INDEX

Explanations

mentions of specific event locations or events

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_fs6144-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_6144

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

6,144

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

 GOODMAN

-0.83

lihood

-0.75

acious

-0.68

 IMAGES

-0.65

TING

-0.62

 guiActiveUn

-0.61

 Hungarian

-0.61

ulia

-0.60

 Trou

-0.59

ICAN

-0.59

POSITIVE LOGITS

ffee

1.28

verage

1.17

xon

1.17

aching

1.13

ffe

1.07

aches

1.07

operative

1.07

asters

1.05

pper

1.05

herent

1.03

Activations Density 0.099%

mentions of specific event locations or events

No Comments

No Known Activations