INDEX

Explanations

patterns related to testing or assertions in code

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

 صوتيه

-0.77

 queſta

-0.75

ThroughAttribute

-0.75

IntoConstraints

-0.73

 MainAxisSize

-0.70

 protoimpl

-0.66

 betweenstory

-0.66

੝

-0.66

Tikang

-0.65

 ostavi

-0.65

POSITIVE LOGITS

 proyec

0.34

 pruebas

0.33

 conmigo

0.32

 kirj

0.32

 sanitarias

0.30

 ویکی‌پدیا

0.30

toBeDefined

0.29

 test

0.29

 Überprüfung

0.29

 Xunit

0.28

Activations Density 0.006%

patterns related to testing or assertions in code

No Comments

No Known Activations