INDEX

Explanations

terms related to programming constructs and data processing

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

esModule

-0.46

setcounter

-0.43

WebServlet

-0.43

 setuptools

-0.43

URLConnection

-0.41

RTDA

-0.41

 AttributeSet

-0.40

 CreateTagHelper

-0.40

prepareStatement

-0.38

+#+

-0.38

POSITIVE LOGITS

LabelTagHelper

0.50

 goal

0.43

 activities

0.42

 lenker

0.42

 events

0.42

 interactions

0.41

 goals

0.40

 barrios

0.40

tanleria

0.40

 parís

0.40

Activations Density 0.065%

terms related to programming constructs and data processing

No Comments

No Known Activations