INDEX

Explanations

keywords related to numerical data such as percentages and measurements

oai_token-act-pair · gpt-3.5-turbo

currency values represented in cents

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

urally

-0.85

ultan

-0.77

icio

-0.77

urned

-0.75

accompanied

-0.74

population

-0.71

ographically

-0.71

ependence

-0.70

essa

-0.70

strate

-0.69

POSITIVE LOGITS

¹

0.77

Â¢

0.75

ï¸

0.75

ï¸ı

0.75

TION

0.74

³

0.73

·

0.72

ternity

0.69

ATH

0.69

GST

0.68

Activations Density 0.015%

keywords related to numerical data such as percentages and measurements

currency values represented in cents

No Comments

No Known Activations