INDEX

Explanations

proper names or entities

oai_token-act-pair · gpt-3.5-turbo

elements related to specific identifiers or labels, likely in a social media or web context

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

Not in Any Lists

No Comments

Negative Logits

 coli

-0.91

ÄŁ

-0.65

lance

-0.63

venants

-0.63

heit

-0.61

ãĤ¼ãĤ¦ãĤ¹

-0.60

 Facility

-0.58

ryu

-0.57

gypt

-0.56

zona

-0.55

POSITIVE LOGITS

itiveness

0.62

XL

0.59

Ban

0.57

ongyang

0.52

Ki

0.51

ody

0.50

Rih

0.49

 Hats

0.49

HRC

0.49

 litter

0.47

Activations Density 1.132%

proper names or entities

elements related to specific identifiers or labels, likely in a social media or web context

No Comments

No Known Activations