INDEX

Explanations

structured data representations with specific formatting and layout attributes

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

httphttps

-0.81

webElementXpaths

-0.63

OGND

-0.62



-0.61

TagMode

-0.60

 Италијани

-0.57

 يتيمه

-0.56

цезда

-0.54

SequentialGroup

-0.54

 nahilalakip

-0.53

POSITIVE LOGITS

IntoConstraints

0.72

InteropServices

0.49

 ideal

0.46

ideal

0.46

RectangleBorder

0.40

Ideal

0.39

Essentials

0.39

 IDEAL

0.39

 Ideal

0.39

UpInside

0.39

Activations Density 0.029%

structured data representations with specific formatting and layout attributes

No Comments

No Known Activations