INDEX

Explanations

technical terms related to computing and programming

oai_token-act-pair · gpt-3.5-turbo

warnings or advisories regarding potential dangers or issues

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

Not in Any Lists

No Comments

Negative Logits

 phys

-0.52

)</

-0.51

)."

-0.51

 inferior

-0.49

 malfunction

-0.47

Downloadha

-0.46

 predators

-0.45

 physical

-0.44

 nearby

-0.44

 destro

-0.44

POSITIVE LOGITS

 concludes

0.61

tarians

0.56

bernatorial

0.52

 summarizes

0.51

ecided

0.50

 summarized

0.49

cipled

0.49

Political

0.49

omics

0.48

 GOODMAN

0.48

Activations Density 2.608%

technical terms related to computing and programming

warnings or advisories regarding potential dangers or issues

No Comments

No Known Activations