INDEX

Explanations

phrases that involve negating the existence of certain concepts or items

oai_token-act-pair · gpt-3.5-turbo

phrases asserting the non-existence of certain concepts or entities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

uay

-0.80

rik

-0.71

alde

-0.70

 Flavoring

-0.70

ipedia

-0.69

olate

-0.69

igi

-0.69

venue

-0.68

osterone

-0.68

 Murd

-0.67

POSITIVE LOGITS

 thing

1.15

 luck

0.99

thing

0.82

 luxury

0.77

 fate

0.69

 fancy

0.67

 designation

0.66

 exact

0.65

 pesky

0.64

ities

0.64

Activations Density 0.040%

phrases that involve negating the existence of certain concepts or items

phrases asserting the non-existence of certain concepts or entities

No Comments

No Known Activations