INDEX

Explanations

phrases related to various actions and the outcomes of those actions

oai_token-act-pair · gpt-3.5-turbo

concepts related to transformation and change

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

hemat

-0.51

)"

-0.48

aving

-0.45

talk

-0.43

grave

-0.43

"?

-0.42

 âĢ¦"

-0.40

 drained

-0.40

)",

-0.40

 ..."

-0.39

POSITIVE LOGITS

ibly

0.51

ãĤ¨ãĥ«

0.49

ãĤ©

0.49

nesday

0.49

emonium

0.48

 guiActiveUn

0.47

bilt

0.47

orpor

0.47

ÃĽ

0.46

iously

0.45

Activations Density 0.987%

phrases related to various actions and the outcomes of those actions

concepts related to transformation and change

No Comments

No Known Activations