INDEX

Explanations

phrases indicating the quantity "two"

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 10-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_10/width_16k/average_l0_77

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.10.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

ly

-1.01

 vaisselle

-0.72

čaj

-0.68

 argint

-0.67

orteur

-0.66

er

-0.65

 sagesse

-0.65

mer

-0.65

ñores

-0.63

QLD

-0.61

POSITIVE LOGITS

two

1.14

Two

1.13

 CreateTagHelper

1.12

Two

1.07

TWO

1.07

TWO

1.05

 dozen

1.05

Према

1.01

two

0.98

醐

0.97

Activations Density 0.158%

phrases indicating the quantity "two"

No Comments

No Known Activations