INDEX

Explanations

structured document sections and headings specifically related to legal or procedural text

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

LookAnd

-0.77

 surla

-0.68

Aholisi

-0.66

 Italijanski

-0.66

 InputDecoration

-0.66

点此举报

-0.62

出版年

-0.62

Әдәбият

-0.61

béco

-0.61

 GenerationType

-0.61

POSITIVE LOGITS

 automatic

0.31

 Zweifel

0.27

 honra

0.27

 Flä

0.27

เก

0.26

 Rady

0.26

fjspx

0.25

 exhi

0.25

บรร

0.25

Activations Density 0.134%

structured document sections and headings specifically related to legal or procedural text

No Comments

No Known Activations