INDEX

Explanations

words related to writing or authorship

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs49152-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

49,152

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

ourke

-0.82

 tert

-0.79

eele

-0.78

ICAN

-0.77

 prest

-0.75

ngth

-0.74

areth

-0.74

oret

-0.71

 rall

-0.71

uez

-0.68

POSITIVE LOGITS

pedia

0.86

eworthy

0.77

haven

0.73

 Creek

0.70

 Bullets

0.65

ookie

0.65

 Report

0.64

beat

0.63

 Monkey

0.63

 Patch

0.63

Activations Density 0.016%

words related to writing or authorship

No Comments

No Known Activations

words related to writing or authorship

No Comments

No Known Activations