INDEX

Explanations

references to multimedia content, particularly videos and audio within discussions or presentations

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

audio/video related to arguments

np_acts-logits-general · gemini-2.0-flash

oral video audio spoken words

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-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

DockStyle

-0.57

 autorytatywna

-0.57

 éditoriale

-0.50

__(/*!

-0.47

InputBorder

-0.47

 FileWriter

-0.45

 الحره

-0.45

wpdb

-0.44

BorderFactory

-0.43

toHave

-0.42

POSITIVE LOGITS

 oral

0.62

 audio

0.58

hyrchwyd

0.56

 video

0.54

Oral

0.54

 Oral

0.53

 videos

0.52

Audio

0.51

 spoken

0.50

VIDEO

0.50

Activations Density 0.864%

references to multimedia content, particularly videos and audio within discussions or presentations

audio/video related to arguments

oral video audio spoken words

No Comments

No Known Activations