INDEX

Explanations

words related to nullification or declaring something void

oai_token-act-pair · gpt-3.5-turbo

terms related to nullification or invalidation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

 toget

-0.80

DonaldTrump

-0.77

 Remastered

-0.68

ilitating

-0.67

 livest

-0.63

 millenn

-0.63

 hemor

-0.62

ãĤ·ãĥ£

-0.62

 Harlem

-0.62

heny

-0.62

POSITIVE LOGITS

ptr

1.32

ifying

1.22

ification

1.21

ified

1.15

ifies

1.14

ifier

1.13

ify

1.07

ifiers

1.07

ative

0.96

locked

0.92

Activations Density 0.022%

words related to nullification or declaring something void

terms related to nullification or invalidation

No Comments

No Known Activations