INDEX

Explanations

words related to indications or signs of various states or conditions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 0-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_0/width_16k/average_l0_105

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.0.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

 pitié

-0.83

 présidenti

-0.79

 neceff

-0.78

 poffe

-0.75

 poffible

-0.74

 auroit

-0.69

Demikian

-0.69

 dentaire

-0.69

 Cæsar

-0.68

 quæ

-0.68

POSITIVE LOGITS

 Sign

0.85

 exchange

0.82

 indication

0.82

 INDIC

0.81

Indicator

0.81

 Exchange

0.80

 exchanges

0.80

Tro

0.79

 sign

0.78

exchange

0.78

Activations Density 0.084%

words related to indications or signs of various states or conditions

No Comments

No Known Activations