INDEX

Explanations

punctuation and hashtags

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

 challeng

-0.72

 faculties

-0.71

 lect

-0.70

 Suite

-0.68

 hinge

-0.67

 nuances

-0.67

 Pyramid

-0.65

 capacities

-0.65

 citiz

-0.65

pse

-0.64

POSITIVE LOGITS

soDeliveryDate

0.91

tnc

0.84

TRUMP

0.83

ï¸ı

0.82

channelAvailability

0.79

Success

0.79

????

0.78

Trump

0.78

destroy

0.77

?????

0.76

Activations Density 0.012%

punctuation and hashtags

No Comments

No Known Activations