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

 describ

-0.89

Sov

-0.84

 behavi

-0.74

\\\\

-0.71

 explan

-0.70

ħĭ

-0.68

 acknow

-0.65

 begg

-0.64

 mosqu

-0.63

à¹

-0.63

POSITIVE LOGITS

ials

0.74

undai

0.73

ono

0.73

rons

0.70

onis

0.70

iar

0.70

phis

0.69

ndum

0.69

0.68

rica

0.68

Activations Density 0.000%

No Known Activations

This feature has no known activations.