INDEX

Explanations

phrases related to time duration, specifically focusing on the second half

oai_token-act-pair · gpt-3.5-turbo

references to specific time periods, particularly the second half of events or periods

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

Reviewed

-0.72

incinn

-0.66

urat

-0.66

apt

-0.65

 confir

-0.65

pload

-0.64

inates

-0.62

mercial

-0.61

Hop

-0.61

 notor

-0.61

POSITIVE LOGITS

pipe

0.85

enium

0.78

tone

0.70

âĢĳ

0.68

Cec

0.67

 Engels

0.67

nings

0.66

Beck

0.64

 Pablo

0.64

mortem

0.64

Activations Density 0.012%

phrases related to time duration, specifically focusing on the second half

references to specific time periods, particularly the second half of events or periods

No Comments

No Known Activations