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

STRUCT

-0.66

 Dudley

-0.65

LM

-0.65

lad

-0.63

ector

-0.63

loo

-0.61

 Kidd

-0.61

 Seymour

-0.60

heng

-0.60

okane

-0.60

POSITIVE LOGITS

tics

0.72

hyde

0.71

vity

0.69

 rehe

0.68

 immortality

0.68

Cause

0.68

 soluble

0.67

anium

0.66

ó

0.65

theless

0.65

Activations Density 0.000%

No Known Activations

This feature has no known activations.