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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

eele

-0.82

 specificity

-0.68

 Belgian

-0.65

 Cambodia

-0.65

 coincidence

-0.63

faced

-0.63

 Jasper

-0.62

 Sagan

-0.62

 Dates

-0.61

Cah

-0.60

POSITIVE LOGITS

ãĤ±

0.74

urga

0.74

ream

0.69

swer

0.68

 oath

0.67

 whine

0.66

qus

0.66

nyder

0.66

©¶æ¥µ

0.64

 encour

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.