INDEX

Explanations

terms indicating significant or noteworthy experiences

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

memorable things

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

IntoConstraints

-0.77

 հղումներ

-0.62

 utafitiHapana

-0.59



-0.59

 שוליים

-0.57

Hochspringen

-0.57

XMLSchema

-0.56

expandindo

-0.54

 CallOverrides

-0.54

 Monfieur

-0.53

POSITIVE LOGITS

 memorable

1.10

 unforgettable

1.00

forgettable

0.86

 inolvid

0.77

 mémor

0.75

 memories

0.70

 memor

0.69

memor

0.69

 remembers

0.59

 remembered

0.59

Activations Density 0.003%

terms indicating significant or noteworthy experiences

memorable things

No Comments

No Known Activations