INDEX

Explanations

proper nouns and organizational affiliations

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

 تضيفلها

-0.50

 المعيارى

-0.47

IsMutable

-0.46

 cherchés

-0.45

новниш

-0.45

 سكانية

-0.44

最快更新

-0.43

 hikariConfig

-0.43

ագրություններ

-0.43

 насељу

-0.43

POSITIVE LOGITS

CodeAttribute

0.44

Xinhua

0.40



0.39

putnik

0.39

 Numerade

0.39

 purpoſe

0.38

ьаж

0.37

 FishBase

0.37

ualaikum

0.37

elemField

0.36

Activations Density 0.001%

proper nouns and organizational affiliations

No Comments

No Known Activations