INDEX

Explanations

mentions of specific years between 1999 and 2001

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

-0.71

senal

-0.68

warts

-0.67

aste

-0.65

artifacts

-0.64

netflix

-0.62

venge

-0.61

 indisp

-0.61

 stead

-0.61

semble

-0.61

POSITIVE LOGITS

istar

0.81

mAh

0.79

å¹

0.75

 ãĤµãĥ¼ãĥĨãĤ£ãĥ¯ãĥ³

0.73

visors

0.71

rpm

0.71

ãĤ¦ãĤ¹

0.70

ixties

0.69

sburg

0.69

 onwards

0.68

Activations Density 0.026%

mentions of specific years between 1999 and 2001

No Comments

No Known Activations