INDEX

Explanations

terms related to admission processes and terminology

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

IFrame

Link

Negative Logits

 kasarigan

-0.72

carolina

-0.65

errorCode

-0.56

keletal

-0.54

Yann

-0.54

Pyrene

-0.54

 Yann

-0.52

 Bourke

-0.52

 Funky

-0.52

山崎

-0.52

POSITIVE LOGITS

 admission

0.93

 Admission

0.91

Admission

0.79

admission

0.71

 Admit

0.71

 admit

0.70

 admitted

0.66

 admittance

0.63

 Admissions

0.62

 admissions

0.60

Activations Density 0.012%

terms related to admission processes and terminology

No Known Activations