Code
Hub
Workspaces
Following
Trending
Connect
MCP
copy
Create free account
hub
/
github.com/QwenLM/ParScale
/ types & classes
Types & classes
14 in github.com/QwenLM/ParScale
⨍
Functions
52
◇
Types & classes
14
↳
Endpoints
1
↓ 4 callers
Class
Qwen2Model
Transformer decoder consisting of *config.num_hidden_layers* layers. Each layer is a [`Qwen2DecoderLayer`] Args: config: Qwen2ParSca
modeling_qwen2_parscale.py:532
↓ 3 callers
Class
Qwen2RMSNorm
modeling_qwen2_parscale.py:269
↓ 1 callers
Class
ParscaleCache
modeling_qwen2_parscale.py:133
↓ 1 callers
Class
Qwen2Attention
Multi-headed attention from 'Attention Is All You Need' paper
modeling_qwen2_parscale.py:166
↓ 1 callers
Class
Qwen2DecoderLayer
modeling_qwen2_parscale.py:289
↓ 1 callers
Class
Qwen2MLP
modeling_qwen2_parscale.py:46
↓ 1 callers
Class
Qwen2RotaryEmbedding
modeling_qwen2_parscale.py:346
Class
KwargsForCausalLM
modeling_qwen2_parscale.py:825
Class
Qwen2ForQuestionAnswering
modeling_qwen2_parscale.py:1148
Class
Qwen2ForSequenceClassification
modeling_qwen2_parscale.py:964
Class
Qwen2ForTokenClassification
modeling_qwen2_parscale.py:1060
Class
Qwen2ParScaleConfig
r""" This is the configuration class to store the configuration of a [`Qwen2Model`]. It is used to instantiate a Qwen2 model according to the
configuration_qwen2_parscale.py:11
Class
Qwen2ParScaleForCausalLM
modeling_qwen2_parscale.py:828
Class
Qwen2PreTrainedModel
modeling_qwen2_parscale.py:428