INDEX

Explanations

numbers and abbreviations related to specific topics or industries

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

 Bulg

-0.66

 Hera

-0.63

 Bris

-0.63

Nun

-0.62

 Remem

-0.62

Io

-0.61

terday

-0.61

 constitu

-0.60

ãĤ¡

-0.59

 Isles

-0.59

POSITIVE LOGITS

etermin

1.36

ividual

1.29

etermined

1.27

etermination

1.26

ownt

1.23

ynam

1.23

imensional

1.21

olph

1.19

iverse

1.19

rones

1.19

Activations Density 0.337%

numbers and abbreviations related to specific topics or industries

No Comments

No Known Activations