INDEX

Explanations

connections between different objects, concepts, or location

oai_token-act-pair · gpt-3.5-turbo

phrases containing comparative statements about regions or countries

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

escription

-0.89

 escort

-0.68

 evac

-0.65

oslav

-0.65

 evacuation

-0.64

 timely

-0.63

 redes

-0.63

iphate

-0.63

 liaison

-0.63

 volunte

-0.62

POSITIVE LOGITS

racuse

0.70

inventoryQuantity

0.66

Downloadha

0.65

flation

0.65

SOURCE

0.64

Nevertheless

0.64

Still

0.63

depending

0.63

 Brilliant

0.62

BT

0.62

Activations Density 0.511%

connections between different objects, concepts, or location

phrases containing comparative statements about regions or countries

No Comments

No Known Activations