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

laws

-0.70

endars

-0.68

rencies

-0.68

 prevailed

-0.67

 bills

-0.65

portation

-0.65

NetMessage

-0.63

 transm

-0.61

 circ

-0.60

 exch

-0.60

POSITIVE LOGITS

Sok

0.74

SAM

0.73

meric

0.68

 Drill

0.68

ocratic

0.67

Lie

0.66

endon

0.65

LEASE

0.64

URI

0.64

 Kenobi

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.