INDEX

Explanations

numerical values related to measurements and medical typing

oai_token-act-pair · gpt-4o-mini

New Auto-Interp

Configuration

chanind/gemma-2-2b-batch-topk-matryoshka-saes-w-32k-l0-40/standard/blocks.12.hook_resid_post

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

32,768

Data Type

float32

Hook Name

blocks.12.hook_resid_post

Architecture

jumprelu

Context Size

1,024

Dataset

chanind/pile-uncopyrighted-gemma-1024-abbrv-1B

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 domy

-0.63

'-'

-0.61

 opération

-0.61

lý

-0.60

lí

-0.57

Wider

-0.57

 possible

-0.56

 sessionFactory

-0.56

جویان

-0.55

mbol

-0.54

POSITIVE LOGITS

 percentile

0.87

percentile

0.77

Efq

0.64

NewRow

0.63

migrationBuilder

0.61

entile

0.59

 Perc

0.59

centiles

0.56

 saites

0.54

NewGuid

0.53

Activations Density 0.014%

numerical values related to measurements and medical typing

No Comments

No Known Activations