INDEX

Explanations

words related to actions or events that involve manipulation or alteration

oai_token-act-pair · gpt-3.5-turbo

prefixes or roots of words related to actions or states

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

 guiActiveUnfocused

-0.81

 consequential

-0.77

DragonMagazine

-0.69

 generic

-0.68

Disclaimer

-0.64

soDeliveryDate

-0.62

 inclusive

-0.62

 Lauder

-0.62

 mandatory

-0.62

unknown

-0.62

POSITIVE LOGITS

izes

1.23

wrote

1.12

itates

1.11

ifies

1.09

ulates

0.96

ises

0.95

ook

0.95

rolled

0.94

aciously

0.93

ounced

0.93

Activations Density 0.324%

words related to actions or events that involve manipulation or alteration

prefixes or roots of words related to actions or states

No Comments

No Known Activations