INDEX

Explanations

negative descriptors or terms indicating poor quality or morality

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

Follows the word "bad"

np_acts-logits-general · gemini-2.0-flash

negative attributes

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

\{\\

-0.74

RectangleBorder

-0.67

/*

-0.64

انظر

-0.63

 محفوظة

-0.61

stateProvider

-0.61

onAttach

-0.61

openqa

-0.60

ResumeLayout

-0.59

bkz

-0.59

POSITIVE LOGITS

bad

0.73

 luck

0.70

Bad

0.70

gering

0.70

Bad

0.69

dies

0.68

ajoz

0.66

inage

0.66

 news

0.65

bad

0.65

Activations Density 0.110%

negative descriptors or terms indicating poor quality or morality

Follows the word "bad"

negative attributes

No Comments

No Known Activations