INDEX

Explanations

statistical terms and mathematical expressions

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_fs1536-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_1536

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

1,536

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

arella

-0.72

etsk

-0.68

eus

-0.68

 loophole

-0.63

 loopholes

-0.62

plets

-0.62

olulu

-0.61

Of

-0.60

skirts

-0.59

 volumes

-0.59

POSITIVE LOGITS

 utmost

0.73

thood

0.72

 imagination

0.67

 nutshell

0.66

 millennium

0.65

 universe

0.63

il

0.63

 Invention

0.63

 century

0.62

 Ancients

0.61

Activations Density 3.828%

statistical terms and mathematical expressions

No Comments

No Known Activations