INDEX

Explanations

structured data related to classifications and categories

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Configuration

google/gemma-scope-9b-pt-res/layer_20/width_131k/average_l0_114

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

MigrationBuilder

-0.56

 препратки

-0.54

queryInterface

-0.52

Autoritní

-0.52

AndEndTag

-0.51

tagHelperRunner

-0.50

postsleuth

-0.49

 NSError

-0.44

TagMode

-0.43

ArrowToggle

-0.43

POSITIVE LOGITS

новниш

0.50

此

0.44

 فريبيس

0.44

Étymologie

0.42

kwds

0.42

glieder

0.40

RUE

0.38

 conglomer

0.38

ształ

0.37

Tracce

0.37

Activations Density 0.230%

structured data related to classifications and categories

No Comments

No Known Activations