INDEX

Explanations

terms related to errors and API requests

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

prime factors and common divisors

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

Diwedd

-0.60

GEBURTSDATUM

-0.58

uxxxx

-0.54

 autorytatywna

-0.54

WriteTagHelper

-0.53

mybatisplus

-0.52

 виправивши

-0.52

__*/

-0.49

fromnode

-0.49

 nakalista

-0.48

POSITIVE LOGITS

Wut

0.41

URED

0.41

ITECH

0.41

Hentet

0.40

AISSEE

0.40

 Foundation

0.40

MNA

0.40

ation

0.40

 European

0.39

”

0.39

Activations Density 0.063%

terms related to errors and API requests

prime factors and common divisors

No Comments

No Known Activations