INDEX

Explanations

No Explanations Found

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

ugi

-0.75

orious

-0.70

alities

-0.69

ality

-0.68

icky

-0.65

vous

-0.65

Aim

-0.64

eless

-0.63

iful

-0.63

 Syndrome

-0.63

POSITIVE LOGITS

çīĪ

0.75

sym

0.71

 Palo

0.71

vertisements

0.65

ĻĤ

0.65

ä½ľ

0.65

ajo

0.64

 blanket

0.64

 discovery

0.63

 pioneer

0.63

Activations Density 0.000%

No Known Activations

This feature has no known activations.