INDEX

Explanations

technical terms and keywords related to user interfaces and data management

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

loss of job / resolution

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

 Bradford

-0.51

xane

-0.50

tizia

-0.50

Cid

-0.50

 createState

-0.49

terre

-0.48

 Task

-0.47

 Trades

-0.47

Meh

-0.47

fisk

-0.47

POSITIVE LOGITS

 Roskov

0.66

تقاوى

0.63

oredCriteria

0.62

tanleria

0.57

utilisons

0.56

 transfieras

0.55

ніципа

0.55

 незавершена

0.53

Хьажоргаш

0.49

 autorytatywna

0.47

Activations Density 0.000%

technical terms and keywords related to user interfaces and data management

loss of job / resolution

No Comments

No Known Activations