INDEX

Explanations

actions or abilities that can have an impact or effect

oai_token-act-pair · gpt-3.5-turbo

actions or capabilities associated with doing or taking something

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

 Mell

-0.62

Jur

-0.61

Cly

-0.60

}"

-0.60

Kub

-0.60

BuyableInstoreAndOnline

-0.59

ĸļ

-0.58

Suz

-0.58

Ruk

-0.58

 Sabb

-0.57

POSITIVE LOGITS

 safely

0.99

 easily

0.90

ibly

0.80

ivably

0.80

 anytime

0.78

 muster

0.78

 freely

0.75

 anywhere

0.72

estine

0.72

 efficiently

0.71

Activations Density 0.505%

actions or abilities that can have an impact or effect

actions or capabilities associated with doing or taking something

No Comments

No Known Activations