INDEX

Explanations

terms related to guidance and direction in various contexts

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

guiding and guidance

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 9-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.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

classnames

-0.44

 paard

-0.42

 broken

-0.39

 notorious

-0.39

 scom

-0.38

 busted

-0.37

 broke

-0.37

broke

-0.37

 notoriously

-0.36

 noxious

-0.36

POSITIVE LOGITS

 guiding

1.23

 guide

1.20

 guided

1.19

 guidance

1.19

 Guidance

1.17

 Guiding

1.13

 guides

1.12

 guider

1.11

 Guide

1.10

Guidance

1.10

Activations Density 0.019%

terms related to guidance and direction in various contexts

guiding and guidance

No Comments

No Known Activations