INDEX

Explanations

statements regarding medical treatments and their effects

oai_token-act-pair · gpt-3.5-turbo

keywords related to medical treatment efficacy and clinical trial results

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

orno

-0.83

vironment

-0.81

é»Ĵ

-0.69

ycle

-0.69

 Verse

-0.67

 Taxes

-0.67

luaj

-0.67

cean

-0.67

iery

-0.66

elaide

-0.66

POSITIVE LOGITS

 efficacy

1.52

 remission

1.35

 antidepressant

1.23

 placebo

1.22

 effectiveness

1.17

 tolerated

1.16

 effic

1.16

 therapeutic

1.15

 pharmac

1.14

 atten

1.12

Activations Density 0.154%

statements regarding medical treatments and their effects

keywords related to medical treatment efficacy and clinical trial results

No Comments

No Known Activations