INDEX

Explanations

phrases indicating alternatives or comparisons

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 9-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_9/width_16k/average_l0_73

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.9.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

 femininas

-0.68

UnusedPrivate

-0.68

клопе

-0.68

 femininos

-0.67

CascadeType

-0.65

 mußten

-0.63

DebuggerNonUser

-0.62

JspWriter

-0.61

 virgen

-0.61

 correctes

-0.61

POSITIVE LOGITS

Instead

1.35

 Instead

1.32

 instead

1.27

instead

1.06

 zamiast

1.02

 Rather

0.99

tdessen

0.96

Rather

0.96

 Statt

0.90

 вместо

0.90

Activations Density 0.137%

phrases indicating alternatives or comparisons

No Comments

No Known Activations