INDEX

Explanations

keywords related to unauthorized access or hacking

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

angan

-0.88

mong

-0.81

odor

-0.73

bridge

-0.73

iola

-0.72

pled

-0.71

queue

-0.71

ocracy

-0.68

circle

-0.68

mouth

-0.67

POSITIVE LOGITS

CLASSIFIED

0.96

 Parenthood

0.81

 EntityItem

0.74

aturdays

0.70

IENT

0.68

nesday

0.67

REAM

0.67

ONSORED

0.64

 ILCS

0.64

ãĤ´ãĥ³

0.64

Activations Density 0.030%

keywords related to unauthorized access or hacking

No Comments

No Known Activations