INDEX

Explanations

words related to physical health and condition

oai_token-act-pair · gpt-3.5-turbo

descriptors of physical well-being and health status

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

atars

-0.70

utters

-0.70

avorite

-0.70

arching

-0.69

eeper

-0.63

rench

-0.59

 Badge

-0.59

 Ging

-0.59

rap

-0.59

Ban

-0.59

POSITIVE LOGITS

 Dragonbound

0.90

 resembling

0.71

nered

0.70

 infancy

0.68

 handwriting

0.68

 financially

0.67

arial

0.66

igue

0.66

mia

0.66

manship

0.65

Activations Density 0.075%

words related to physical health and condition

descriptors of physical well-being and health status

No Comments

No Known Activations