INDEX

Explanations

phrases that start with 'San Diego'

oai_token-act-pair · gpt-3.5-turbo

locations and associated details

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

 carving

-0.86

 regenerate

-0.78

 paralle

-0.75

 quitting

-0.75

 freel

-0.74

nesday

-0.73

 chewing

-0.73

 crem

-0.72

 snowball

-0.72

 underestimate

-0.72

POSITIVE LOGITS

Status

1.33

Price

1.28

Rating

1.27

Duration

1.25

Tags

1.24

Weight

1.22

Runtime

1.20

Subject

1.17

Gender

1.17

Language

1.16

Activations Density 0.221%

phrases that start with 'San Diego'

locations and associated details

No Comments

No Known Activations