INDEX

Explanations

phrases related to quantity, comparison, or exception

oai_token-act-pair · gpt-3.5-turbo

many instances or references to exceptions or notable cases

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

lass

-0.87

raf

-0.80

lio

-0.80

Newsletter

-0.76

deen

-0.76

here

-0.75

lyak

-0.69

atech

-0.69

adobe

-0.68

ideshow

-0.68

POSITIVE LOGITS

 exceptions

1.83

 exception

1.71

 caveats

1.46

 caveat

1.30

 emphasis

1.18

 attendant

1.12

 accompanying

1.11

 hindsight

1.11

 notable

1.10

 implications

1.10

Activations Density 0.252%

phrases related to quantity, comparison, or exception

many instances or references to exceptions or notable cases

No Comments

No Known Activations