INDEX

Explanations

names related to laboratories

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "Lab" and its variations or related terms, indicating a focus on laboratory contexts or entities

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

icy

-0.80

NetMessage

-0.80

cy

-0.74

 ILCS

-0.68

 STATS

-0.65

 Cthulhu

-0.65

 escort

-0.64

cir

-0.64

 clutch

-0.64

nia

-0.63

POSITIVE LOGITS

Lab

2.98

lab

2.71

 Horowitz

1.83

LAB

1.64

label

1.00

DN

0.98

imeo

0.98

abus

0.90

 labs

0.89

actory

0.89

Activations Density 0.042%

names related to laboratories

occurrences of the word "Lab" and its variations or related terms, indicating a focus on laboratory contexts or entities

No Comments

No Known Activations