INDEX

Explanations

sentences or discussions regarding the absence or negation of concepts

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

Code, references, and names

np_acts-logits-general · gemini-2.0-flash

Swedish self-reference

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.18.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

<eos>

-0.55

is

-0.47

Was

-0.46

aarrggbb

-0.46

AndEndTag

-0.45

 protoimpl

-0.45

 Will

-0.45

 Play

-0.44

 useCallback

-0.43

POSITIVE LOGITS

ſelves

0.92

ſelf

0.89

 Reverso

0.79

 itſelf

0.78

 doubtnut

0.77

ειτουργ

0.74

 ་་

0.72

 greateſt

0.72

neſs

0.71

 themſelves

0.71

Activations Density 0.084%

sentences or discussions regarding the absence or negation of concepts

Code, references, and names

Swedish self-reference

No Comments

No Known Activations