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

Fram

-0.72

undle

-0.67

Soft

-0.66

prototype

-0.65

 Ampl

-0.65

angular

-0.65

 Calm

-0.64

soType

-0.63

phrase

-0.62

ciating

-0.61

POSITIVE LOGITS

 behalf

0.71

acters

0.68

aminer

0.66

 Tenn

0.64

glers

0.63

 redeem

0.62

®

0.59

 justice

0.59

 recip

0.59

kins

0.58

Activations Density 0.000%

No Known Activations

This feature has no known activations.