INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

 Chronicles

-0.61

 Stars

-0.60

 Evolution

-0.60

 transitioned

-0.60

EVs

-0.60

Eve

-0.59

 neutron

-0.59

 Cups

-0.58

tit

-0.58

Plex

-0.57

POSITIVE LOGITS

RAG

0.88

ESE

0.82

ieri

0.81

yrics

0.76

ãĤ®

0.75

ople

0.73

uld

0.73

oral

0.71

 Cheong

0.70

ADA

0.69

Activations Density 0.000%

No Known Activations

This feature has no known activations.