INDEX

Explanations

elements related to evidence, models, and conflict in various contexts

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

patterns from listssupply, conflicts, interactions, caused, replacement, inflammatory, college

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

AndEndTag

-0.54

 вікісторінку

-0.46

 raiſ

-0.40

 chofe

-0.39

 informée

-0.39

 Comprometido

-0.39

 dép

-0.37

 Référence

-0.36

 للاسماء

-0.35

uxxxx

-0.34

POSITIVE LOGITS

richTextPanel

0.57

 الرياضيه

0.51

ParallelGroup

0.50

internalType

0.50

pit

0.49

 Numerade

0.49

MessageOf

0.48

GTCX

0.48

sidemargin

0.47

urna

0.47

Activations Density 0.396%

elements related to evidence, models, and conflict in various contexts

patterns from listssupply, conflicts, interactions, caused, replacement, inflammatory, college

No Comments

No Known Activations