INDEX

Explanations

phrases related to individual names or specific proper nouns

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

catentry

-0.74

 fict

-0.69

ivably

-0.65

UCT

-0.62

 nephew

-0.62

ulatory

-0.62

ascript

-0.62

 merc

-0.61

 wearer

-0.61

PLIED

-0.60

POSITIVE LOGITS

inski

1.04

insky

0.94

nik

0.94

chev

0.88

chuk

0.87

gaard

0.85

zie

0.84

yna

0.83

sky

0.83

 Lumpur

0.83

Activations Density 0.105%

phrases related to individual names or specific proper nouns

No Comments

No Known Activations