INDEX

Explanations

phrases related to medical conditions, specifically Parkinson's disease and paralysis

oai_token-act-pair · gpt-3.5-turbo

terms related to Parkinson's disease and experiences of wandering or loss of control

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

wards

-0.80

vironment

-0.80

perature

-0.73

arian

-0.70

ezvous

-0.70

mine

-0.68

nery

-0.68

orgetown

-0.68

galitarian

-0.67

boarding

-0.66

POSITIVE LOGITS

Interstitial

0.74

 Mongolia

0.70

region

0.70

govtrack

0.68

Pig

0.67

orne

0.66

airo

0.66

stretched

0.66

oup

0.65

UNCH

0.65

Activations Density 0.059%

phrases related to medical conditions, specifically Parkinson's disease and paralysis

terms related to Parkinson's disease and experiences of wandering or loss of control

No Comments

No Known Activations