INDEX

Explanations

signs of starting or introducing a new topic or idea

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 13-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_13/width_65k/average_l0_75

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.13.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

 myſelf

-1.13

 ―――――

-1.13

 itſelf

-1.03

 Monfieur

-1.03

 Houſe

-1.02

 Majefty

-1.01

 ſeveral

-1.01

 Reſ

-0.99

IntoConstraints

-0.96

 pleaſure

-0.96

POSITIVE LOGITS

 alternatively

0.85

 Alternatively

0.80

Alternatively

0.74

if

0.63

 же

0.61

干脆

0.60

 alternativ

0.60

逆に

0.59

or

0.59

 αντί

0.59

Activations Density 0.088%

signs of starting or introducing a new topic or idea

No Comments

No Known Activations