INDEX

Explanations

medical terms and treatment-related concepts

oai_token-act-pair · gpt-3.5-turbo

scientific terms and phrases related to medical research and effects on subjects

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

soDeliveryDate

-0.87

 looms

-0.73

vernment

-0.70

Ü

-0.66

 Citizen

-0.65

 commons

-0.65

////////////////////////////////

-0.63

claimer

-0.63

udget

-0.63

arium

-0.62

POSITIVE LOGITS

 decreased

1.07

RESULTS

1.05

 placebo

0.97

 (âĪĴ

0.92

 remission

0.88

 significantly

0.88

 metabolites

0.88

 correlated

0.86

Compared

0.86

 persisted

0.85

Activations Density 0.521%

medical terms and treatment-related concepts

scientific terms and phrases related to medical research and effects on subjects

No Comments

No Known Activations