INDEX

Explanations

numeric values and their comparisons in textual data

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

draws or ties

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 мәкал

-0.69

 صوتيه

-0.69

arangay

-0.68

MigrationBuilder

-0.67

 Geiſt

-0.66

adaptiveStyles

-0.65

 aikana

-0.63

 dieſem

-0.62

ſelben

-0.61

 ſei

-0.60

POSITIVE LOGITS

subsubsection

0.54

MM

0.52

//

0.51

WW

0.50

CC

0.49

ww

0.49

mm

0.48

HH

0.48

HH

0.48

ff

0.47

Activations Density 0.063%

numeric values and their comparisons in textual data

draws or ties

No Comments

No Known Activations