INDEX

Explanations

indications of importance or emphasis on specific phrases or actions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.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

ремен

-0.44

xp

-0.44

 Kirche

-0.43

hour

-0.43

zyn

-0.42

 followed

-0.42

 Ferne

-0.41

jdź

-0.40

рачи

-0.40

hodné

-0.40

POSITIVE LOGITS

 فريبيس

0.99

HasForeignKey

0.91

UrlResolution

0.89

 auprès

0.85

 समीक्षक

0.84

Efq

0.83

Tembelea

0.82

Handlung

0.82

CloseOperation

0.82

sendTo

0.81

Activations Density 0.289%

indications of importance or emphasis on specific phrases or actions

No Comments

No Known Activations