INDEX

Explanations

mentions of the name "Milo" at varying activations

oai_token-act-pair · gpt-3.5-turbo

occurrences of the term "Milo"

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.4.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

marked

-0.80

race

-0.70

paren

-0.69

mark

-0.68

master

-0.68

marks

-0.67

gradient

-0.67

lain

-0.67

groups

-0.66

frames

-0.65

POSITIVE LOGITS

cean

1.11

Å¡

1.03

zzi

0.95

 Princ

0.94

fty

0.92

quit

0.86

zar

0.85

quet

0.84

zek

0.84

leans

0.82

Activations Density 0.008%

mentions of the name "Milo" at varying activations

occurrences of the term "Milo"

No Comments

No Known Activations