INDEX

Explanations

instances of the word "Nothing."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 1-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_1/width_16k/average_l0_102

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.1.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

 more

-0.54

 other

-0.53

 nice

-0.50

 large

-0.49

 even

-0.49

 BoxDecoration

-0.49

萌

-0.47

 small

-0.46

saraba

-0.46

 quality

-0.46

POSITIVE LOGITS

 Nothing

1.15

 Anything

1.13

 Something

1.11

 Such

1.09

Nothing

1.08

 Things

1.07

Any

1.06

Something

1.02

 Anybody

1.01

Anything

0.96

Activations Density 0.171%

instances of the word "Nothing."

No Comments

No Known Activations