INDEX

Explanations

updates or clarifications in news articles

oai_token-act-pair · gpt-3.5-turbo

references to specific dates and events

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

often

-0.82

Often

-0.76

 nurture

-0.73

Imagine

-0.72

²¾

-0.72

ĸļ

-0.71

 stereotypical

-0.71

pires

-0.71

 often

-0.70

Years

-0.67

POSITIVE LOGITS

 UPDATE

1.21

UPDATE

1.20

Update

1.09

 confirms

1.08

 confirming

1.07

PDATED

1.06

 updated

1.03

!]

1.01

 corrected

1.00

 confirmed

0.99

Activations Density 0.598%

updates or clarifications in news articles

references to specific dates and events

No Comments

No Known Activations