INDEX

Explanations

words related to engagements or involvement

oai_token-act-pair · gpt-3.5-turbo

references to engagement 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 @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

pmwiki

-0.70

IFIED

-0.69

ifications

-0.68

istically

-0.68

 Leilan

-0.66

FUL

-0.63

andra

-0.62

itarian

-0.60

 Tale

-0.60

fully

-0.59

POSITIVE LOGITS

raved

1.29

agement

1.24

ravings

1.21

rish

1.11

agements

1.08

rams

1.06

lishes

1.06

endered

1.00

inse

0.95

rave

0.94

Activations Density 0.029%

words related to engagements or involvement

references to engagement in various contexts

No Comments

No Known Activations