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

agascar

-0.95

 toget

-0.82

 helicop

-0.79

 inval

-0.74

éĹĺ

-0.74

 rall

-0.74

 conduc

-0.74

"},"

-0.72

ãĥĺ

-0.68

ikuman

-0.68

POSITIVE LOGITS

 paraph

0.62

 sketch

0.62

dot

0.59

DIY

0.59

 Brian

0.59

 Serpent

0.59

VR

0.58

onz

0.58

 Tatt

0.58

DV

0.57

Activations Density 0.000%

No Known Activations

This feature has no known activations.