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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 recomp

-0.68

 reconstruct

-0.68

 reconstruction

-0.66

 fine

-0.66

 steam

-0.65

 plateau

-0.65

 exits

-0.64

 rebuilt

-0.63

 increments

-0.62

 bounce

-0.61

POSITIVE LOGITS

âĹ¼

1.49

AU

0.80

SPONSORED

0.77

XL

0.74

Kin

0.74

Tumblr

0.74

Myth

0.74

 Labrador

0.73

Human

0.72

Uncommon

0.72

Activations Density 0.000%

No Known Activations

This feature has no known activations.