INDEX

Explanations

years in the 2000s mentioned in references

oai_token-act-pair · gpt-3.5-turbo

references to publication years in academic citations

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

imer

-0.73

pee

-0.71

mir

-0.70

pees

-0.64

unal

-0.64

Bay

-0.64

isec

-0.63

ype

-0.63

ety

-0.63

venge

-0.63

POSITIVE LOGITS

å¹

1.08

 Edition

0.72

 EDITION

0.70

é¾

0.70

 Diversity

0.69

 Highlights

0.69

 doctoral

0.69

-)

0.68

)—

0.67

UX

0.65

Activations Density 0.067%

years in the 2000s mentioned in references

references to publication years in academic citations

No Comments

No Known Activations