INDEX

Explanations

references to specific years

oai_token-act-pair · gpt-3.5-turbo

references to the year 2000

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

atories

-0.74

warts

-0.71

akening

-0.67

senal

-0.66

bon

-0.66

aste

-0.64

Else

-0.64

venge

-0.64

atory

-0.64

alth

-0.63

POSITIVE LOGITS

mAh

0.81

 ãĤµãĥ¼ãĥĨãĤ£ãĥ¯ãĥ³

0.79

istar

0.78

visors

0.75

rpm

0.74

 Leban

0.70

ixties

0.69

rpm

0.68

Hz

0.68

hz

0.67

Activations Density 0.012%

references to specific years

references to the year 2000

No Comments

No Known Activations