INDEX

Explanations

the use of the pronoun "I"

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

 Reverse

-0.66

tnc

-0.66

pires

-0.62

groupon

-0.59

Gap

-0.59

ses

-0.59

INGTON

-0.57

 indistinguishable

-0.56

 excess

-0.56

interstitial

-0.56

POSITIVE LOGITS

'm

1.40

've

1.30

'll

1.17

'd

1.13

 suppose

1.11

 dunno

1.10

ronic

1.05

 guess

1.03

 mean

1.00

 hope

1.00

Activations Density 0.160%

the use of the pronoun "I"

No Comments

No Known Activations