INDEX

Explanations

words related to medical terms and conditions

oai_token-act-pair · gpt-3.5-turbo

terms associated with negative or harmful impacts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

 Subtle

-0.78

 Lack

-0.72

 Izan

-0.71

 Lines

-0.70

 Cheong

-0.67

Ll

-0.67

 Chap

-0.66

 Traff

-0.66

 Hurt

-0.66

 Accountability

-0.65

POSITIVE LOGITS

omorph

0.93

id

0.88

odon

0.88

ouse

0.86

arious

0.85

ouses

0.85

ophobic

0.85

otype

0.85

astic

0.85

ect

0.85

Activations Density 0.252%

words related to medical terms and conditions

terms associated with negative or harmful impacts

No Comments

No Known Activations