INDEX

Explanations

mentions of the entity "Us"

oai_token-act-pair · gpt-3.5-turbo

references to the term "Us" 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

arden

-0.58

 trooper

-0.58

seed

-0.57

bleacher

-0.57

 marker

-0.57

 coat

-0.57

patch

-0.55

 draft

-0.55

 chambers

-0.54

 facial

-0.54

POSITIVE LOGITS

Us

3.92

Us

2.12

us

1.24

 Them

1.20

Me

1.18

 Tradable

1.10

Me

1.06

Veh

1.02

Him

0.92

userc

0.87

Activations Density 0.011%

mentions of the entity "Us"

references to the term "Us" in various contexts

No Comments

No Known Activations