INDEX

Explanations

words related to a specific entity, likely named "Host"

oai_token-act-pair · gpt-3.5-turbo

instances of the word "host" and its variations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

-0.89

anguage

-0.82

Downloadha

-0.73

utic

-0.67

Oro

-0.66

ucket

-0.66

inois

-0.65

 Imper

-0.64

-0.63

utherford

-0.60

POSITIVE LOGITS

ilities

1.06

Host

1.03

ility

0.96

 Host

0.92

host

0.88

ship

0.88

esses

0.85

named

0.83

0.82

strate

0.80

Activations Density 0.008%

words related to a specific entity, likely named "Host"

instances of the word "host" and its variations

No Comments

No Known Activations