INDEX

Explanations

negations or phrases emphasizing the absence of something

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 7-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_7/width_16k/average_l0_69

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.7.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

codehaus

-0.83

ässe

-0.69

Moose

-0.68

λους

-0.68

 Trasp

-0.66

Datuak

-0.65

"`

-0.65

 mxArray

-0.65

⇑

-0.65

いらっしゃ

-0.64

POSITIVE LOGITS

 nothing

1.65

nothing

1.64

NOTHING

1.62

 Nothing

1.60

 NOTHING

1.55

Nothing

1.54

 anything

1.20

 Anything

1.19

anything

1.19

 nothin

1.17

Activations Density 0.052%

negations or phrases emphasizing the absence of something

No Comments

No Known Activations