INDEX

Explanations

words related to leadership and political figures

oai_token-act-pair · gpt-3.5-turbo

repeated phrases that emphasize the state of being or conditions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

ranged

-0.69

luaj

-0.63

ribe

-0.63

die

-0.62

Finish

-0.60

kee

-0.60

opathic

-0.60

ire

-0.59

 bearer

-0.59

oji

-0.57

POSITIVE LOGITS

 unclear

1.45

 doubtful

1.09

 unsur

1.04

 ironic

1.03

 speculated

0.98

 understandable

0.92

 uncertain

0.91

 unlikely

0.91

 conceivable

0.88

 clear

0.88

Activations Density 0.181%

words related to leadership and political figures

repeated phrases that emphasize the state of being or conditions

No Comments

No Known Activations