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

 Imran

-0.68

 Tycoon

-0.65

inyl

-0.62

 prolet

-0.61

 Levi

-0.60

mosp

-0.60

 Flavoring

-0.59

XL

-0.58

âĸł

-0.58

 habit

-0.58

POSITIVE LOGITS

yip

0.77

ousy

0.77

ãĥ¼ãĥĨ

0.76

aez

0.76

 revisions

0.74

ãĥĥãĥĪ

0.74

indal

0.74

iosyncr

0.74

urst

0.72

 nodd

0.71

Activations Density 0.000%

No Known Activations

This feature has no known activations.