INDEX

Explanations

phrases related to the act of taking action

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "to"

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

 Alleg

-0.59

 laps

-0.59

 Alive

-0.57

 nevertheless

-0.56

 disclosures

-0.56

 Detected

-0.56

 instit

-0.55

 incurred

-0.55

 indeed

-0.55

 appropri

-0.55

POSITIVE LOGITS

wered

1.47

ggles

1.46

asted

1.46

pless

1.35

othy

1.30

ffee

1.27

asts

1.26

asty

1.21

psy

1.19

asters

1.17

Activations Density 0.128%

phrases related to the act of taking action

occurrences of the word "to"

No Comments

No Known Activations