INDEX

Explanations

phrases related to the initiation or overview of a topic or discussion

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

first one

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

GEBURTSDATUM

-0.78

 kasarigan

-0.65

 queſta

-0.60

numerusform

-0.59

 Walkover

-0.59

Obrázky

-0.57

__*/

-0.57

 ſta

-0.56

 uſed

-0.56

 Chriftian

-0.55

POSITIVE LOGITS

 inaugural

0.36

まずは

0.35

まず

0.35

 easiest

0.34

unächst

0.34

 first

0.34

 foremost

0.34

Firstly

0.34

First

0.33

 First

0.32

Activations Density 0.034%

phrases related to the initiation or overview of a topic or discussion

first one

No Comments

No Known Activations