INDEX

Explanations

proper nouns, specifically names and titles of individuals

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-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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

écnicas

-0.43

 PARSER

-0.41

的一部分

-0.39

Lup

-0.39

 karet

-0.38

sistors

-0.38

 AssemblyCulture

-0.38

adpleegd

-0.38

TagMode

-0.38

싶

-0.38

POSITIVE LOGITS

Obrázky

0.50

hbs

0.48

TextWatcher

0.45

 Numerade

0.45

setVerticalGroup

0.44

WebElementEntity

0.44

Према

0.42

лтемелер

0.41

 صوتيه

0.40

IBOutlet

0.40

Activations Density 0.055%

proper nouns, specifically names and titles of individuals

No Comments

No Known Activations