INDEX

Explanations

phrases containing the word "ancer"

oai_token-act-pair · gpt-3.5-turbo

references to cancer and related terms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ween

-0.72

itability

-0.69

¤

-0.68

shall

-0.67

shr

-0.65

pled

-0.64

sett

-0.64

keys

-0.63

arantine

-0.63

 Bundy

-0.63

POSITIVE LOGITS

ancers

0.82

ancer

0.81

NetMessage

0.80

 Mechdragon

0.73

 Arts

0.72

utical

0.69

ous

0.69

xual

0.68

 Starts

0.67

 recovers

0.66

Activations Density 0.028%

phrases containing the word "ancer"

references to cancer and related terms

No Comments

No Known Activations