INDEX

Explanations

statements of identity or self-identification

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

forms of to be

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

 ſche

-0.85

ngrx

-0.81

FormTagHelper

-0.75

ScopeManager

-0.74

ſchaft

-0.73

 beſch

-0.73

ésultats

-0.72

 NavController

-0.71

ſicht

-0.71

<unused8>

-0.71

POSITIVE LOGITS

0.33

toBe

0.30

Be

0.29

being

0.29

be

0.28

 super

0.28

 verdaderos

0.28

be

0.28

 быть

0.28

ir

0.27

Activations Density 0.072%

statements of identity or self-identification

forms of to be

No Comments

No Known Activations

statements of identity or self-identification

forms of to be

No Comments

No Known Activations