INDEX

Explanations

terms related to industrial work such as factories, manufacturing, and specific job roles

oai_token-act-pair · gpt-3.5-turbo

references to specific industries and occupations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

ACL

-0.70

 polarized

-0.70

 Paran

-0.68

 Torrent

-0.68

 IMAGES

-0.67

RIS

-0.64

 fragmentation

-0.64

inventoryQuantity

-0.64

 streaks

-0.64

gression

-0.63

POSITIVE LOGITS

 manufacturing

1.39

 factories

1.35

 industries

1.23

 factory

1.22

 labou

1.16

Factory

1.16

 industry

1.14

 Manufacturing

1.12

manufact

1.12

 manufactures

1.11

Activations Density 1.317%

terms related to industrial work such as factories, manufacturing, and specific job roles

references to specific industries and occupations

No Comments

No Known Activations