Model Configuration
Analyzing model architecture...
Loading configuration...
MoE Enabled
Analysis time: ms
Parameters
Total
Active
Main backbone
Main active
Embeddings
Transformer
Model Params
Quant Metadata
Checkpoint Total
Struct Diff
Computation
FLOPs / token
Attention Type
Layers
Memory Traffic
Memory / token
Hidden Size
Data Type
float16
Mixture of Experts (MoE)
MoE Layers
Active Experts / Token
Total Experts
Raw PyTorch Modules (print(model))
Module Classification
| Module Name | Class | Parameters | Category |
|---|