INDEX

Explanations

phrases related to health and safety warnings

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

"Cor" or "Kor" followed by text

np_acts-logits-general · gemini-2.0-flash

cor, correlations, corrector

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 22-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_22/width_16k/average_l0_72

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.22.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

 Saito

-0.63

 recl

-0.60

 виправивши

-0.60

 Beatty

-0.57

 Buckley

-0.57

Mue

-0.56

:✨

-0.55

 Multiplier

-0.55

 Kazan

-0.53

 Haywood

-0.52

POSITIVE LOGITS

COR

2.14

Cor

2.13

Kor

2.12

Cor

2.07

COR

2.07

cor

2.04

cor

1.97

Kor

1.96

kor

1.88

kor

1.86

Activations Density 0.469%

phrases related to health and safety warnings

"Cor" or "Kor" followed by text

cor, correlations, corrector

No Comments

No Known Activations