INDEX

Explanations

negative expressions and emphatic denials

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

Negation, possibility, or disbelief

np_acts-logits-general · gemini-2.0-flash

never / cannot / not + verb

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

ījum

-0.53

Obras

-0.51

 manifestly

-0.48

LayoutConstraint

-0.48

 beneficial

-0.48

 MessageBoxIcon

-0.48

прия

-0.47

 berke

-0.47

 senang

-0.47

SuppressLint

-0.46

POSITIVE LOGITS

 disappoint

0.82

 disappointment

0.76

 disappointed

0.72

 disappointments

0.71

 Dull

0.70

 оригіналу

0.66

 regrets

0.65

 boring

0.65

 Boring

0.65

 disap

0.64

Activations Density 0.165%

negative expressions and emphatic denials

Negation, possibility, or disbelief

never / cannot / not + verb

No Comments

No Known Activations