INDEX

Explanations

references to inventions or inventions themselves

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 3-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_3/width_16k/average_l0_59

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.3.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

tagHelperRunner

-0.76

neſs

-0.75

 ſche

-0.74

ſelf

-0.70

ſelves

-0.68

 themſelves

-0.68

 purpoſe

-0.65

 itſelf

-0.65

 neceſſ

-0.65

 ſte

-0.64

POSITIVE LOGITS

ofthe

0.63

 della

0.62

 dell

0.59

del

0.57

GraphicsUnit

0.55

 thereof

0.54

OfThe

0.53

HasBeenSet

0.53

 metade

0.52

of

0.52

Activations Density 0.293%

references to inventions or inventions themselves

No Comments

No Known Activations