INDEX

Explanations

negation terms or phrases such as "nor" and "neither."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 11-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_11/width_16k/average_l0_80

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.11.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

fwd

-0.68

pezi

-0.66

ittens

-0.66

 ARAB

-0.65

SAX

-0.62

wiście

-0.60

 bulk

-0.60

 Brag

-0.59

 Turn

-0.58

tı

-0.58

POSITIVE LOGITS

neither

1.24

Nor

1.20

 neither

1.18

nor

1.17

nor

1.12

NOR

1.12

Neither

1.09

 Neither

1.08

Nor

1.07

theless

1.05

Activations Density 0.068%

negation terms or phrases such as "nor" and "neither."

No Comments

No Known Activations