INDEX

Explanations

proper nouns and names, particularly those related to individuals and places

oai_token-act-pair · gpt-4o-mini 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

ipeg

-0.83

angers

-0.82

req

-0.76

endiary

-0.74

ending

-0.73

ends

-0.71

psons

-0.71

eties

-0.69

anger

-0.68

ield

-0.65

POSITIVE LOGITS

lehem

0.89

nai

0.81

cia

0.74

utical

0.72

hedral

0.71

htaking

0.71

 Crist

0.70

zza

0.69

 Walls

0.67

Constructed

0.66

Activations Density 0.052%

proper nouns and names, particularly those related to individuals and places

No Comments

No Known Activations