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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

nces

-0.96

theless

-0.85

omorph

-0.85

tein

-0.80

otom

-0.79

pora

-0.78

xual

-0.77

uana

-0.76

roma

-0.73

eday

-0.72

POSITIVE LOGITS

0.72

 Shea

0.70

Rib

0.65

 motto

0.63

 Pict

0.63

felt

0.62

0.61

0.60

 Nasa

0.60

iff

0.59

Activations Density 0.000%

No Known Activations

This feature has no known activations.