INDEX

Explanations

phrases related to statements or propositions

oai_token-act-pair · gpt-3.5-turbo

recurring themes of propositions and locations

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

raham

-0.76

relation

-0.74

ãĥĥãĥĪ

-0.72

cius

-0.68

 clearance

-0.67

rely

-0.66

rel

-0.65

chem

-0.65

rations

-0.65

 crawl

-0.65

POSITIVE LOGITS

enance

0.85

eenth

0.81

sun

0.76

VERTISEMENT

0.76

âĢ¢âĢ¢âĢ¢âĢ¢

0.75

ä½¿

0.74

een

0.73

atoon

0.71

 Carbuncle

0.70

\\\\\\\\

0.70

Activations Density 0.075%

phrases related to statements or propositions

recurring themes of propositions and locations

No Known Activations