INDEX

Explanations

phrases related to advertising and marketing

oai_token-act-pair · gpt-3.5-turbo

references to advertisements and their impact on behavior or perception

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

thal

-0.89

 goalt

-0.71

 droid

-0.68

hug

-0.66

 favor

-0.64

 Saiyan

-0.63

Override

-0.63

 honored

-0.63

 vacation

-0.63

 mete

-0.62

POSITIVE LOGITS

 Whilst

1.26

 Ulster

1.11

 Firstly

1.05

 Labour

0.95

 UKIP

0.94

 Alternatively

0.93

NHS

0.92

NZ

0.91

Scotland

0.91

 Miliband

0.91

Activations Density 1.710%

phrases related to advertising and marketing

references to advertisements and their impact on behavior or perception

No Comments

No Known Activations