INDEX

Explanations

phrases related to location and setting

oai_token-act-pair · gpt-3.5-turbo

contexts related to behaviors and actions occurring in specific places or under certain conditions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

paio

-0.57

letes

-0.55

rogens

-0.53

avorite

-0.52

apego

-0.51

brids

-0.51

yss

-0.50

rehensive

-0.50

 sear

-0.49

lete

-0.48

POSITIVE LOGITS

 guise

0.80

 fashion

0.65

 manner

0.64

ãĤ½

0.62

 vicinity

0.60

 obscurity

0.59

ividual

0.58

 infancy

0.57

 context

0.54

 mode

0.53

Activations Density 2.447%

phrases related to location and setting

contexts related to behaviors and actions occurring in specific places or under certain conditions

No Comments

No Known Activations