INDEX

Explanations

hashtags

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs3072-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_3072

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

3,072

Data Type

torch.float32

Hook Point

blocks.8.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

 Gors

-0.80

 cler

-0.74

 tabl

-0.74

 Sylv

-0.73

 foremost

-0.73

Vie

-0.73

oun

-0.73

 slips

-0.72

 Ambrose

-0.70

 destro

-0.70

POSITIVE LOGITS

define

1.37

DIV

1.26

!/

1.23

Gamer

1.22

################################

1.21

include

1.15

########

1.10

DonaldTrump

1.09

RIP

1.09

Ask

1.08

Activations Density 0.281%

hashtags

No Comments

No Known Activations