INDEX

Explanations

past actions or experiences

oai_token-act-pair · gpt-3.5-turbo

references to personal experiences and memories related to food and drinking

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

ansky

-0.65

 relent

-0.63

interstitial

-0.62

 rematch

-0.56

RHP

-0.53

 Payne

-0.53

 extradition

-0.52

 reporters

-0.52

 billboards

-0.51

LLOW

-0.51

POSITIVE LOGITS

 previous

1.22

 previously

1.12

 prior

1.01

 beforehand

0.94

 preceding

0.93

 recently

0.86

 ancestors

0.86

or

0.85

ago

0.84

 before

0.81

Activations Density 0.511%

past actions or experiences

references to personal experiences and memories related to food and drinking

No Comments

No Known Activations