INDEX

Explanations

scientific terms and locations

oai_token-act-pair · gpt-3.5-turbo

references to specific scientific classifications or designations

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

 Mile

-0.77

oppers

-0.71

oof

-0.68

 fide

-0.67

apple

-0.67

gerald

-0.67

Ã´

-0.66

ãĤ´ãĥ³

-0.66

vous

-0.66

phalt

-0.66

POSITIVE LOGITS

SC

3.61

SC

2.40

SL

1.46

SK

1.43

Sc

1.37

SW

1.29

TC

1.29

ST

1.28

LS

1.28

BW

1.27

Activations Density 0.007%

scientific terms and locations

references to specific scientific classifications or designations

No Comments

No Known Activations