INDEX

Explanations

keywords related to rules, limitations, and legal terms

oai_token-act-pair · gpt-3.5-turbo

phrases that indicate exceptions or limitations in discussions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

rawdownloadcloneembedreportprint

-0.48

 smack

-0.47

PAL

-0.47

 achie

-0.45

 Souls

-0.45

 WARRANT

-0.44

 adam

-0.42

Ont

-0.42

 progressing

-0.42

 compliant

-0.41

POSITIVE LOGITS

ccording

0.65

astical

0.62

anwhile

0.58

agine

0.54

ĺħ

0.53

sci

0.52

Origin

0.51

usual

0.51

arte

0.50

etheless

0.50

Activations Density 0.464%

keywords related to rules, limitations, and legal terms

phrases that indicate exceptions or limitations in discussions

No Comments

No Known Activations