INDEX

Explanations

updates or progress reports related to scientific research or software development

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

shows that or lists people

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_31/width_131k/average_l0_109

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.31.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

 móvel

-0.46

 práci

-0.45

 kapsul

-0.44

 ilman

-0.42

 rodzic

-0.42

Coordonnées

-0.42

 améli

-0.41

Synonymes

-0.40

 konsumsi

-0.40

 amanecer

-0.40

POSITIVE LOGITS

 <=",

0.57

 <<<<<<<<<<<<<<

0.56

 relevant

0.56

relevant

0.56

 Relevant

0.53

 official

0.52

 reportedly

0.51

yscale

0.50

 successively

0.50

 rele

0.49

Activations Density 0.005%

updates or progress reports related to scientific research or software development

shows that or lists people

No Comments

No Known Activations