INDEX

Explanations

names of brands and models related to technology and automotive

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

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

/*

-0.50

OGND

-0.50

/**

-0.49

encodeWith

-0.48

Diweddarwch

-0.48

hyrchwyd

-0.46

GEBURTSDATUM

-0.44

Autoritní

-0.44

tagHelperRunner

-0.44

 typelib

-0.43

POSITIVE LOGITS

 Reſ

0.52

 Inſ

0.52

 Conſ

0.49

 pleaſure

0.48

 Diſ

0.47

 Houſe

0.47

 Perſ

0.44

 actionPerformed

0.43

 ſta

0.42

 prate

0.42

Activations Density 0.748%

names of brands and models related to technology and automotive

No Comments

No Known Activations