INDEX

Explanations

instances of significant numerical values or statistics within the text

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

 troubles

-0.73

isoft

-0.68

yan

-0.67

 occupancy

-0.65

 reorgan

-0.64

 exoner

-0.63

roit

-0.63

 "$:/

-0.62

undai

-0.62

IPM

-0.62

POSITIVE LOGITS

 Provided

1.26

 https

0.95

 http

0.88

↵Âł

0.84

YES

0.79

Liter

0.73

http

0.72

Thompson

0.71

 Whereas

0.70

 Logged

0.70

Activations Density 0.077%

instances of significant numerical values or statistics within the text

No Comments

No Known Activations