INDEX

Explanations

mentions of monuments or landmarks

oai_token-act-pair · gpt-3.5-turbo

references to monuments and significant landmarks

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

erm

-0.79

FP

-0.68

ENG

-0.67

comes

-0.67

Roy

-0.66

Liquid

-0.65

ooters

-0.63

iencies

-0.63

cery

-0.63

rosso

-0.62

POSITIVE LOGITS

 monuments

1.14

 monument

1.09

 erected

0.98

alist

0.84

ajo

0.81

anan

0.81

 statues

0.80

arium

0.78

 commemor

0.77

 paintings

0.77

Activations Density 0.016%

mentions of monuments or landmarks

references to monuments and significant landmarks

No Comments

No Known Activations