INDEX

Explanations

references to geographical locations and demographic details related to people

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.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

BeginContext

-0.65

tagHelperRunner

-0.50

ען

-0.48

rrggbb

-0.45

ฤษ

-0.45

sti

-0.43

bezier

-0.43

uestamente

-0.42

σια

-0.42

)<\

-0.40

POSITIVE LOGITS

 living

2.50

living

2.21

 lived

2.15

 live

2.10

Living

2.06

 residing

2.04

 Living

1.98

 reside

1.98

 lives

1.91

 LIVING

1.84

Activations Density 0.478%

references to geographical locations and demographic details related to people

No Comments

No Known Activations