INDEX

Explanations

phrases related to interference or involvement in various contexts

oai_token-act-pair · gpt-3.5-turbo

instances of the word "interfere" and its related phrases

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

Label

-0.71

DragonMagazine

-0.69

lli

-0.67

ndra

-0.67

 "$:/

-0.66

iller

-0.66

elsen

-0.66

FN

-0.65

cki

-0.65

tek

-0.64

POSITIVE LOGITS

 interstate

0.69

 havoc

0.65

ateral

0.65

ãĤ¹ãĥĪ

0.65

 Myanmar

0.64

 interfere

0.63

 Territories

0.62

 obstruct

0.61

 unlawfully

0.61

 vegetation

0.60

Activations Density 0.138%

phrases related to interference or involvement in various contexts

instances of the word "interfere" and its related phrases

No Comments

No Known Activations