INDEX

Explanations

references to documents and data structures

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

Tokens followed by numbers

np_acts-logits-general · gemini-2.0-flash

HCC organizations

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.18.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

<bos>

-0.65

кну

-0.54

ab

-0.54

-0.50

():

-0.49

ürk

-0.48

cla

-0.48

co

-0.48

-0.47

자인

-0.47

POSITIVE LOGITS

 للاسماء

0.95

 שוליים

0.87

EndInit

0.83

 Roskov

0.82

 <<<<<<<<<<<<<<

0.80

 فريبيس

0.78

]")]

0.78

 متعلقه

0.78

0.75

AndEndTag

0.74

Activations Density 0.550%

references to documents and data structures

Tokens followed by numbers

HCC organizations

No Comments

No Known Activations