INDEX

Explanations

references to programming components or elements in a technical context

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

hematical

-0.76

erapeutic

-0.72

ividual

-0.68

rupulous

-0.65

orithms

-0.65

litude

-0.65

orithm

-0.65

intenance

-0.65

pectives

-0.64

ervations

-0.63

POSITIVE LOGITS

om

0.63

ang

0.55

 getItemId

0.54

ow

0.54

el

0.54

est

0.53

ash

0.53

 createSprite

0.52

ot

0.52

 raiſ

0.52

Activations Density 0.518%

references to programming components or elements in a technical context

No Comments

No Known Activations