INDEX

Explanations

punctuation and specific formatting symbols or commands in the text

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

programming constructs

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

principalColumn

-0.54

 فريبيس

-0.53

RenderAtEndOf

-0.52

 estekak

-0.48

 useAppContext

-0.47

PerformLayout

-0.47

 صوتيه

-0.46

hyrchwyd

-0.46

Życiorys

-0.44

MessageTagHelper

-0.44

POSITIVE LOGITS

Rational

0.44

Donut

0.43

Hollywood

0.43

sag

0.43

omal

0.42

<eos>

0.41

Fordítás

0.41

casian

0.41

Radial

0.40

@}

0.39

Activations Density 0.012%

punctuation and specific formatting symbols or commands in the text

programming constructs

No Comments

No Known Activations