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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 capacity

-0.68

 experien

-0.66

 misses

-0.63

 omission

-0.62

 coverage

-0.61

 respect

-0.60

 confir

-0.59

pps

-0.57

 reception

-0.57

 malf

-0.57

POSITIVE LOGITS

sic

0.78

å§«

0.76

enture

0.75

Scroll

0.73

issors

0.73

rift

0.73

rub

0.70

0000000000000000

0.70

VERTISEMENT

0.69

SF

0.67

Activations Density 0.000%

No Known Activations

This feature has no known activations.