INDEX

Explanations

entities referring to people or specific roles in social contexts

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

TagMode

-0.83

存于互联网档案馆

-0.77

 disponibilités

-0.74

OGND

-0.71

 yourselves

-0.68

uxxxx

-0.68

AutoScaleMode

-0.65

AddField

-0.64

 sẻ

-0.63

yourself

-0.63

POSITIVE LOGITS

1.22

an

1.10

 some

1.04

the

1.00

 plenty

0.81

 another

0.78

 access

0.75

 their

0.75

 something

0.75

DockStyle

0.73

Activations Density 0.191%

entities referring to people or specific roles in social contexts

No Comments

No Known Activations