INDEX

Explanations

instances of the word "rogue."

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 goodbye

-0.64

Fas

-0.61

Fey

-0.60

aved

-0.60

birth

-0.56

peed

-0.55

USE

-0.54

âķĲ

-0.54

 Machina

-0.54

AAP

-0.54

POSITIVE LOGITS

raphic

1.38

raphics

1.23

raph

1.17

aming

1.02

allery

1.01

rams

0.97

roup

0.95

iov

0.91

rog

0.89

atory

0.88

Activations Density 0.003%

instances of the word "rogue."

No Comments

No Known Activations