INDEX

Explanations

quantities or numerical values represented in various formats

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-2b-pt-res/layer_2/width_16k/average_l0_141

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.2.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

Datuak

-0.63

Daz

-0.50

moveToFirst

-0.49

 raso

-0.48

↵↵

-0.48

izi

-0.47

po

-0.47

 нах

-0.45

مصادر

-0.45

 Aires

-0.45

POSITIVE LOGITS

 AssemblyCulture

0.71

iastes

0.70

исленность

0.70

haustible

0.69

reness

0.67

ghijkl

0.66

 useAppContext

0.66

 continuant

0.65

 anhydride

0.64

domada

0.64

Activations Density 0.936%

quantities or numerical values represented in various formats

No Comments

No Known Activations