INDEX

Explanations

adjectives followed by 'all'

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_fs24576-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

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

pu

-0.66

lf

-0.61

uay

-0.61

 virginity

-0.58

avorite

-0.57

 proverb

-0.57

cel

-0.57

 Ferdinand

-0.57

geries

-0.56

algia

-0.56

POSITIVE LOGITS

 expense

0.88

 levels

0.88

 angles

0.84

ocating

0.83

 times

0.82

å¸

0.78

onge

0.76

 seams

0.74

 wavelengths

0.74

 hazards

0.74

Activations Density 0.024%

adjectives followed by 'all'

No Comments

No Known Activations