INDEX

Explanations

mentions of official titles and positions

oai_token-act-pair · gpt-3.5-turbo

numerical values and statistics related to events

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

etheless

-0.81

halla

-0.81

poon

-0.80

 surely

-0.75

¥µ

-0.72

ãĤ´ãĥ³

-0.69

dyn

-0.67

yt

-0.66

omics

-0.65

 nowhere

-0.65

POSITIVE LOGITS

FILE

1.58

window

1.13

 Photographer

0.95

 Employees

0.93

People

0.89

 Visitors

0.87

 Protesters

0.87

 Participants

0.85

Members

0.85

 Customers

0.83

Activations Density 0.039%

mentions of official titles and positions

numerical values and statistics related to events

No Comments

No Known Activations