INDEX

Explanations

terms related to multiple instances or occurrences of something

oai_token-act-pair · gpt-3.5-turbo

references to structured group arrangements or configurations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

iron

-0.68

 Wilde

-0.61

tips

-0.60

baugh

-0.59

 whence

-0.58

 Freeman

-0.57

 Stout

-0.54

 Hague

-0.51

EST

-0.51

atos

-0.51

POSITIVE LOGITS

multiple

1.50

 multiple

1.50

Multiple

1.32

 Multiple

1.28

 consecut

1.15

 simultaneous

1.07

 overlapping

1.06

 simultaneously

1.03

multi

1.02

ãĥ¯ãĥ³

1.02

Activations Density 0.573%

terms related to multiple instances or occurrences of something

references to structured group arrangements or configurations

No Comments

No Known Activations