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

pload

-0.78

OUR

-0.76

RECT

-0.73

leased

-0.70

autions

-0.70

RAFT

-0.69

ãĤ§

-0.68

Synopsis

-0.68

ãĤ¦ãĤ¹

-0.67

avascript

-0.67

POSITIVE LOGITS

 slam

0.74

bang

0.69

 insurgency

0.64

 widow

0.63

oglobin

0.63

buck

0.61

 blob

0.60

 aber

0.58

 incub

0.58

chrom

0.57

Activations Density 0.000%

No Known Activations

This feature has no known activations.