INDEX

Explanations

proper nouns or terms related to a specific company or entity

oai_token-act-pair · gpt-3.5-turbo

mentions of the company "Ex."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

 Sawyer

-0.67

 blindly

-0.66

 Rough

-0.61

bud

-0.60

 learning

-0.58

 grasp

-0.58

 shade

-0.57

 first

-0.56

 scaff

-0.56

FIG

-0.55

POSITIVE LOGITS

terior

1.41

cerpt

1.38

haust

1.33

isting

1.29

pected

1.28

cessive

1.26

istence

1.23

changes

1.23

ternally

1.22

ceptions

1.18

Activations Density 0.018%

proper nouns or terms related to a specific company or entity

mentions of the company "Ex."

No Comments

No Known Activations