INDEX

Explanations

proper nouns with unusual characters or words written in all caps

oai_token-act-pair · gpt-3.5-turbo

references to types or categories of content, particularly related to news or reporting

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

angan

-0.67

aunder

-0.67

 patri

-0.62

Oracle

-0.61

 disenfranch

-0.61

chance

-0.60

 downt

-0.59

 flawed

-0.59

olt

-0.59

soType

-0.59

POSITIVE LOGITS

 IMAGES

1.47

TY

1.09

ENCY

1.01

 WATCHED

0.90

URE

0.81

MU

0.79

URES

0.79

STEM

0.77

LES

0.75

APE

0.75

Activations Density 0.007%

proper nouns with unusual characters or words written in all caps

references to types or categories of content, particularly related to news or reporting

No Comments

No Known Activations