INDEX

Explanations

image-related content or references to images

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

image captions

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

 snippetHide

-0.59

AddTagHelper

-0.55

########.

-0.52

bootstrapcdn

-0.52

 autorytatywna

-0.52

Rohy

-0.51

bukkit

-0.49

nestjs

-0.48

 getItemId

-0.48

tangentMode

-0.47

POSITIVE LOGITS

Télécharger

0.46

LEGGI

0.44

 roulant

0.44

 sanitarias

0.44

 adicionales

0.44

toMatchSnapshot

0.43

 artístico

0.43

 paralelo

0.43

UrlResolution

0.42

 Jurí

0.42

Activations Density 0.005%

image-related content or references to images

image captions

No Comments

No Known Activations