INDEX

Explanations

words and phrases related to specific details, configurations, or identifiers within contexts

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

IRST

-0.49

Coe

-0.49

 devotee

-0.46

otal

-0.45

pegno

-0.44

Rem

-0.44

jad

-0.43

 samp

-0.42

Oto

-0.42

rad

-0.42

POSITIVE LOGITS

richTextPanel

0.50

nezeu

0.48

 Bewußt

0.45

tagHelperRunner

0.44

etCode

0.43

webElementXpaths

0.42

etcode

0.42

 Paglinawan

0.41

 pouvoit

0.41

Personendaten

0.41

Activations Density 1.219%

words and phrases related to specific details, configurations, or identifiers within contexts

No Comments

No Known Activations