INDEX

Explanations

noun phrases related to medication

oai_token-act-pair · gpt-3.5-turbo

references to prescribed medication

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

bread

-0.76

İ

-0.75

 Horn

-0.74

 Democr

-0.69

ECD

-0.68

 Eliot

-0.68

 Hollande

-0.68

bour

-0.68

estamp

-0.66

 Guth

-0.66

POSITIVE LOGITS

 medication

1.38

 medications

1.32

 pills

1.03

 dosage

1.02

 regimen

1.01

 drugs

0.99

 overdose

0.99

 prescriptions

0.95

 treatments

0.93

 anesthesia

0.92

Activations Density 0.015%

noun phrases related to medication

references to prescribed medication

No Comments

No Known Activations