INDEX

Explanations

mentions of medication dosages

oai_token-act-pair · gpt-3.5-turbo

references to dosage in medical or pharmaceutical contexts

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

Rail

-0.67

 Borders

-0.66

auga

-0.64

inates

-0.64

atters

-0.63

sections

-0.63

penter

-0.63

 Archives

-0.62

relations

-0.62

American

-0.61

POSITIVE LOGITS

 dose

1.14

 doses

1.02

 dosage

0.85

 regimen

0.83

 orally

0.79

 aspirin

0.76

 weights

0.76

 ingested

0.74

 laced

0.73

inh

0.69

Activations Density 0.012%

mentions of medication dosages

references to dosage in medical or pharmaceutical contexts

No Comments

No Known Activations