INDEX

Explanations

phrases related to the city of Tampa Bay and its sports teams, particularly the Buccaneers

oai_token-act-pair · gpt-3.5-turbo

mentions of the Tampa Bay area and its sports teams

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

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

theless

-0.93

naire

-0.87

places

-0.83

Ô

-0.76

acerb

-0.75

Ü

-0.72

ASC

-0.72

çĦ

-0.70

ynt

-0.69

ettings

-0.68

POSITIVE LOGITS

Bay

1.32

 Tribune

1.02

pole

1.02

 Tampa

0.95

ignt

0.92

bay

0.88

 Buccaneers

0.87

 Rays

0.83

Bay

0.81

 Bucs

0.78

Activations Density 0.012%

phrases related to the city of Tampa Bay and its sports teams, particularly the Buccaneers

mentions of the Tampa Bay area and its sports teams

No Comments

No Known Activations