INDEX

Explanations

names and terms related to Star Wars characters and movies

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

esville

-0.75

iard

-0.74

Vog

-0.74

IED

-0.72

Ou

-0.69

orio

-0.68

imen

-0.65

Org

-0.65

ball

-0.64

IGF

-0.64

POSITIVE LOGITS

 Skywalker

1.15

 Kenobi

0.93

 trooper

0.92

 Vader

0.92

 Trooper

0.89

 Awakens

0.84

 troopers

0.84

 lightsaber

0.83

 Rebels

0.80

 pilot

0.76

Activations Density 0.063%

names and terms related to Star Wars characters and movies

No Comments

No Known Activations