INDEX

Explanations

mentions of the term "Roman"

oai_token-act-pair · gpt-3.5-turbo

references to Roman history and culture

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

intosh

-1.11

mble

-0.96

ramid

-0.85

NetMessage

-0.85

*/(

-0.83

olulu

-0.83

ilater

-0.80

razil

-0.80

lessly

-0.79

lesh

-0.78

POSITIVE LOGITS

 Catholic

1.01

 Catholicism

0.86

 Reign

0.85

 Torch

0.84

 numer

0.82

 Catholics

0.79

 Inquisition

0.78

 Roman

0.77

 Pont

0.75

 Emperor

0.75

Activations Density 0.012%

mentions of the term "Roman"

references to Roman history and culture

No Comments

No Known Activations