INDEX

Explanations

phrases related to academic or professional affiliations, particularly at institutions or organizations

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "the"

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

 engulf

-0.77

 rever

-0.76

 ensued

-0.73

rats

-0.72

 preceded

-0.69

 segreg

-0.69

blow

-0.68

 transpired

-0.68

 substit

-0.68

 attempting

-0.67

POSITIVE LOGITS

 Institute

1.36

 University

1.23

 Center

1.21

 National

1.16

 Brookings

1.15

 Department

1.09

 Smithsonian

1.07

 Centers

1.06

 Centre

1.06

 Association

1.01

Activations Density 0.218%

phrases related to academic or professional affiliations, particularly at institutions or organizations

occurrences of the word "the"

No Comments

No Known Activations