INDEX

Explanations

specific phrases related to actions and events

oai_token-act-pair · gpt-3.5-turbo

phrases indicating actions or activities involving movement or 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

 anymore

-0.51

tan

-0.49

 whichever

-0.48

base

-0.47

iddle

-0.47

yond

-0.47

 Chapters

-0.47

MI

-0.47

entric

-0.46

affected

-0.46

POSITIVE LOGITS

 yesterday

0.60

 last

0.59

 earlier

0.53

 Thursday

0.52

 captcha

0.49

 extensively

0.49

Roz

0.48

 Wednesday

0.48

 Horowitz

0.48

ax

0.48

Activations Density 1.103%

specific phrases related to actions and events

phrases indicating actions or activities involving movement or change

No Comments

No Known Activations