INDEX

Explanations

references to specific code abstractions or object-oriented programming concepts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.18.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

er

-1.17

-0.87

checkable

-0.71

Activités

-0.70

ER

-0.67

ers

-0.63

eX

-0.63

OX

-0.63

Dixon

-0.61

st

-0.61

POSITIVE LOGITS

NameInMap

1.02

 disambiguazione

0.99

 Paglinawan

0.96

 defaultstate

0.95

EDEFAULT

0.91

 متعلقه

0.86

}))

0.86

RenderAtEndOf

0.85

########.

0.82

Zitat

0.82

Activations Density 0.590%

references to specific code abstractions or object-oriented programming concepts

No Comments

No Known Activations