INDEX

Explanations

occurrences of the word "pronounce" and its variants

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

describing something to a high degree

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_31/width_131k/average_l0_109

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.31.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

BagLayout

-0.49

talks

-0.47

🏾

-0.47

Tikang

-0.47

Dữ

-0.44

Kam

-0.44

Ei

-0.44

ThroughAttribute

-0.44

 Seeker

-0.44

Wes

-0.43

POSITIVE LOGITS

 pronounce

0.57

 iNdEx

0.55

 Pron

0.49

able

0.48

Pron

0.46

__*/

0.44

ments

0.43

NamedQuery

0.42

ably

0.42

XmlAccessType

0.41

Activations Density 0.008%

occurrences of the word "pronounce" and its variants

describing something to a high degree

No Comments

No Known Activations