INDEX

Explanations

mentions of physical units of measurement

oai_token-act-pair · gpt-3.5-turbo

references to "units" in the context of measurement or quantity

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

vous

-0.83

romeda

-0.76

é»Ĵ

-0.74

 Sloan

-0.71

Ark

-0.69

 Shack

-0.68

CHO

-0.68

=-=-

-0.67

bone

-0.67

opard

-0.66

POSITIVE LOGITS

 units

0.90

imet

0.90

mate

0.87

mates

0.81

rity

0.78

 unit

0.75

matically

0.74

iations

0.74

imately

0.73

itatively

0.72

Activations Density 0.013%

mentions of physical units of measurement

references to "units" in the context of measurement or quantity

No Comments

No Known Activations