INDEX

Explanations

words with prefixes indicating negation or opposition

oai_token-act-pair · gpt-3.5-turbo

sections that indicate a presence or absence of specific emphatic or critical elements in the text

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

ulhu

-0.80

 Doodle

-0.70

Wiz

-0.69

 modification

-0.66

AAC

-0.66

 badges

-0.65

 Rica

-0.65

 Boys

-0.65

 ranks

-0.65

 Barker

-0.63

POSITIVE LOGITS

perfect

1.25

zero

1.16

sighted

1.16

const

1.15

surface

1.15

term

1.13

average

1.11

exclusive

1.09

distance

1.07

daily

1.06

Activations Density 0.023%

words with prefixes indicating negation or opposition

sections that indicate a presence or absence of specific emphatic or critical elements in the text

No Comments

No Known Activations