INDEX

Explanations

references to advertising or the act of advertisement

oai_token-act-pair · gpt-3.5-turbo

terms related to advertising

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

 incomplete

-0.66

mt

-0.65

 prepared

-0.64

Kal

-0.64

 modular

-0.64

OC

-0.62

 hastily

-0.62

SC

-0.62

 widow

-0.62

 wood

-0.61

POSITIVE LOGITS

vertising

4.54

vertisements

1.88

vertisement

1.78

vertis

1.70

advertising

1.45

 Advertising

1.42

 Advertisement

1.25

 advertising

1.14

ibaba

1.07

VERTISEMENT

1.06

Activations Density 0.018%

references to advertising or the act of advertisement

terms related to advertising

No Comments

No Known Activations