INDEX

Explanations

references to the word "pent" or variations of it like "pentagon" or "pentathlon"

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

ãģĦ

-0.82

OAD

-0.76

Ø©

-0.67

ISSION

-0.65

 tune

-0.64

 Mahar

-0.63

 Friendly

-0.63

ãĥīãĥ©ãĤ´ãĥ³

-0.61

 Shake

-0.61

 disabling

-0.60

POSITIVE LOGITS

agon

1.29

agonal

1.23

agram

1.20

obar

1.10

agons

1.09

athlon

1.06

ipl

1.06

acles

1.03

etr

1.03

inct

1.00

Activations Density 0.042%

references to the word "pent" or variations of it like "pentagon" or "pentathlon"

No Comments

No Known Activations