INDEX

Explanations

phrases related to wire and fraud

oai_token-act-pair · gpt-3.5-turbo

references to wire-related topics or concepts

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ivia

-0.81

eston

-0.70

BIP

-0.70

 Flav

-0.65

 Scotia

-0.65

 contestant

-0.65

 Fraz

-0.63

anchester

-0.63

 Taste

-0.62

CVE

-0.62

POSITIVE LOGITS

lessly

1.25

tap

1.13

frame

1.08

fram

1.07

frames

0.95

wire

0.94

kefeller

0.83

 wire

0.83

haired

0.83

OGR

0.82

Activations Density 0.017%

phrases related to wire and fraud

references to wire-related topics or concepts

No Comments

No Known Activations