INDEX

Explanations

scientific and medical terms, including statistics and research findings

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

soDeliveryDate

-0.70

orld

-0.67

 Vaugh

-0.67

ufact

-0.66

 thous

-0.62

 loopholes

-0.59

gallery

-0.59

 princ

-0.59

 Franch

-0.58

Resources

-0.58

POSITIVE LOGITS

 antidepressant

1.00

 placebo

0.97

 biomark

0.96

 allele

0.94

 clinically

0.88

 (âĪĴ

0.88

 hippocamp

0.87

 cortisol

0.86

 metabolic

0.85

 dietary

0.85

Activations Density 0.864%

scientific and medical terms, including statistics and research findings

No Comments

No Known Activations