INDEX

Explanations

topics or subjects in written documents

oai_token-act-pair · gpt-3.5-turbo

nouns related to reports, bills, and other formal documents or items

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.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

•Voight-Kampff

No Comments

Negative Logits

Unknown

-0.71

ĪĴ

-0.67

 Wrong

-0.64

 forgetting

-0.62

 wrong

-0.61

dden

-0.60

dropping

-0.59

wrong

-0.58

 Remain

-0.57

true

-0.57

POSITIVE LOGITS

 consists

1.26

 consisted

1.21

 comprises

1.13

 revolves

1.11

 contains

1.09

 debuted

1.07

 includes

1.02

 utilizes

0.98

 premiered

0.97

 combines

0.97

Activations Density 0.343%

topics or subjects in written documents

nouns related to reports, bills, and other formal documents or items

No Comments

No Known Activations