INDEX

Explanations

references to specific geographical locations and administrative areas

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 22-gemmascope-mlp-16k

Configuration

google/gemma-scope-2b-pt-mlp/layer_22/width_16k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.22.hook_mlp_out

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 métallique

-0.65

 ProtoMessage

-0.63

出版年

-0.59

 wireType

-0.58

SharedDtor

-0.57

HasAnnotation

-0.55

 vectorielles

-0.54

MergeFrom

-0.54

featureID

-0.54

 clearColor

-0.54

POSITIVE LOGITS

Local

0.52

!*\

0.51

Eloquent

0.48

 Local

0.48

apollo

0.46

HasOne

0.46

local

0.45

locu

0.45

CodeDom

0.45

@"/

0.45

Activations Density 0.148%

references to specific geographical locations and administrative areas

No Comments

No Known Activations