INDEX

Explanations

mentions of physical or abstract signs, such as signals, indications, or markers

oai_token-act-pair · gpt-3.5-turbo

the term "signs" in various contexts, often indicating indications or symptoms of a situation or condition

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

inion

-0.74

udeb

-0.66

milo

-0.64

ifle

-0.62

ranged

-0.62

ency

-0.61

olate

-0.60

ania

-0.60

hw

-0.59

 Exile

-0.59

POSITIVE LOGITS

 signs

4.13

 Signs

3.20

 sign

2.48

 signage

1.86

 indications

1.83

sign

1.63

 Sign

1.59

Sign

1.51

 billboards

1.48

 clues

1.48

Activations Density 0.017%

mentions of physical or abstract signs, such as signals, indications, or markers

the term "signs" in various contexts, often indicating indications or symptoms of a situation or condition

No Comments

No Known Activations