INDEX

Explanations

proper nouns, possibly related to news articles or reports

oai_token-act-pair · gpt-3.5-turbo

terms related to official documents and identification systems

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

cean

-0.91

hex

-0.77

kiss

-0.75

utch

-0.71

camp

-0.71

ãĥĥ

-0.68

lahoma

-0.68

idae

-0.67

isen

-0.67

lings

-0.65

POSITIVE LOGITS

TON

1.31

RY

1.29

NESS

1.24

LY

1.23

TERN

1.21

IUM

1.21

ION

1.21

ER

1.19

LAND

1.16

IAL

1.15

Activations Density 0.046%

proper nouns, possibly related to news articles or reports

terms related to official documents and identification systems

No Comments

No Known Activations