INDEX

Explanations

product names or models related to technology or software, especially those with the term "Pro" in them

oai_token-act-pair · gpt-3.5-turbo

mentions of professional software or products

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 halls

-0.79

ãĤ¼ãĤ¦ãĤ¹

-0.75

 straw

-0.75

 Cornell

-0.67

assetsadobe

-0.66

owship

-0.64

erella

-0.64

 Carrie

-0.64

 oranges

-0.63

ibrary

-0.63

POSITIVE LOGITS

digy

1.44

posal

1.30

secution

1.29

gression

1.26

verbs

1.17

clamation

1.15

portion

1.13

blems

1.13

ceed

1.10

pose

1.07

Activations Density 0.018%

product names or models related to technology or software, especially those with the term "Pro" in them

mentions of professional software or products

No Comments

No Known Activations