INDEX

Explanations

words related to forgiveness

oai_token-act-pair · gpt-3.5-turbo

words related to forgetting or failing to remember, as well as terms associated with legal or financial consequences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

NESS

-0.68

 athletes

-0.67

 IMAGES

-0.67

 detachment

-0.66

 feder

-0.65

 negotiators

-0.64

GROUP

-0.64

 VICE

-0.63

RB

-0.62

 fury

-0.61

POSITIVE LOGITS

veland

0.98

otton

0.89

ave

0.87

ãģ®éŃĶ

0.85

avis

0.83

andra

0.81

ames

0.80

 forg

0.79

hetto

0.79

omal

0.78

Activations Density 0.018%

words related to forgiveness

words related to forgetting or failing to remember, as well as terms associated with legal or financial consequences

No Comments

No Known Activations