INDEX

Explanations

phrases indicating a solution, action, or method to address an issue

oai_token-act-pair · gpt-3.5-turbo

prepositions indicating agency or means in actions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

ational

-0.77

itto

-0.68

upon

-0.64

imental

-0.64

bian

-0.64

ertain

-0.63

kson

-0.63

asy

-0.63

>[

-0.62

ginx

-0.62

POSITIVE LOGITS

 virtue

1.30

laws

1.10

products

0.97

 fiat

0.85

gone

0.85

 proxy

0.83

 leaps

0.81

akuya

0.79

catch

0.79

 multiplying

0.79

Activations Density 0.164%

phrases indicating a solution, action, or method to address an issue

prepositions indicating agency or means in actions

No Comments

No Known Activations