INDEX

Explanations

proper nouns

oai_token-act-pair · gpt-3.5-turbo

instances of the letter 'N'

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

 surve

-0.67

 briefs

-0.66

 difficulty

-0.66

 overe

-0.63

 something

-0.61

 doub

-0.60

 inund

-0.60

 bearing

-0.60

 nowhere

-0.59

 lengths

-0.59

POSITIVE LOGITS

umerous

1.47

uclear

1.44

atalie

1.40

guyen

1.32

orman

1.32

igel

1.30

ATIONAL

1.28

athan

1.28

ancy

1.26

OVA

1.26

Activations Density 0.035%

proper nouns

instances of the letter 'N'

No Comments

No Known Activations