INDEX

Explanations

references to Java or Apache libraries and utilities

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

 חיצוניים

-0.69

ViewFeatures

-0.61

Tikang

-0.60

ReusableCell

-0.60

 GenerationType

-0.60

 Wiktionnaire

-0.58

GeneratedValue

-0.57



-0.57

 EconPapers

-0.56

RenderAtEndOf

-0.55

POSITIVE LOGITS

autogui

0.38

 judgment

0.36

Jeografia

0.35

 judge

0.35

quis

0.34

レット

0.34

 Mittelalter

0.33

 medieval

0.32

judgment

0.32

 forensic

0.31

Activations Density 0.002%

references to Java or Apache libraries and utilities

No Comments

No Known Activations