INDEX

Explanations

references to specific names or identifiers related to individuals or products

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

 snippetHide

-0.73

 BoxFit

-0.62

 ſein

-0.62

 ſont

-0.62

 ſei

-0.60

 iNdEx

-0.59

 ویکی‌پدی

-0.59

verifyException

-0.59

 increí

-0.58

ſcher

-0.56

POSITIVE LOGITS

方

0.36

endregion

0.34

 only

0.34

0.32

gewicht

0.32

gnore

0.32

 bera

0.32

 finally

0.31

Activations Density 0.007%

references to specific names or identifiers related to individuals or products

No Comments

No Known Activations