INDEX

Explanations

distinct keywords and terms related to formal documents and identifiers

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

FREEDOM, LUCK, LEWIS, FILTER, GRANT, LINK

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 9-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.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

alyptus

-0.38

 تكبرها

-0.37

-0.36

OMM

-0.35

liant

-0.35

 ache

-0.34

minded

-0.34

enment

-0.34

 Vulcan

-0.34

nps

-0.34

POSITIVE LOGITS

----</

0.54

addPreferredGap

0.53

IVEREF

0.53

 masiva

0.51

uxxxx

0.50

 nahilalakip

0.46

aarrggbb

0.45

 كومونز

0.45

 detenidos

0.45

ніципалі

0.45

Activations Density 0.475%

distinct keywords and terms related to formal documents and identifiers

FREEDOM, LUCK, LEWIS, FILTER, GRANT, LINK

No Comments

No Known Activations