INDEX

Explanations

mentions of specific locations, particularly junctions

oai_token-act-pair · gpt-3.5-turbo

references to junctions and their associated contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

Winner

-0.78

 Panic

-0.77

rek

-0.71

yre

-0.70

liquid

-0.68

anyahu

-0.67

ogun

-0.66

isable

-0.65

galitarian

-0.64

istor

-0.63

POSITIVE LOGITS

jun

1.44

cture

1.29

 junction

1.04

ctions

0.89

iors

0.84

eteenth

0.84

iper

0.83

ames

0.74

uke

0.74

iet

0.73

Activations Density 0.008%

mentions of specific locations, particularly junctions

references to junctions and their associated contexts

No Comments

No Known Activations