INDEX

Explanations

references to molecules, specifically molecules with a high activation value indicating a strong match

oai_token-act-pair · gpt-3.5-turbo

terms related to molecules and chemical structures

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

į

-0.82

swick

-0.77

Disney

-0.74

²

-0.74

san

-0.70

ĳå£«

-0.70

GM

-0.70

shire

-0.69

ership

-0.69

 Bald

-0.69

POSITIVE LOGITS

 molecule

1.34

 molecules

1.29

cules

1.27

 nitrogen

0.92

cule

0.91

ynthesis

0.87

 oxygen

0.85

 oxide

0.81

 ammonia

0.81

 tiss

0.80

Activations Density 0.017%

references to molecules, specifically molecules with a high activation value indicating a strong match

terms related to molecules and chemical structures

No Comments

No Known Activations