INDEX

Explanations

references to the beginning or start of something

oai_token-act-pair · gpt-3.5-turbo

instances of the word "opening."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

ieving

-0.92

tics

-0.90

cia

-0.88

rior

-0.79

lean

-0.76

cles

-0.74

zin

-0.72

rome

-0.72

lust

-0.71

cci

-0.70

POSITIVE LOGITS

 ceremonies

0.86

 Doors

0.82

 ceremony

0.78

 doors

0.78

 salv

0.77

 opening

0.74

 weekend

0.73

 crawl

0.73

 openings

0.71

 Weekend

0.71

Activations Density 0.024%

references to the beginning or start of something

instances of the word "opening."

No Comments

No Known Activations