INDEX

Explanations

occurrences of the word "false."

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

%)$

-0.70

μαι

-0.68

yntaxException

-0.65

 aufnehmen

-0.65

 Kirkwood

-0.63

akaian

-0.62

WireFormatLite

-0.62

spricht

-0.62

 referenties

-0.61

lingo

-0.61

POSITIVE LOGITS

False

0.73

 False

0.71

 false

0.71

FALSE

0.67

stdbool

0.66

false

0.64

 مرئيه

0.64

 FALSE

0.64

GraphicsUnit

0.64

 falsely

0.60

Activations Density 0.047%

occurrences of the word "false."

No Comments

No Known Activations