INDEX

Explanations

mentions of the word "Blood" in the text

oai_token-act-pair · gpt-3.5-turbo

references to blood 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

 monop

-0.84

 forwarding

-0.74

Â¨

-0.72

 predic

-0.72

 directional

-0.68

 indoors

-0.68

 flat

-0.68

piv

-0.67

 pivot

-0.67

essen

-0.67

POSITIVE LOGITS

 Blood

3.67

Blood

2.92

blood

2.11

 blood

1.92

 Bloody

1.51

 Bones

1.49

 Flesh

1.41

 Bone

1.39

 Breath

1.30

 Vampire

1.29

Activations Density 0.018%

mentions of the word "Blood" in the text

references to blood in various contexts

No Comments

No Known Activations