INDEX

Explanations

terms related to health and medical specialties, specifically "gynecology."

oai_token-act-pair · gpt-3.5-turbo

references to gynecology or related medical fields

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

ACTED

-0.93

ortium

-0.88

éĹĺ

-0.82

 Hurricanes

-0.74

 Wikileaks

-0.72

angered

-0.70

 Austral

-0.69

ional

-0.69

incial

-0.68

phabet

-0.67

POSITIVE LOGITS

gy

1.38

Gy

1.23

Gy

0.95

 corrid

0.94

psy

0.92

rosc

0.87

romy

0.82

Ĥª

0.75

mpire

0.74

¼

0.73

Activations Density 0.004%

terms related to health and medical specialties, specifically "gynecology."

references to gynecology or related medical fields

No Comments

No Known Activations