INDEX

Explanations

words related to actions or things that are considered unnecessary

oai_token-act-pair · gpt-3.5-turbo

mention of unnecessary actions or items

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.2.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

ebus

-0.93

ramid

-0.80

rix

-0.78

ois

-0.78

eming

-0.77

odan

-0.77

mitt

-0.75

ieving

-0.74

quart

-0.72

opsis

-0.72

POSITIVE LOGITS

 unnecessary

1.15

 duplication

0.93

 complication

0.89

 unnecess

0.88

 precaution

0.83

 superflu

0.82

aneous

0.82

 expense

0.81

 waste

0.81

 guiActiveUn

0.79

Activations Density 0.009%

words related to actions or things that are considered unnecessary

mention of unnecessary actions or items

No Comments

No Known Activations