INDEX

Explanations

references to data, statistics, or numerical information tied to research or surveys

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

statistical figures respondents

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

WithIOException

-0.55

 calendriers

-0.50

 ſta

-0.50

Tembelea

-0.50

HomeAsUpEnabled

-0.48

 ſte

-0.48

ynchronously

-0.47

 purpoſe

-0.45

enderror

-0.45

+:+

-0.45

POSITIVE LOGITS

styleable

0.65

 survey

0.49

 Survey

0.48

 respondents

0.47

 polled

0.47

 statistics

0.45

 surveyed

0.44

 preferences

0.41

ViewFeatures

0.41

 Statistics

0.40

Activations Density 0.296%

references to data, statistics, or numerical information tied to research or surveys

statistical figures respondents

No Comments

No Known Activations