INDEX

Explanations

phrases related to conflicting situations or elements

oai_token-act-pair · gpt-3.5-turbo

phrases that imply conditions and qualifications

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

eva

-0.86

".

-0.71

ombat

-0.68

miss

-0.66

illance

-0.66

istor

-0.66

FTWARE

-0.65

bart

-0.65

assies

-0.65

angu

-0.65

POSITIVE LOGITS

 admittedly

0.85

 ostensibly

0.81

 preferably

0.77

 hitherto

0.77

 lacking

0.75

 understandably

0.73

)]

0.73

 originating

0.72

 perhaps

0.72

 notoriously

0.72

Activations Density 0.295%

phrases related to conflicting situations or elements

phrases that imply conditions and qualifications

No Comments

No Known Activations