INDEX

Explanations

mentions of the word "lo" at a high activation level

oai_token-act-pair · gpt-3.5-turbo

repeated occurrences of the term "lo" 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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

mM

-0.70

tox

-0.67

 Mercury

-0.67

ATIVE

-0.59

 Batt

-0.59

 gestation

-0.58

 Dietary

-0.57

 immun

-0.57

 Ceres

-0.57

 Sherman

-0.56

POSITIVE LOGITS

lo

4.52

LO

1.80

Lo

1.66

la

1.53

los

1.52

li

1.45

Lo

1.34

loe

1.31

loop

1.29

lic

1.22

Activations Density 0.009%

mentions of the word "lo" at a high activation level

repeated occurrences of the term "lo" in various contexts

No Comments

No Known Activations