INDEX

Explanations

occurrences of the letter 's' in various contexts

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

past decades

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

iſen

-0.54

 skär

-0.54

IsContent

-0.51

Hentet

-0.50

iſten

-0.50

請繼續往下閱讀

-0.49

ustimmung

-0.48

httphttps

-0.48

IMPORTED

-0.47

processable

-0.47

POSITIVE LOGITS

0.69

 years

0.58

<bos>

0.57

 Roskov

0.57

 twenties

0.57

 weeks

0.56

0.53

 decades

0.52

 months

0.52

CDs

0.50

Activations Density 0.011%

occurrences of the letter 's' in various contexts

past decades

No Comments

No Known Activations