INDEX

Explanations

the word "debris" with a high degree of activation

oai_token-act-pair · gpt-3.5-turbo

references to debris and related materials 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 @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

XM

-0.76

igne

-0.71

Brew

-0.69

hey

-0.69

 Rate

-0.68

ribe

-0.68

tu

-0.68

IQ

-0.67

lik

-0.67

YA

-0.65

POSITIVE LOGITS

 debris

3.62

 wreckage

2.29

 rubble

2.24

bris

1.69

 ashes

1.43

 fragments

1.38

 rubbish

1.38

 remnants

1.36

 clutter

1.34

 dust

1.28

Activations Density 0.021%

the word "debris" with a high degree of activation

references to debris and related materials in various contexts

No Comments

No Known Activations