INDEX

Explanations

references to individuals or groups of people

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

following "person" or "Person"

np_acts-logits-general · gemini-2.0-flash

person and its role

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.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

ArgsConstructor

-1.11

BagLayout

-0.97

argout

-0.91

 WebDriverWait

-0.87

 EconPapers

-0.85

ConstraintMaker

-0.81

PerformLayout

-0.80

 WaitForSeconds

-0.80

 RouterModule

-0.79

 doubtnut

-0.78

POSITIVE LOGITS

 Persons

0.90

 PERSONS

0.84

who

0.82

Persons

0.82

 Person

0.80

 PERSON

0.80

 persons

0.79

persons

0.74

 person

0.71

ification

0.69

Activations Density 0.059%

references to individuals or groups of people

following "person" or "Person"

person and its role

No Comments

No Known Activations