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

æĿ

-0.81

è¦ļéĨĴ

-0.76

ken

-0.74

ordes

-0.73

±

-0.69

Redd

-0.67

çĭ

-0.67

ãĤ¦ãĤ¹

-0.67

tee

-0.66

glers

-0.65

POSITIVE LOGITS

orie

0.69

 accelerator

0.67

 placeholder

0.65

binary

0.65

 extension

0.64

 orbit

0.63

icker

0.63

odes

0.62

iatrics

0.61

indle

0.59

Activations Density 0.000%

No Known Activations

This feature has no known activations.