INDEX

Explanations

occurrences of the word "under."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 17-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_17/width_65k/average_l0_125

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.17.hook_resid_post

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

databind

-0.83

 FetchType

-0.82

 MainAxisSize

-0.72

 CURIAM

-0.71

 ComVisible

-0.71

 initComponents

-0.71

BeginContext

-0.67

 MonoBehaviour

-0.67

__":

-0.65

 JpaRepository

-0.65

POSITIVE LOGITS

the

0.93

neath

0.79

 various

0.75

}}^

0.68

0.66

mantel

0.65

lining

0.65

 sotto

0.64

any

0.62

under

0.61

Activations Density 0.044%

occurrences of the word "under."

No Comments

No Known Activations