INDEX

Explanations

references to locations, particularly cities and countries

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

itsch

-0.80

————————

-0.77

edin

-0.75

 horizont

-0.74

ports

-0.72

dress

-0.70

TextColor

-0.70

hart

-0.70

emark

-0.69

manship

-0.68

POSITIVE LOGITS

Ľ

1.03

Ļ

0.97

ivable

0.94

³

0.87

È

0.83

arest

0.83

auld

0.82

rates

0.82

Ł

0.80

Ĥİ

0.80

Activations Density 0.017%

references to locations, particularly cities and countries

No Comments

No Known Activations