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

acists

-0.73

atorium

-0.69

 Strongh

-0.66

abet

-0.66

nis

-0.66

alist

-0.65

oms

-0.63

isol

-0.63

thirds

-0.62

yll

-0.61

POSITIVE LOGITS

zzle

0.67

 landfall

0.66

 Rush

0.65

asking

0.64

icho

0.63

imentary

0.63

vernment

0.61

Ö¼

0.61

 deliveries

0.58

_-

0.57

Activations Density 0.000%

No Known Activations

This feature has no known activations.