INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

ilk

-0.78

AMA

-0.72

mine

-0.68

izabeth

-0.67

anco

-0.66

agos

-0.64

 blockers

-0.62

argo

-0.62

ansen

-0.61

²¾

-0.60

POSITIVE LOGITS

minster

0.77

rists

0.75

bou

0.70

ropolitan

0.67

prem

0.65

 Fancy

0.65

typ

0.64

sych

0.64

)",

0.63

 presumptive

0.62

Activations Density 0.000%

No Known Activations

This feature has no known activations.