INDEX

Explanations

phrases related to physical spaces or locations

oai_token-act-pair · gpt-3.5-turbo

references to physical or conceptual spaces

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

 Simpson

-0.69

 damning

-0.64

ocide

-0.64

hood

-0.63

OK

-0.62

 neurot

-0.62

 pathological

-0.61

 Lomb

-0.61

PUT

-0.60

 punishing

-0.60

POSITIVE LOGITS

 spaces

1.41

 Spaces

1.12

hips

0.97

dayName

0.92

 space

0.91

inventoryQuantity

0.91

space

0.85

 gaps

0.85

paces

0.84

hops

0.84

Activations Density 0.008%

phrases related to physical spaces or locations

references to physical or conceptual spaces

No Comments

No Known Activations