INDEX

Explanations

negative or contradictory statements

oai_token-act-pair · gpt-3.5-turbo

repetitive negative conjunctions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

DragonMagazine

-0.75

ULAR

-0.62

imil

-0.61

gnu

-0.60

Pie

-0.59

 Mushroom

-0.58

 horm

-0.57

igr

-0.57

trap

-0.56

ÙĴ

-0.56

POSITIVE LOGITS

wegian

1.13

nor

0.88

hovah

0.85

berto

0.78

 necessarily

0.71

forth

0.68

theless

0.67

fleet

0.66

alysed

0.66

Else

0.66

Activations Density 0.016%

negative or contradictory statements

repetitive negative conjunctions

No Comments

No Known Activations