INDEX

Explanations

variations of the word "increase" in different forms and contexts

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

¥µ

-0.83

ĪĴ

-0.74

 Skydragon

-0.69

ĸļ

-0.69

akeru

-0.66

 Aman

-0.66

 Unch

-0.66

 Masquerade

-0.65

ICAN

-0.65

RTX

-0.64

POSITIVE LOGITS

orporated

1.66

orpor

1.41

arcer

1.37

orrect

1.35

identally

1.34

redible

1.32

idental

1.25

hester

1.21

redibly

1.19

itement

1.18

Activations Density 0.004%

variations of the word "increase" in different forms and contexts

No Comments

No Known Activations