INDEX

Explanations

actions related to system modifications or code changes

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

removing or stopping something

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

PreferredItem

-0.40

 ſtate

-0.40

 clarinet

-0.39

 ſche

-0.39

 للمعارف

-0.39

 itſelf

-0.38

StatusBadRequest

-0.38

 becauſe

-0.38

ControllerAdvice

-0.38

 ویکی‌آمباردا

-0.36

POSITIVE LOGITS

 nakalista

0.49

principalTable

0.42

 reverted

0.41

合衆

0.41

REMOV

0.41

 Seitz

0.41

 mijne

0.40

┏

0.40

 exposed

0.40

Życiorys

0.40

Activations Density 0.031%

actions related to system modifications or code changes

removing or stopping something

No Comments

No Known Activations