INDEX

Explanations

mentions of the word "fresh."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 0-gemmascope-mlp-65k

Configuration

google/gemma-scope-2b-pt-mlp/layer_0/width_65k/average_l0_72

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.0.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

 estekak

-0.90

 مشين

-0.81

 fresh

-0.68

تفصیلات

-0.66

 originais

-0.62

uginosa

-0.60

EndGlobalSection

-0.58

 Италијани

-0.58

 incomplète

-0.57

CWE

-0.56

POSITIVE LOGITS

 arras

0.63

Turismo

0.57

<h6>

0.54

casio

0.54

\{\\

0.53

 debo

0.52

гри

0.51

 Based

0.51

版

0.51

Platforms

0.51

Activations Density 0.014%

mentions of the word "fresh."

No Comments

No Known Activations