INDEX

Explanations

terms related to viability and safety in various contexts

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

non-standard or technical identifiers

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

VersionUID

-0.37

 hyper

-0.36

DrawerToggle

-0.36

 KIDS

-0.35

 Ivanov

-0.35

minente

-0.35

 Bianchi

-0.34

 terminated

-0.34

 cross

-0.34

dib

-0.34

POSITIVE LOGITS

 Jefus

0.62

 оригіналу

0.60

AndEndTag

0.60

 يتيمه

0.57

 himſelf

0.56

tagHelperRunner

0.56

 ſche

0.56

haikusbot

0.56

 ſeveral

0.56

 Houſe

0.56

Activations Density 0.010%

terms related to viability and safety in various contexts

non-standard or technical identifiers

No Comments

No Known Activations