INDEX

Explanations

terms related to medical treatments and experimental conditions

oai_token-act-pair · gpt-3.5-turbo

references to placebo effects in medical contexts

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

hani

-0.73

orian

-0.71

dar

-0.70

laws

-0.68

Allen

-0.67

lar

-0.67

odes

-0.66

 creditors

-0.65

ORN

-0.64

don

-0.64

POSITIVE LOGITS

 placebo

1.11

 analges

0.84

veyard

0.84

nesday

0.72

 acupuncture

0.71

 aspirin

0.69

lette

0.68

 mosqu

0.68

aneously

0.67

 augment

0.66

Activations Density 0.012%

terms related to medical treatments and experimental conditions

references to placebo effects in medical contexts

No Comments

No Known Activations