INDEX

Explanations

prefixes related to negation or reversal, particularly those starting with "un."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 14-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_14/width_16k/average_l0_84

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.14.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

 ainfi

-1.04

 MainAxisSize

-1.03

 wireType

-0.95

 Theſe

-0.92

 Berikut

-0.91

 ſche

-0.90

 becauſe

-0.89

Berikut

-0.89

 itſelf

-0.89

 auffi

-0.87

POSITIVE LOGITS

Un

1.42

UN

1.29

un

1.27

Un

1.25

un

1.19

UN

1.13

Dun

0.89

0.88

Bun

0.86

Al

0.83

Activations Density 0.107%

prefixes related to negation or reversal, particularly those starting with "un."

No Comments

No Known Activations