INDEX

Explanations

references to the app "Hooked" and related discussions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 19-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_19/width_16k/average_l0_73

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.19.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

 Krä

-0.70

stand

-0.60

tain

-0.56

out

-0.56

stands

-0.56

po

-0.55

setScene

-0.55

IGNS

-0.54

ائم

-0.53

やかな

-0.53

POSITIVE LOGITS

 useCallback

1.02

 فريبيس

0.93

ATR

0.86

Huk

0.84

 Kishi

0.81

Sia

0.79

 Hooks

0.79

 للاسماء

0.79

 Sarm

0.75

 Eccle

0.75

Activations Density 0.441%

references to the app "Hooked" and related discussions

No Comments

No Known Activations