INDEX

Explanations

expressions of gratitude

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

iche

-0.68

conserv

-0.68

å¥³

-0.63

 projected

-0.62

 projecting

-0.62

place

-0.61

atform

-0.61

inese

-0.60

inance

-0.55

avis

-0.55

POSITIVE LOGITS

giving

1.39

 goodness

0.89

SG

0.89

ESCO

0.88

 heavens

0.88

 gracious

0.86

rats

0.82

gements

0.81

god

0.79

 acknowled

0.79

Activations Density 0.597%

expressions of gratitude

No Comments

No Known Activations