INDEX

Explanations

phrases related to growth, development, or progression

oai_token-act-pair · gpt-3.5-turbo

instances of the word "become" and similar verbs indicating change or evolution

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

MJ

-0.70

rish

-0.68

 Later

-0.67

article

-0.64

uez

-0.64

abet

-0.62

emate

-0.60

apo

-0.60

umbn

-0.60

morrow

-0.59

POSITIVE LOGITS

 remarkably

1.15

 steadily

1.11

 lately

1.09

 tremendously

1.03

 since

1.02

 considerably

1.00

 markedly

0.96

 enormously

0.95

 leaps

0.94

 countless

0.93

Activations Density 0.372%

phrases related to growth, development, or progression

instances of the word "become" and similar verbs indicating change or evolution

No Comments

No Known Activations