INDEX

Explanations

terms related to data security and protection

oai_token-act-pair · gpt-3.5-turbo

terms related to data security and protection technologies

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

çĭ

-0.68

inventoryQuantity

-0.64

Attempts

-0.62

ãĤ¦ãĤ¹

-0.61

ãĥ´

-0.61

Ô

-0.60

Fit

-0.59

 unnatural

-0.58

 uncont

-0.58

Eastern

-0.58

POSITIVE LOGITS

ophon

0.78

itory

0.72

anyon

0.69

etting

0.66

ancy

0.65

ieve

0.65

mingham

0.65

itaire

0.63

nas

0.62

iences

0.62

Activations Density 0.361%

terms related to data security and protection

terms related to data security and protection technologies

No Comments

No Known Activations