INDEX

Explanations

elements related to academic publications and their revisions

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

article versions or updates

np_acts-logits-general · gemini-2.0-flash

article updates and versions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 24-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_24/width_16k/average_l0_114

Prompts (Dashboard)

24,576 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

AccessorTable

-0.47

principalTable

-0.44

</thead>

-0.43

 Explo

-0.41

۩

-0.40

 تكبرها

-0.39

__":

-0.39

IBOutlet

-0.39

 buzz

-0.38

(;;

-0.38

POSITIVE LOGITS

 originally

0.54

 originalmente

0.52

originally

0.48

这篇

0.47

本文

0.44

 reprinted

0.43

 republished

0.43

esterno

0.43

 statistique

0.43

 tulisan

0.42

Activations Density 0.404%

elements related to academic publications and their revisions

article versions or updates

article updates and versions

No Comments

No Known Activations