INDEX

Explanations

proper nouns such as names of locations and organizations

oai_token-act-pair · gpt-3.5-turbo

periods and punctuation marks

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

hiba

-0.78

eatures

-0.72

 eleph

-0.68

cius

-0.67

 advis

-0.66

 fulfillment

-0.66

 appropriate

-0.66

 refres

-0.65

idious

-0.63

rimp

-0.62

POSITIVE LOGITS

 Pool

0.78

McC

0.77

0.76

 PARK

0.73

 Pryor

0.71

MX

0.71

 Mellon

0.70

 Tribe

0.68

orks

0.68

Activations Density 0.027%

proper nouns such as names of locations and organizations

periods and punctuation marks

No Comments

No Known Activations