INDEX

Explanations

phrases related to engagement

oai_token-act-pair · gpt-3.5-turbo

mentions of engagement or related concepts in texts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

IFIED

-0.74

pmwiki

-0.73

ifications

-0.71

 Leilan

-0.60

 rise

-0.60

 tumble

-0.59

 propriet

-0.58

 Tale

-0.58

fully

-0.58

istically

-0.57

POSITIVE LOGITS

agement

1.40

raved

1.36

ravings

1.28

lishes

1.23

rish

1.22

agements

1.22

rams

1.07

endered

1.07

ulf

1.06

lished

1.02

Activations Density 0.052%

phrases related to engagement

mentions of engagement or related concepts in texts

No Comments

No Known Activations