INDEX

Explanations

proper nouns, particularly names and locations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 12-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_12/width_16k/average_l0_130

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.12.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

 Wiktionnaire

-0.64

iastes

-0.58

cetines

-0.58

formin

-0.53

 pazi

-0.53

 telefónica

-0.51

 rozho

-0.51

riezmann

-0.50

 häls

-0.50

 nervioso

-0.49

POSITIVE LOGITS

PageRoute

0.59

 Milo

0.53

InputBorder

0.52

httphttps

0.52

era

0.51

JScrollPane

0.50

Era

0.50

CascadeType

0.50

startY

0.50

 Maro

0.50

Activations Density 2.077%

proper nouns, particularly names and locations

No Comments

No Known Activations