INDEX

Explanations

organizations or companies

oai_token-act-pair · gpt-3.5-turbo

the word "which" associated with various organizations or entities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

ATH

-0.72

mma

-0.65

Interstitial

-0.64

 VIDEOS

-0.64

 forgetting

-0.60

istance

-0.59

 Conversation

-0.59

 Passage

-0.59

 awakening

-0.59

obyl

-0.58

POSITIVE LOGITS

 owns

1.48

 oversees

1.45

 specializes

1.39

 operates

1.37

 regulates

1.33

 publishes

1.24

 oversaw

1.19

 promotes

1.18

 employs

1.17

 dominates

1.17

Activations Density 0.128%

organizations or companies

the word "which" associated with various organizations or entities

No Comments

No Known Activations