INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

 Fields

-0.72

 wards

-0.69

iffe

-0.69

ample

-0.66

mentioned

-0.62

byn

-0.61

 annexed

-0.60

 Chatt

-0.60

 fields

-0.59

campus

-0.58

POSITIVE LOGITS

ļé

0.82

itialized

0.75

OSH

0.68

ãĤ´ãĥ³

0.67

Mich

0.67

gm

0.67

areth

0.66

ogene

0.65

god

0.64

 Caps

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.