INDEX

Explanations

instances of the word or variations of 'watch'

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

esan

-0.82

ãĥ´

-0.77

interstitial

-0.76

eno

-0.72

wealth

-0.70

enos

-0.68

enture

-0.66

phi

-0.65

hesion

-0.62

 reusable

-0.62

POSITIVE LOGITS

dogs

1.26

dog

1.19

tower

1.11

 Watching

0.96

 helpless

0.94

 videos

0.93

 unfold

0.89

tv

0.85

 clips

0.83

 watching

0.80

Activations Density 0.199%

instances of the word or variations of 'watch'

No Comments

No Known Activations