INDEX

Explanations

phrases related to completeness or totality

oai_token-act-pair · gpt-3.5-turbo

the word "Complete" and its variations in different 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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

tem

-0.78

 gigs

-0.72

)]

-0.69

 outsiders

-0.68

gey

-0.67

gig

-0.67

 emitting

-0.64

tern

-0.64

 witness

-0.63

Fn

-0.63

POSITIVE LOGITS

 Complete

3.32

Complete

2.49

complete

1.81

 Definitive

1.57

 COMPLE

1.48

 Completed

1.14

COMPLE

1.12

 Collection

1.11

 Remastered

1.09

Completed

1.02

Activations Density 0.017%

phrases related to completeness or totality

the word "Complete" and its variations in different contexts

No Comments

No Known Activations