INDEX

Explanations

the concept of "opposite"

oai_token-act-pair · gpt-3.5-turbo

the term "opposite" in various contexts

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

zan

-0.79

ITED

-0.72

 Annotations

-0.71

 Beans

-0.70

 blat

-0.68

ortium

-0.65

ulic

-0.64

hens

-0.64

estones

-0.64

utters

-0.64

POSITIVE LOGITS

 sexes

0.98

 direction

0.95

sex

0.90

 poles

0.85

 extremes

0.84

 directions

0.79

 sides

0.79

 side

0.79

 pole

0.78

 gender

0.77

Activations Density 0.049%

the concept of "opposite"

the term "opposite" in various contexts

No Comments

No Known Activations