INDEX

Explanations

references to equality, equivalence, or balancing across various contexts

oai_token-act-pair · gpt-3.5-turbo

terms related to equivalence or equality

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

doms

-0.72

Assembly

-0.71

ZI

-0.70

 Sandwich

-0.70

 Soda

-0.67

dom

-0.64

 mustard

-0.64

TY

-0.63

 Glass

-0.63

ãģ¦

-0.62

POSITIVE LOGITS

ivil

1.25

itably

1.08

ivalent

1.05

ifax

1.01

als

1.00

ilib

0.97

ilateral

0.97

alled

0.95

aling

0.94

aled

0.93

Activations Density 0.013%

references to equality, equivalence, or balancing across various contexts

terms related to equivalence or equality

No Comments

No Known Activations