INDEX

Explanations

numerical values with 'x' as a unit of measurement

oai_token-act-pair · gpt-3.5-turbo

references to magnification levels and resolutions in visual or imaging contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

 Aval

-0.73

kson

-0.71

 TODAY

-0.69

 abiding

-0.69

 Garry

-0.67

ogy

-0.66

 strut

-0.65

 Verge

-0.64

ogical

-0.63

sburgh

-0.62

POSITIVE LOGITS

digit

1.02

1.00

0.99

0.98

LP

0.96

0.94

0.93

0.92

0.91

Activations Density 0.017%

numerical values with 'x' as a unit of measurement

references to magnification levels and resolutions in visual or imaging contexts

No Comments

No Known Activations