INDEX

Explanations

references to news coverage or reporting

oai_token-act-pair · gpt-3.5-turbo

phrases related to journalistic coverage of 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

Negative Logits

 teasp

-0.94

Timer

-0.74

ĸļ

-0.71

imei

-0.69

TRY

-0.69

terness

-0.69

Iterator

-0.69

 exting

-0.68

©¶æ¥µ

-0.67

externalActionCode

-0.66

POSITIVE LOGITS

 developments

1.04

 scandals

1.02

 Benghazi

0.99

 controversies

0.94

hillary

0.88

stories

0.85

 Hurricane

0.85

 debates

0.83

 Watergate

0.83

 Ebola

0.82

Activations Density 0.308%

references to news coverage or reporting

phrases related to journalistic coverage of events

No Known Activations