INDEX

Explanations

instances of the word "in," indicating a focus on locations or contexts

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

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 ویکی‌پدی

-0.65

 صفحۀ

-0.57

Erreferentziak

-0.55

Aholisi

-0.53

 Италијани

-0.52

__*/

-0.52

httphttps

-0.50

__":

-0.47

 PagesJaunes

-0.46

__":

-0.46

POSITIVE LOGITS

kuuta

0.60

 vuonna

0.60

 early

0.57

 July

0.56

 בשנת

0.56

 năm

0.54

 June

0.53

since

0.53

 January

0.52

umumkan

0.51

Activations Density 0.069%

instances of the word "in," indicating a focus on locations or contexts

No Comments

No Known Activations