INDEX

Explanations

names or titles associated with individuals or entities in a biographical or historical context

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 ainfi

-0.77

 acrylique

-0.77

 وتسجيلات

-0.75

rungsseite

-0.74

 avoient

-0.74

 chré

-0.74

 feroit

-0.73

 étoient

-0.70

 enfans

-0.68

 auroit

-0.68

POSITIVE LOGITS

0.89

̍t

0.63

いざ

0.60

 दें

0.55

AsUp

0.54

<bos>

0.54

ッシャー

0.53

 care

0.53

udesta

0.52

Weaknesses

0.51

Activations Density 0.367%

names or titles associated with individuals or entities in a biographical or historical context

No Comments

No Known Activations