INDEX

Explanations

references to investigations, assessments, and inquiries

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 13-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_13/width_65k/average_l0_75

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.13.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

PhysRevLett

-0.64

 DriverManager

-0.64

mappedBy

-0.55

cipar

-0.53

 Wiktionnaire

-0.51

meldungen

-0.50

lapsingToolbar

-0.49

 disponibilités

-0.47

 unselfish

-0.46

yelitis

-0.45

POSITIVE LOGITS

ientôt

0.68

ViewFeatures

0.65

 questioning

0.62

 inquiry

0.61

 explore

0.61

 investigate

0.60

 inquire

0.60

 evaluate

0.60

 exploring

0.60

principalTable

0.59

Activations Density 0.152%

references to investigations, assessments, and inquiries

No Comments

No Known Activations