INDEX

Explanations

phrases with the format "first-degree [noun]"

oai_token-act-pair · gpt-3.5-turbo

phrases containing "first" followed by various contextual references, often related to legal or significant events

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

 Canaver

-0.65

âĸĴ

-0.63

 Restrict

-0.63

âĸĵ

-0.62

 shroud

-0.62

kson

-0.61

âĶģ

-0.61

 Richards

-0.61

MENTS

-0.61

stanbul

-0.60

POSITIVE LOGITS

degree

1.15

generation

1.14

class

1.02

level

1.00

contact

0.99

century

0.96

year

0.95

order

0.95

party

0.95

hand

0.94

Activations Density 0.031%

phrases with the format "first-degree [noun]"

phrases containing "first" followed by various contextual references, often related to legal or significant events

No Comments

No Known Activations