INDEX

Explanations

terms related to specific companies or entities

oai_token-act-pair · gpt-3.5-turbo

specific brands, companies, or online platforms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

worldly

-0.77

cedes

-0.75

ourning

-0.74

requires

-0.73

pees

-0.72

-+-+

-0.71

umenthal

-0.71

alore

-0.68

vered

-0.67

uria

-0.67

POSITIVE LOGITS

 playbook

1.05

 logo

1.00

 homepage

0.93

 Foundation

0.90

 team

0.90

 saga

0.87

 Institute

0.87

 universe

0.86

 debacle

0.86

 crew

0.86

Activations Density 0.388%

terms related to specific companies or entities

specific brands, companies, or online platforms

No Comments

No Known Activations