INDEX

Explanations

references to weight or measurements

oai_token-act-pair · gpt-3.5-turbo

expressions related to weight measurements

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

issance

-0.79

cknow

-0.76

choes

-0.73

opol

-0.73

eday

-0.73

actionDate

-0.72

 WARN

-0.70

istically

-0.69

vironment

-0.69

 externalToEVAOnly

-0.68

POSITIVE LOGITS

 sterling

1.10

 gorilla

1.00

 weight

0.91

 heavier

0.91

cake

0.90

lbs

0.88

lbs

0.86

 overweight

0.83

 pounds

0.82

Weight

0.81

Activations Density 0.042%

references to weight or measurements

expressions related to weight measurements

No Comments

No Known Activations