INDEX

Explanations

No Explanations Found

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 advent

-0.90

 causal

-0.70

 hemor

-0.70

ionage

-0.68

 espionage

-0.68

 raids

-0.68

iltr

-0.67

 clos

-0.66

atically

-0.66

 occurrence

-0.65

POSITIVE LOGITS

 Kinn

0.79

holding

0.69

 Genesis

0.67

 Canyon

0.65

 Gentleman

0.65

quit

0.63

drawn

0.63

 Lomb

0.62

ciation

0.62

ken

0.62

Activations Density 0.000%

No Known Activations

This feature has no known activations.