Python module
max.pipelines.architectures.phi3
Phi-3 transformer architecture for text generation.
Phi3Model
class max.pipelines.architectures.phi3.Phi3Model(pipeline_config, session, devices, kv_cache_config, weights, adapter=None, return_logits=ReturnLogits.LAST_TOKEN)
Bases: LlamaModelBase
Phi 3 pipeline model implementation.
-
Parameters:
-
- pipeline_config (PipelineConfig) – The configuration for this pipeline.
- session (InferenceSession) – The container for the runtime for this model.
- devices (list[Device])
- kv_cache_config (KVCacheConfig)
- weights (Weights)
- adapter (WeightsAdapter | None)
- return_logits (ReturnLogits)
norm_method
norm_method: Literal['rms_norm'] | Literal['layer_norm'] = 'rms_norm'
Normalization layer.
Was this page helpful?
Thank you! We'll create more content like this.
Thank you for helping us improve!