INDEX

Explanations

codes or identifiers, such as product codes or numbers

oai_token-act-pair · gpt-3.5-turbo

identifiers and codes related to products or services

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

âĢ¢âĢ¢âĢ¢âĢ¢

-0.75

met

-0.71

||||

-0.67

ships

-0.66

atu

-0.65

mons

-0.65

chev

-0.65

ful

-0.65

Cla

-0.64

ãĥ£

-0.64

POSITIVE LOGITS

FORMATION

0.91

DERR

0.81

IRD

0.80

ULTS

0.79

INGTON

0.71

OTAL

0.70

OTA

0.67

INFO

0.65

ennes

0.64

Scene

0.64

Activations Density 0.013%

codes or identifiers, such as product codes or numbers

identifiers and codes related to products or services

No Comments

No Known Activations