INDEX

Explanations

locations or places

oai_token-act-pair · gpt-3.5-turbo

occurrences of the letter "d"

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

=-=-

-0.67

 Corvette

-0.65

 Remem

-0.65

CRIP

-0.64

å§«

-0.62

 adversity

-0.62

 recy

-0.61

WAYS

-0.61

OIL

-0.59

 constitu

-0.58

POSITIVE LOGITS

ynamic

1.26

ividual

1.26

aniel

1.20

orf

1.19

yssey

1.16

iamond

1.15

ynam

1.15

elta

1.12

roid

1.09

ynasty

1.09

Activations Density 0.045%

locations or places

occurrences of the letter "d"

No Comments

No Known Activations