x)
service/vllm_inf/telechat_12B.py:91
Method
forward
( self, positions: torch.Tensor, hidden_states: torch.Tensor, kv_cache: torch.
service/vllm_inf/telechat_12B.py:192
Method
forward
( self, positions: torch.Tensor, hidden_states: torch.Tensor, kv_cache: torch.
service/vllm_inf/telechat_12B.py:259
Method
forward
( self, input_ids: Optional[torch.Tensor], positions: torch.Tensor, kv_caches:
service/vllm_inf/telechat_12B.py:313
Method
forward
( self, input_ids: torch.Tensor, positions: torch.Tensor, kv_caches: List[torc
service/vllm_inf/telechat_12B.py:398
Method
forward
(self, x, seq_dim=0, seq_len=None)
models/7B_8bit/modeling_telechat.py:106
Method
forward
(self, hidden_states)
models/7B_8bit/modeling_telechat.py:148
Method
forward
Implements the multihead softmax attention. Arguments --------- q, k, v: The tensor containing the query, key, and valu
models/7B_8bit/modeling_telechat.py:178
Method
forward
(ctx, input: torch.Tensor)
models/7B_8bit/modeling_telechat.py:299
Method
forward
(self, x: torch.Tensor)
models/7B_8bit/modeling_telechat.py:322
Method
forward
( self, hidden_states: torch.Tensor, residual: torch.Tensor,
models/7B_8bit/modeling_telechat.py:399
Method
forward
(self, hidden_states: torch.Tensor, residual: torch.Tensor)
models/7B_8bit/modeling_telechat.py:502
Method
forward
( self, hidden_states: torch.Tensor, attention_mask: torch.Tensor,
models/7B_8bit/modeling_telechat.py:524
Method
forward
( self, input_ids: Optional[torch.LongTensor] = None, past_key_values:
models/7B_8bit/modeling_telechat.py:636
Method
forward
( self, input_ids: Optional[torch.LongTensor] = None, past_key_values:
models/7B_8bit/modeling_telechat.py:781
Method
forward
(self, x, seq_dim=0, seq_len=None)
models/12B_4bit/modeling_telechat.py:106
Method
forward
(self, hidden_states)
models/12B_4bit/modeling_telechat.py:148
Method
forward
Implements the multihead softmax attention. Arguments --------- q, k, v: The tensor containing the query, key, and value.
models/12B_4bit/modeling_telechat.py:178
Method
forward
(ctx, input: torch.Tensor)
models/12B_4bit/modeling_telechat.py:299
Method
forward
(self, x: torch.Tensor)
models/12B_4bit/modeling_telechat.py:322
Method
forward
( self, hidden_states: torch.Tensor, residual: torch.Tensor, a
models/12B_4bit/modeling_telechat.py:399
Method
forward
(self, hidden_states: torch.Tensor, residual: torch.Tensor)
models/12B_4bit/modeling_telechat.py:502
Method
forward
( self, hidden_states: torch.Tensor, attention_mask: torch.Tensor,
models/12B_4bit/modeling_telechat.py:524
Method
forward
( self, input_ids: Optional[torch.LongTensor] = None, past_key_values: Opt
models/12B_4bit/modeling_telechat.py:636
Method
forward
( self, input_ids: Optional[torch.LongTensor] = None, past_key_values: Opt
models/12B_4bit/modeling_telechat.py:781
Method
forward
(self, x, seq_dim=0, seq_len=None)
models/7B/modeling_telechat.py:106
Method
forward
(self, hidden_states)
models/7B/modeling_telechat.py:148
Method
forward
Implements the multihead softmax attention. Arguments --------- q, k, v: The tensor containing the query, key, and valu
models/7B/modeling_telechat.py:178
Method
forward
(ctx, input: torch.Tensor)
models/7B/modeling_telechat.py:299
Method
forward
(self, x: torch.Tensor)
models/7B/modeling_telechat.py:322
Method
forward
( self, hidden_states: torch.Tensor, residual: torch.Tensor,
models/7B/modeling_telechat.py:399
Method
forward
(self, hidden_states: torch.Tensor, residual: torch.Tensor)
models/7B/modeling_telechat.py:502
Method
forward
( self, hidden_states: torch.Tensor, attention_mask: torch.Tensor,
models/7B/modeling_telechat.py:524
Method
forward
( self, input_ids: Optional[torch.LongTensor] = None, past_key_values:
models/7B/modeling_telechat.py:636
Method
forward
( self, input_ids: Optional[torch.LongTensor] = None, past_key_values:
models/7B/modeling_telechat.py:781
Method
forward
(self, x, seq_dim=0, seq_len=None)
models/12B/modeling_telechat.py:106
Method
forward
(self, hidden_states)
models/12B/modeling_telechat.py:148
Method
forward
Implements the multihead softmax attention. Arguments --------- q, k, v: The tensor containing the query, key, and valu
models/12B/modeling_telechat.py:178
Method
forward
(ctx, input: torch.Tensor)
models/12B/modeling_telechat.py:299
Method
forward
(self, x: torch.Tensor)
models/12B/modeling_telechat.py:322
Method
forward
( self, hidden_states: torch.Tensor, residual: torch.Tensor,
models/12B/modeling_telechat.py:399
Method
forward
(self, hidden_states: torch.Tensor, residual: torch.Tensor)
models/12B/modeling_telechat.py:502
Method
forward
( self, hidden_states: torch.Tensor, attention_mask: torch.Tensor,
models/12B/modeling_telechat.py:524
Method
forward
( self, input_ids: Optional[torch.LongTensor] = None, past_key_values:
models/12B/modeling_telechat.py:636
Method
forward
( self, input_ids: Optional[torch.LongTensor] = None, past_key_values:
models/12B/modeling_telechat.py:781
Method
forward
(self, x, seq_dim=0, seq_len=None)
models/7B_4bit/modeling_telechat.py:106
Method
forward
(self, hidden_states)
models/7B_4bit/modeling_telechat.py:148
Method
forward
Implements the multihead softmax attention. Arguments --------- q, k, v: The tensor containing the query, key, and valu
models/7B_4bit/modeling_telechat.py:178
Method
forward
(ctx, input: torch.Tensor)
models/7B_4bit/modeling_telechat.py:299
Method
forward
(self, x: torch.Tensor)
models/7B_4bit/modeling_telechat.py:322
Method
forward
( self, hidden_states: torch.Tensor, residual: torch.Tensor,
models/7B_4bit/modeling_telechat.py:399
Method
forward
(self, hidden_states: torch.Tensor, residual: torch.Tensor)
models/7B_4bit/modeling_telechat.py:502
Method
forward
( self, hidden_states: torch.Tensor, attention_mask: torch.Tensor,
models/7B_4bit/modeling_telechat.py:524
Method
forward
( self, input_ids: Optional[torch.LongTensor] = None, past_key_values:
models/7B_4bit/modeling_telechat.py:636
Method
forward
( self, input_ids: Optional[torch.LongTensor] = None, past_key_values:
models/7B_4bit/modeling_telechat.py:781
Method
forward
(self, x, seq_dim=0, seq_len=None)
models/12B-V2_8bit/modeling_telechat.py:106
Method
forward
(self, hidden_states)
models/12B-V2_8bit/modeling_telechat.py:148
Method
forward
Implements the multihead softmax attention. Arguments --------- q, k, v: The tensor containing the query, key, and value.
models/12B-V2_8bit/modeling_telechat.py:178
Method
forward
(ctx, input: torch.Tensor)
models/12B-V2_8bit/modeling_telechat.py:299
Method
forward
(self, x: torch.Tensor)
models/12B-V2_8bit/modeling_telechat.py:322
Method
forward
( self, hidden_states: torch.Tensor, residual: torch.Tensor, a
models/12B-V2_8bit/modeling_telechat.py:399
Method
forward
(self, hidden_states: torch.Tensor, residual: torch.Tensor)
models/12B-V2_8bit/modeling_telechat.py:502
Method
forward
( self, hidden_states: torch.Tensor, attention_mask: torch.Tensor,
models/12B-V2_8bit/modeling_telechat.py:524
Method
forward
( self, input_ids: Optional[torch.LongTensor] = None, past_key_values: Opt
models/12B-V2_8bit/modeling_telechat.py:636
Method
forward
( self, input_ids: Optional[torch.LongTensor] = None, past_key_values: Opt
models/12B-V2_8bit/modeling_telechat.py:781
Method
forward
(self, x, seq_dim=0, seq_len=None)
models/12B_8bit/modeling_telechat.py:106
Method
forward
(self, hidden_states)
models/12B_8bit/modeling_telechat.py:148
Method
forward
Implements the multihead softmax attention. Arguments --------- q, k, v: The tensor containing the query, key, and value.
models/12B_8bit/modeling_telechat.py:178
Method
forward
(ctx, input: torch.Tensor)
models/12B_8bit/modeling_telechat.py:299
Method
forward
(self, x: torch.Tensor)
models/12B_8bit/modeling_telechat.py:322
Method
forward
( self, hidden_states: torch.Tensor, residual: torch.Tensor, a
models/12B_8bit/modeling_telechat.py:399
Method
forward
(self, hidden_states: torch.Tensor, residual: torch.Tensor)
models/12B_8bit/modeling_telechat.py:502
Method
forward
( self, hidden_states: torch.Tensor, attention_mask: torch.Tensor,
models/12B_8bit/modeling_telechat.py:524
Method
forward
( self, input_ids: Optional[torch.LongTensor] = None, past_key_values: Opt
models/12B_8bit/modeling_telechat.py:636
Method
forward
( self, input_ids: Optional[torch.LongTensor] = None, past_key_values: Opt
models/12B_8bit/modeling_telechat.py:781
Method
forward
(self, x, seq_dim=0, seq_len=None)
models/12B-V2/modeling_telechat.py:106
Method
forward
(self, hidden_states)
models/12B-V2/modeling_telechat.py:148
Method
forward
Implements the multihead softmax attention. Arguments --------- q, k, v: The tensor containing the query, key, and value.
models/12B-V2/modeling_telechat.py:178
Method
forward
(ctx, input: torch.Tensor)
models/12B-V2/modeling_telechat.py:299
Method
forward
(self, x: torch.Tensor)
models/12B-V2/modeling_telechat.py:322
Method
forward
( self, hidden_states: torch.Tensor, residual: torch.Tensor, a
models/12B-V2/modeling_telechat.py:399
Method
forward
(self, hidden_states: torch.Tensor, residual: torch.Tensor)
models/12B-V2/modeling_telechat.py:502
Method
forward
( self, hidden_states: torch.Tensor, attention_mask: torch.Tensor,
models/12B-V2/modeling_telechat.py:524
Method
forward
( self, input_ids: Optional[torch.LongTensor] = None, past_key_values: Opt
models/12B-V2/modeling_telechat.py:636
Method
forward
( self, input_ids: Optional[torch.LongTensor] = None, past_key_values: Opt
models/12B-V2/modeling_telechat.py:781
Function
get_all_reduce_mean
(tensor)
deepspeed-telechat/utils/utils.py:67
Function
get_dtype_size
(dtype)
deepspeed-telechat/utils/utils.py:28
Method
get_eval_data
(self)
deepspeed-telechat/utils/data/raw_datasets.py:18
Method
get_eval_data
(self)
deepspeed-telechat/utils/data/raw_datasets.py:40
Method
get_input_embeddings
(self)
models/7B_8bit/modeling_telechat.py:612
Method
get_input_embeddings
(self)
models/12B_4bit/modeling_telechat.py:612
Method
get_input_embeddings
(self)
models/12B/modeling_telechat.py:612
Method
get_input_embeddings
(self)
models/7B_4bit/modeling_telechat.py:612
Method
get_input_embeddings
(self)
models/12B-V2_8bit/modeling_telechat.py:612
Method
get_input_embeddings
(self)
models/12B_8bit/modeling_telechat.py:612
Method
get_input_embeddings
(self)
models/12B-V2/modeling_telechat.py:612
← previous
next →
501–600 of 680, ranked by callers