INDEX

Explanations

proper nouns

oai_token-act-pair · gpt-3.5-turbo

specific names or identifiers related to people or entities

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

 Lead

-0.85

iron

-0.81

 Iron

-0.81

 iron

-0.79

 Able

-0.75

 Copper

-0.68

 malaria

-0.67

 arsenic

-0.66

 Stick

-0.66

 Solomon

-0.66

POSITIVE LOGITS

yy

4.16

YY

2.48

nn

1.82

eeee

1.70

NN

1.36

kk

1.36

aaaa

1.33

yah

1.30

hhhh

1.26

aaa

1.26

Activations Density 0.035%

proper nouns

specific names or identifiers related to people or entities

No Comments

No Known Activations