INDEX

Explanations

words related to cutting or reducing in some way

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_fs1536-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_1536

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

1,536

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Organisation

-0.66

 Monitor

-0.66

 Refresh

-0.65

 Dunham

-0.63

Purchase

-0.61

 Kamp

-0.61

 Moder

-0.59

ADRA

-0.58

 Franklin

-0.58

 Global

-0.58

POSITIVE LOGITS

thro

1.28

scenes

1.27

scene

1.25

aneous

1.24

lasses

1.16

lass

0.98

cuts

0.94

tle

0.93

eness

0.93

cheon

0.91

Activations Density 2.173%

words related to cutting or reducing in some way

No Comments

No Known Activations