INDEX

Explanations

terms related to scientific or technical concepts including words ending in '-ation', such as "elimination" or "integration"

oai_token-act-pair · gpt-3.5-turbo

words related to processes, methods, and systemic actions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

herty

-0.78

tiny

-0.62

 cultured

-0.60

 substantive

-0.60

ãĥ£

-0.60

roc

-0.58

 ranc

-0.57

bold

-0.56

 Citiz

-0.56

 predec

-0.56

POSITIVE LOGITS

ary

1.23

naire

0.95

hower

0.89

ism

0.89

xual

0.89

ist

0.87

aries

0.87

ally

0.87

ateral

0.82

ality

0.79

Activations Density 0.129%

terms related to scientific or technical concepts including words ending in '-ation', such as "elimination" or "integration"

words related to processes, methods, and systemic actions

No Comments

No Known Activations