INDEX

Explanations

references to categories and data structures in technical contexts

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

 betweenstory

-0.69

transQ

-0.55

 BoxFit

-0.50

 Feinde

-0.49

ỡng

-0.48

новниш

-0.48

ismet

-0.48

ynnwys

-0.47

 nhớ

-0.44

𝟹

-0.44

POSITIVE LOGITS

 autorytatywna

0.88

 nahilalakip

0.56

Données

0.48

Nationalité

0.46

Biographie

0.44

">//

0.44

 disambiguazione

0.43

 informée

0.43

Referències

0.43

StructEnd

0.43

Activations Density 0.001%

references to categories and data structures in technical contexts

No Comments

No Known Activations