INDEX

Explanations

proper nouns or technical terms

oai_token-act-pair · gpt-3.5-turbo

phrases that introduce or refer to specific entities or concepts that are recognized or labeled

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

odder

-0.88

ertation

-0.81

erva

-0.80

rentice

-0.80

ermanent

-0.78

ŃĶ

-0.76

ixels

-0.75

orsche

-0.75

plet

-0.75

psey

-0.75

POSITIVE LOGITS

Ð»

0.83

ties

0.74

ledged

0.74

cut

0.73

ãĤ¤

0.72

cuts

0.71

itarian

0.70

 Î¼

0.68

 quantity

0.67

 abbrevi

0.66

Activations Density 0.030%

proper nouns or technical terms

phrases that introduce or refer to specific entities or concepts that are recognized or labeled

No Comments

No Known Activations