INDEX

Explanations

mentions of puppets

oai_token-act-pair · gpt-3.5-turbo

references to puppets and puppet-related themes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

IGH

-0.82

 Angels

-0.77

andals

-0.75

-0.74

Ox

-0.74

 MacArthur

-0.70

Hav

-0.69

 Aviation

-0.66

 Forbes

-0.65

 coli

-0.64

POSITIVE LOGITS

 puppet

1.54

 pupp

1.51

uppet

1.40

 Puppet

1.10

eering

1.03

olicy

0.90

etry

0.87

glers

0.85

 masters

0.85

 simul

0.83

Activations Density 0.007%

mentions of puppets

references to puppets and puppet-related themes

No Comments

No Known Activations