INDEX

Explanations

locations or references related to Seattle

oai_token-act-pair · gpt-3.5-turbo

references to Seattle and its geographical context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

 Rouhani

-1.21

 Gaddafi

-1.16

 Allaah

-1.10

 Jindal

-1.06

Iran

-1.04

 Libyan

-1.04

 Tehran

-1.01

 Giul

-0.99

 Hispan

-0.97

 Tunis

-0.96

POSITIVE LOGITS

 Seattle

1.89

Seattle

1.88

 Tacoma

1.83

 Spokane

1.63

 Vancouver

1.43

Portland

1.43

 Portland

1.43

WA

1.39

Pug

1.39

 Bellev

1.36

Activations Density 0.228%

locations or references related to Seattle

references to Seattle and its geographical context

No Comments

No Known Activations