INDEX

Explanations

phrases related to hierarchical structures or organizational levels

oai_token-act-pair · gpt-3.5-turbo

references to ranks or hierarchical structures

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

argo

-0.78

 Infinite

-0.66

 Monstrous

-0.62

 Archangel

-0.62

issance

-0.61

 physic

-0.60

 retina

-0.59

MAD

-0.59

\/\/

-0.59

uras

-0.59

POSITIVE LOGITS

 ranks

0.95

leading

0.81

ename

0.81

 insign

0.80

naires

0.78

naire

0.78

chel

0.77

enance

0.77

cius

0.75

staff

0.72

Activations Density 0.016%

phrases related to hierarchical structures or organizational levels

references to ranks or hierarchical structures

No Comments

No Known Activations