INDEX

Explanations

mentions of medication and related terms

oai_token-act-pair · gpt-3.5-turbo

references to prescribed medications

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

 Horn

-0.78

oir

-0.75

TRUMP

-0.68

EAR

-0.68

gian

-0.67

Wind

-0.67

ORE

-0.66

ENC

-0.66

vol

-0.66

rake

-0.65

POSITIVE LOGITS

 medication

1.46

 medications

1.42

 pills

1.04

 drugs

0.98

 prescriptions

0.97

 dosage

0.95

 anesthesia

0.94

 overdose

0.93

 prescribing

0.93

 regimen

0.93

Activations Density 0.014%

mentions of medication and related terms

references to prescribed medications

No Comments

No Known Activations