INDEX

Explanations

phrases indicating initial actions or steps in a process

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

first step

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.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

 commented

-0.42

 pelúcia

-0.39

 Comentarios

-0.39

 comentario

-0.39

 Coment

-0.38

 embreagem

-0.38

 Ciencias

-0.38

numerusform

-0.38

onChanged

-0.38

věd

-0.37

POSITIVE LOGITS

第一步

0.52

tvguidetime

0.44

IVEREF

0.44

 насељу

0.44

PreExecute

0.44

тьяна

0.43

 steps

0.42

 priority

0.41

 begins

0.41

 prerequisite

0.41

Activations Density 0.018%

phrases indicating initial actions or steps in a process

first step

No Comments

No Known Activations