INDEX

Explanations

moments or instances related to strong emotions or significant 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_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

ãģķ

-0.74

ply

-0.72

ä¸Ĭ

-0.71

ãģĮ

-0.69

sheet

-0.68

pel

-0.64

inav

-0.62

GH

-0.62

GE

-0.60

 redistributed

-0.60

POSITIVE LOGITS

ary

1.14

 Moments

1.01

aneously

0.98

icals

0.95

aneous

0.93

ous

0.91

 moments

0.90

hift

0.90

icity

0.88

aries

0.84

Activations Density 0.875%

moments or instances related to strong emotions or significant events

No Comments

No Known Activations