INDEX

Explanations

terms related to different types of relationships or connections between entities

oai_token-act-pair · gpt-3.5-turbo

terms associated with health and illness

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

pload

-0.85

ONEY

-0.80

KEY

-0.71

natureconservancy

-0.70

Shares

-0.69

ORGE

-0.69

keyes

-0.69

xxxxxxxx

-0.64

liga

-0.64

ocrates

-0.64

POSITIVE LOGITS

quartered

0.89

lihood

0.77

riors

0.72

istance

0.71

 hybrid

0.65

 garbage

0.64

tenance

0.64

 tendencies

0.64

etheless

0.63

omn

0.63

Activations Density 0.162%

terms related to different types of relationships or connections between entities

terms associated with health and illness

No Comments

No Known Activations