INDEX

Explanations

instances of the word "don’t" or its variations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 17-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_17/width_65k/average_l0_125

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.17.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

 itſelf

-1.21

 ModelExpression

-1.19

^(@)

-1.14

BibitemShut

-1.10

 Мексичка

-1.10

Lähteet

-1.09

✨:

-1.08

 للمعارف

-1.08

 CURIAM

-1.08

WithIOException

-1.06

POSITIVE LOGITS

ing

1.17

1.16

ning

1.02

don

0.98

Don

0.93

0.89

don

0.88

0.84

0.82

Don

0.81

Activations Density 0.025%

instances of the word "don’t" or its variations

No Comments

No Known Activations