INDEX

Explanations

phrases related to how individuals are treated or perceived in different situations

oai_token-act-pair · gpt-3.5-turbo

phrases related to treatment or how individuals are treated in various situations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 obser

-0.90

 confir

-0.76

 Archdemon

-0.71

 sugg

-0.70

oln

-0.70

 Rivals

-0.69

uthor

-0.68

 endorsements

-0.65

 endorsement

-0.65

ģĸ

-0.62

POSITIVE LOGITS

 differently

0.85

 sick

0.72

 mentally

0.67

 critically

0.66

 Orient

0.66

 wounds

0.64

 religiously

0.62

utic

0.60

 wounded

0.59

 scant

0.59

Activations Density 0.177%

phrases related to how individuals are treated or perceived in different situations

phrases related to treatment or how individuals are treated in various situations

No Comments

No Known Activations