INDEX

Explanations

phrases related to acronyms or abbreviations with the structure 'NOR'

oai_token-act-pair · gpt-3.5-turbo

references to the NOR designation or acronym in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

Xi

-0.90

 Qiao

-0.85

Hut

-0.79

ibal

-0.78

 Deng

-0.76

ql

-0.75

ugal

-0.75

Woo

-0.74

Xia

-0.72

anian

-0.72

POSITIVE LOGITS

NOR

3.06

NOR

2.95

MH

1.60

MH

1.41

CAN

1.33

 RAND

1.20

NTS

1.17

Norm

1.17

FAA

1.13

SAM

1.11

Activations Density 0.058%

phrases related to acronyms or abbreviations with the structure 'NOR'

references to the NOR designation or acronym in various contexts

No Comments

No Known Activations