INDEX

Explanations

names or terms related to specific authorities or figures

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

Follows the tokens U, M, or Ba

np_acts-logits-general · gemini-2.0-flash

Mziwamadoda Qwabe

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 24-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_24/width_16k/average_l0_73

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.24.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

ConstraintMaker

-0.57

ьаж

-0.55

 يتيمه

-0.51

UAWEI

-0.50

:+:

-0.49

erview

-0.47

Viitteet

-0.47

versa

-0.45

;->

-0.44

httphttps

-0.44

POSITIVE LOGITS

 дописавши

0.60

Autoritní

0.57

 bandou

0.51

قایناق‌لار

0.51

kung

0.51

 Ejecutivo

0.50

ungan

0.49

MessageOf

0.49

 Corbin

0.49

(&:

0.48

Activations Density 0.058%

names or terms related to specific authorities or figures

Follows the tokens U, M, or Ba

Mziwamadoda Qwabe

No Comments

No Known Activations