INDEX

Explanations

instances where people or organizations are being notified about something

oai_token-act-pair · gpt-3.5-turbo

repeated mentions of the term "notified" related to various incidents and authorities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

ingu

-0.72

puff

-0.68

racted

-0.64

nic

-0.62

 vain

-0.61

bers

-0.60

 Huck

-0.59

 wielding

-0.59

 torch

-0.59

serving

-0.59

POSITIVE LOGITS

isance

1.03

ĵĺ

0.84

 subscribers

0.81

 Subscribe

0.77

£ı

0.77

wcs

0.75

daq

0.74

 notify

0.73

ingly

0.73

 beforehand

0.72

Activations Density 0.034%

instances where people or organizations are being notified about something

repeated mentions of the term "notified" related to various incidents and authorities

No Comments

No Known Activations