INDEX

Explanations

phrases related to various challenges and risks

oai_token-act-pair · gpt-3.5-turbo

phrases related to challenges and risks

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

ership

-0.71

onement

-0.69

BN

-0.67

info

-0.67

owder

-0.66

mun

-0.65

 donation

-0.65

MER

-0.61

igun

-0.60

line

-0.60

POSITIVE LOGITS

hooting

1.09

cale

1.06

pring

1.05

hift

1.04

 inherent

0.91

afety

0.90

omething

0.88

cape

0.87

paces

0.87

 abound

0.87

Activations Density 0.111%

phrases related to various challenges and risks

phrases related to challenges and risks

No Comments

No Known Activations