INDEX

Explanations

words related to programming and technical implementation details

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

 Bibliothek

-0.40

untung

-0.38

APPLAUSE

-0.37

 rayures

-0.35

 Umständen

-0.34

 Vorstellung

-0.34

WriteTagHelper

-0.33

 rağmen

-0.33

 tatuajes

-0.32

 chrétien

-0.32

POSITIVE LOGITS

 насељу

0.68

+#+#

0.62

 springfox

0.58

 propOrder

0.55

/*

0.54

HtmlAttribute

0.52

 дописавши

0.52

 tartalomajánló

0.51

intellij

0.51

adaptiveStyles

0.51

Activations Density 0.006%

words related to programming and technical implementation details

No Comments

No Known Activations