INDEX

Explanations

references to individuals and their associated activities or characteristics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Reſ

-0.68

InputTagHelper

-0.63

 purpoſe

-0.63

 Diſ

-0.62

 Anſ

-0.59

 ſche

-0.59

writerow

-0.59

 Conſ

-0.58

 ſtate

-0.56

 constituencies

-0.56

POSITIVE LOGITS

 Zufall

1.09

偶然

1.06

たまたま

1.01

 hasard

0.97

 случайно

0.96

 stumbled

0.93

 chanced

0.91

 fortu

0.90

 chance

0.89

 discovered

0.89

Activations Density 0.389%

references to individuals and their associated activities or characteristics

No Comments

No Known Activations