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

inous

-0.85

 âĨĳ

-0.80

 âĶĤ

-0.77

HTML

-0.76

rys

-0.76

MI

-0.74

Model

-0.74

 Doctrine

-0.72

EMA

-0.72

FT

-0.72

POSITIVE LOGITS

angan

0.93

 permitting

0.77

 dispers

0.75

 choking

0.71

theless

0.71

 encour

0.69

 jour

0.67

 deteriorated

0.66

 Xuan

0.65

 respir

0.63

Activations Density 0.000%

No Known Activations

This feature has no known activations.