INDEX

Explanations

names and terms related to politics and government

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

ailability

-0.68

¿½

-0.67

ICES

-0.65

ãĤ´ãĥ³

-0.64

 satell

-0.64

benefit

-0.63

church

-0.62

 commute

-0.61

FML

-0.61

TPS

-0.59

POSITIVE LOGITS

andise

0.85

orno

0.83

uay

0.82

agall

0.78

raph

0.76

vernment

0.75

nostic

0.72

iani

0.72

uez

0.70

utan

0.68

Activations Density 0.208%

names and terms related to politics and government

No Comments

No Known Activations