INDEX

Explanations

phrases related to location or position

oai_token-act-pair · gpt-3.5-turbo

phrases indicating possession or belonging

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

uador

-0.74

ocene

-0.72

antine

-0.72

igent

-0.71

ower

-0.71

 oneself

-0.71

ghan

-0.70

 Canaver

-0.69

orer

-0.68

untled

-0.67

POSITIVE LOGITS

 glory

0.82

 counterparts

0.81

 selves

0.80

ãĤ¦ãĤ¹

0.79

 namesake

0.76

 heritage

0.73

 mates

0.71

 ancestors

0.69

 dominance

0.69

 counterpart

0.69

Activations Density 0.174%

phrases related to location or position

phrases indicating possession or belonging

No Comments

No Known Activations