INDEX

Explanations

elements related to structured data formats or programming language syntax

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

Text after parentheses

np_acts-logits-general · gemini-2.0-flash

numbers and code

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

bootstrapcdn

-0.81

-0.60

setViewportView

-0.54

 Third

-0.53

 third

-0.51

errHandler

-0.51

 Three

-0.50

TypedDataSet

-0.49

 trio

-0.48

 three

-0.48

POSITIVE LOGITS

ArrowToggle

0.74

 hashes

0.59

getMenuInflater

0.56

ISCHE

0.55

 NSCoder

0.53

 Audiodateien

0.52

Conven

0.50

lossene

0.50

ruti

0.50

łoż

0.50

Activations Density 0.352%

elements related to structured data formats or programming language syntax

Text after parentheses

numbers and code

No Comments

No Known Activations