INDEX

Explanations

items related to numerical data or identifiers

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

 avoient

-0.52

 étoit

-0.51

 pouvoit

-0.45

 disambiguazione

-0.45

 Wiktionnaire

-0.45

出版年

-0.44

сылкі

-0.44

 étoient

-0.43

 fermée

-0.42

webElementXpaths

-0.42

POSITIVE LOGITS

principalTable

0.49

XmlAccessType

0.49

 Cabral

0.49

septic

0.48

dron

0.48

Sto

0.45

oley

0.45

میزی

0.45

carbon

0.45

stag

0.45

Activations Density 0.021%

items related to numerical data or identifiers

No Comments

No Known Activations