INDEX

Explanations

proprietary institutional affiliations and acknowledgments

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 17-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_17/width_65k/average_l0_125

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.17.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

 ExecuteAsync

-0.83

IntoConstraints

-0.69

 initComponents

-0.67

 Numerade

-0.66

 NSCoder

-0.66

 Pingback

-0.64

})`

-0.64

IBar

-0.62

Koa

-0.61

ptonshire

-0.61

POSITIVE LOGITS

UU

0.79

0.73

uu

0.71

UIS

0.69

UB

0.66

 ungu

0.66

UU

0.65

 uVar

0.65

UE

0.65

Ull

0.64

Activations Density 0.931%

proprietary institutional affiliations and acknowledgments

No Comments

No Known Activations