INDEX

Explanations

instances of looking and its variations in different contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_21/width_16k/average_l0_70

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

FormTagHelper

-0.65

WillAppear

-0.57

 externi

-0.56

 Akismet

-0.55

 estekak

-0.55

NameInMap

-0.54

 Cæsar

-0.53

 otomatig

-0.53

 Mystic

-0.50

 comigo

-0.50

POSITIVE LOGITS

 down

1.18

up

0.88

down

0.77

 Down

0.76

 DOWN

0.72

 around

0.68

 back

0.68

 xuống

0.67

Down

0.66

 вниз

0.64

Activations Density 0.044%

instances of looking and its variations in different contexts

No Comments

No Known Activations