INDEX

Explanations

references to dragons and their representations

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

 {}));

-0.45

wertig

-0.45

 frattempo

-0.44

 mínimos

-0.42

=============

-0.42

########.

-0.41

 koniec

-0.41

 wobec

-0.40

ftagPool

-0.40

 EconPapers

-0.40

POSITIVE LOGITS

 Dragon

1.26

Dragon

1.15

 dragon

1.11

 DRAGON

1.09

 Dragons

0.99

dragon

0.97

Dragons

0.93

 dragons

0.91

 dragón

0.79

 dragonfly

0.76

Activations Density 0.006%

references to dragons and their representations

No Comments

No Known Activations