INDEX

Explanations

phrases related to consistency or alignment

oai_token-act-pair · gpt-3.5-turbo

references to being "in line" 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 @ 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

CVE

-0.92

soever

-0.79

rely

-0.74

ilt

-0.74

è¦ļéĨĴ

-0.71

 Seym

-0.67

ulet

-0.67

 Remastered

-0.66

 nodd

-0.64

 livest

-0.63

POSITIVE LOGITS

backer

0.89

breakers

0.78

stad

0.75

breaker

0.71

lines

0.70

dated

0.68

arity

0.66

line

0.66

ups

0.66

ioned

0.66

Activations Density 0.019%

phrases related to consistency or alignment

references to being "in line" in various contexts

No Comments

No Known Activations