INDEX

Explanations

tech-related terms and programming concepts

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

nces

-0.77

ernels

-0.73

ceans

-0.72

xes

-0.72

abouts

-0.72

opes

-0.71

atives

-0.69

riers

-0.69

agles

-0.66

ls

-0.66

POSITIVE LOGITS

 preferably

1.15

or

1.05

whether

0.96

 regardless

0.91

etc

0.87

 irrespective

0.86

 whether

0.86

either

0.85

ardless

0.84

someone

0.84

Activations Density 0.270%

tech-related terms and programming concepts

No Comments

No Known Activations