INDEX

Explanations

references to technical protocols, specifically focusing on the word "protocol" and related technical terms

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

adows

-0.79

inav

-0.73

inning

-0.72

inese

-0.71

inez

-0.71

some

-0.69

reens

-0.68

ahar

-0.68

hiro

-0.67

ying

-0.66

POSITIVE LOGITS

 protocol

1.16

 protocols

1.10

ocol

1.03

eers

1.01

 Protocol

0.95

 handshake

0.88

eer

0.82

ĨĴ

0.77

tcp

0.74

ually

0.73

Activations Density 0.019%

references to technical protocols, specifically focusing on the word "protocol" and related technical terms

No Comments

No Known Activations