INDEX

Explanations

phrases indicating visitor statistics or metrics

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

 scapego

-0.68

 spons

-0.65

Ges

-0.64

 Lever

-0.63

 appell

-0.61

 Reilly

-0.60

 Ezekiel

-0.60

 Dane

-0.60

 Argon

-0.60

 Berger

-0.60

POSITIVE LOGITS

Never

0.90

Uncommon

0.77

quickShipAvailable

0.75

ilded

0.74

arth

0.73

song

0.71

Reason

0.70

wed

0.69

okemon

0.69

ãĥİ

0.69

Activations Density 0.004%

phrases indicating visitor statistics or metrics

No Comments

No Known Activations