INDEX

Explanations

references to trademarks and specific product names

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

 createState

-0.43

 Walkover

-0.39

Manbalar

-0.39

 ferroviaire

-0.37

httphttps

-0.37

 ordem

-0.36

AddTagHelper

-0.36

GEBURTSDATUM

-0.35

 ModelExpression

-0.35

Kjelder

-0.35

POSITIVE LOGITS

®

0.89

®,

0.77

™

0.73

®.

0.71

®

0.69

circledR

0.68

®:

0.68

Ⓡ

0.68

®-

0.66

™,

0.65

Activations Density 0.400%

references to trademarks and specific product names

No Comments

No Known Activations