INDEX

Explanations

phrases related to cultural heritage or historical artifacts

oai_token-act-pair · gpt-3.5-turbo

the word "ad" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 backwards

-0.62

 Sunshine

-0.60

 pole

-0.59

 pity

-0.59

 Borders

-0.59

 sclerosis

-0.58

 corrective

-0.58

NTS

-0.57

 Ãľ

-0.57

 stret

-0.56

POSITIVE LOGITS

visor

1.25

venture

1.19

ventures

1.11

urai

1.11

illac

1.10

rill

1.10

vance

1.08

aily

1.08

apter

1.07

visors

1.05

Activations Density 0.058%

phrases related to cultural heritage or historical artifacts

the word "ad" in various contexts

No Comments

No Known Activations