INDEX

Explanations

proper nouns

oai_token-act-pair · gpt-3.5-turbo

specific names and terms related to geographical locations and individuals

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

ALLY

-0.81

GBT

-0.65

 differential

-0.65

WI

-0.63

 separation

-0.60

HL

-0.59

 Germ

-0.58

VIEW

-0.56

 Drift

-0.56

 pressure

-0.56

POSITIVE LOGITS

aurus

1.65

hip

1.30

ques

1.27

ync

1.25

omething

1.24

ktop

1.23

ophical

1.21

creen

1.18

keleton

1.18

chool

1.16

Activations Density 0.041%

proper nouns

specific names and terms related to geographical locations and individuals

No Comments

No Known Activations