INDEX

Explanations

phrases related to meeting or surpassing expectations

oai_token-act-pair · gpt-3.5-turbo

phrases related to meeting or failing to meet expectations or promises

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

MRI

-0.79

cloth

-0.71

awi

-0.68

leans

-0.66

brow

-0.66

lees

-0.66

ept

-0.65

rians

-0.65

antine

-0.65

apse

-0.63

POSITIVE LOGITS

 promises

1.64

 promise

1.44

 commitments

1.31

 prophecy

1.26

 prophes

1.24

 pledges

1.24

 expectations

1.23

 promised

1.19

 obligations

1.17

 prophe

1.12

Activations Density 0.133%

phrases related to meeting or surpassing expectations

phrases related to meeting or failing to meet expectations or promises

No Comments

No Known Activations