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

agn

-1.00

anners

-0.84

Gall

-0.83

<?

-0.79

GI

-0.76

ulner

-0.75

Ù

-0.74

ODUCT

-0.72

OB

-0.71

boys

-0.71

POSITIVE LOGITS

 Protocol

0.72

 submission

0.69

 pipelines

0.66

arthed

0.66

earch

0.65

 comprehension

0.65

 cave

0.64

 waterways

0.63

 decisions

0.63

 militia

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.