INDEX

Explanations

ways to detect various phenomena or conditions, especially in the context of medical diagnostics or technology

oai_token-act-pair · gpt-3.5-turbo

terms related to detection and identifying specific conditions or entities

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

perty

-0.77

mort

-0.76

creation

-0.74

emate

-0.72

nesia

-0.67

 Vide

-0.66

skill

-0.65

erer

-0.65

)=(

-0.64

erous

-0.63

POSITIVE LOGITS

ibly

0.87

 detect

0.86

ively

0.85

ibles

0.82

ible

0.82

 detects

0.78

wana

0.73

ibility

0.73

 detection

0.72

 detected

0.70

Activations Density 0.028%

ways to detect various phenomena or conditions, especially in the context of medical diagnostics or technology

terms related to detection and identifying specific conditions or entities

No Comments

No Known Activations