INDEX

Explanations

mentions of items or products categorized by their quality grade

oai_token-act-pair · gpt-3.5-turbo

references to the quality or classification of drugs

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

otrop

-0.79

 Bundy

-0.68

doi

-0.67

oos

-0.65

soDeliveryDate

-0.64

acea

-0.62

 prow

-0.62

INESS

-0.60

sle

-0.60

bered

-0.59

POSITIVE LOGITS

grade

1.04

Grade

0.84

ably

0.83

 grade

0.82

grades

0.79

grad

0.78

grading

0.76

 grades

0.74

graded

0.73

able

0.72

Activations Density 0.008%

mentions of items or products categorized by their quality grade

references to the quality or classification of drugs

No Comments

No Known Activations