INDEX

Explanations

references to personal pronouns and indicators of relationships

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

ELD

-0.78

quickShipAvailable

-0.71

 Increasing

-0.71

onday

-0.67

ipping

-0.65

Magazine

-0.64

ebus

-0.64

liest

-0.63

 Associated

-0.63

ccording

-0.62

POSITIVE LOGITS

 breathe

1.05

 shine

1.03

 decide

0.99

 borrow

0.97

 choose

0.94

 handle

0.94

 join

0.92

 drown

0.92

 dictate

0.91

 indulge

0.91

Activations Density 0.050%

references to personal pronouns and indicators of relationships

No Comments

No Known Activations