INDEX

Explanations

expressions of regret or hindsight, where the subject reflects on what they should have done differently

oai_token-act-pair · gpt-3.5-turbo

phrases indicating regret or missed opportunities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

ocol

-0.70

 enchant

-0.61

 muse

-0.58

of

-0.58

opian

-0.57

 waning

-0.56

Mem

-0.55

 hostage

-0.55

orship

-0.53

 Seek

-0.53

POSITIVE LOGITS

 been

1.16

been

1.16

 gotten

1.14

gotten

1.00

 listened

0.88

 waited

0.88

 intervened

0.86

 gone

0.84

 Been

0.83

 behaved

0.83

Activations Density 0.059%

expressions of regret or hindsight, where the subject reflects on what they should have done differently

phrases indicating regret or missed opportunities

No Comments

No Known Activations