INDEX

Explanations

structured data formats and organization in text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 14-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_14/width_16k/average_l0_84

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.14.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

WebVitals

-0.76

 curieux

-0.61

formik

-0.57

 étoient

-0.56

สำ

-0.55

thebibliography

-0.53

 परि

-0.52

 étape

-0.52

 ufficio

-0.52

 intérieure

-0.51

POSITIVE LOGITS

 separate

0.80

separate

0.75

 seperate

0.69

 occupying

0.65

 SEPAR

0.62

 space

0.61

 split

0.60

 spread

0.59

 occupy

0.59

afficheront

0.58

Activations Density 0.487%

structured data formats and organization in text

No Comments

No Known Activations