INDEX

Explanations

negations and negative phrases

oai_token-act-pair · gpt-3.5-turbo

negations and expressions of dissatisfaction

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

Not in Any Lists

No Comments

Negative Logits

iferation

-0.70

ashtra

-0.70

cens

-0.64

ructose

-0.63

oos

-0.63

aston

-0.63

ousy

-0.63

uders

-0.62

itton

-0.62

but

-0.60

POSITIVE LOGITS

 necessarily

0.85

etheless

0.83

 nonetheless

0.81

 guarantee

0.81

 conclusive

0.79

 guarantees

0.73

 concrete

0.70

 enough

0.70

 infall

0.66

 convin

0.66

Activations Density 0.326%

negations and negative phrases

negations and expressions of dissatisfaction

No Comments

No Known Activations