INDEX

Explanations

specific scientific or biological terms related to entities or concepts, particularly those involving acronyms and notations

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

Two or three letter acronyms

np_acts-logits-general · gemini-2.0-flash

secret codes

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-2b-pt-res/layer_16/width_16k/average_l0_78

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.16.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

-1.15

-1.02

-1.00

-0.99

-0.98

-0.95

ng

-0.84

-0.83

lin

-0.80

-0.78

POSITIVE LOGITS

 poffe

1.01

elfare

1.00

posedge

0.99

 Jefus

0.98

wiſe

0.97

 raiſ

0.97

annica

0.96

 myſelf

0.95

łucha

0.94

LDA

0.94

Activations Density 1.001%

specific scientific or biological terms related to entities or concepts, particularly those involving acronyms and notations

Two or three letter acronyms

secret codes

No Comments

No Known Activations