INDEX

Explanations

instances where an action or event is expected to happen

oai_token-act-pair · gpt-3.5-turbo

expressions related to expectations and predictions

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

farious

-0.66

raint

-0.64

é¾įå¥ĳå£«

-0.63

ayer

-0.63

ocative

-0.61

grounds

-0.59

 Acting

-0.57

enthal

-0.55

 Ariel

-0.52

GGGGGGGG

-0.52

POSITIVE LOGITS

 hear

1.56

see

1.47

 encounter

1.27

 find

1.24

 receive

1.18

 encount

1.11

 seeing

1.10

get

1.04

 hears

1.01

 witness

1.00

Activations Density 0.328%

instances where an action or event is expected to happen

expressions related to expectations and predictions

No Comments

No Known Activations