INDEX

Explanations

phrases related to connections or associations between entities or concepts

oai_token-act-pair · gpt-3.5-turbo

mentions of relationships between various entities or concepts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

hemy

-0.79

sk

-0.79

milo

-0.79

sky

-0.74

Dou

-0.73

enic

-0.72

rpm

-0.69

fl

-0.69

haps

-0.68

upiter

-0.67

POSITIVE LOGITS

 relationship

0.94

 relationships

0.92

 intimately

0.89

ually

0.87

 between

0.84

 Relationship

0.81

 relations

0.78

 partner

0.76

hips

0.74

 dynamics

0.73

Activations Density 0.025%

phrases related to connections or associations between entities or concepts

mentions of relationships between various entities or concepts

No Comments

No Known Activations