INDEX

Explanations

key points, highlights, or articles in a story

oai_token-act-pair · gpt-3.5-turbo

content related to news highlights and key points

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

ogly

-0.79

 nurs

-0.73

tails

-0.73

abase

-0.67

 independ

-0.66

olesc

-0.65

 toget

-0.65

ŃĶ

-0.65

ĪĴ

-0.65

ãĤ¦ãĤ¹

-0.65

POSITIVE LOGITS

ï

0.77

 Thousands

0.75

 Authorities

0.74

 Surveillance

0.73

 Timeline

0.73

 Kaine

0.73

 Comed

0.72

 Researchers

0.71

 Highlights

0.71

 Investigators

0.71

Activations Density 0.010%

key points, highlights, or articles in a story

content related to news highlights and key points

No Comments

No Known Activations