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

ä½ľ

-0.82

bery

-0.68

selves

-0.64

fter

-0.64

geries

-0.63

 ropes

-0.62

 TOTAL

-0.61

hetic

-0.60

gery

-0.60

stals

-0.60

POSITIVE LOGITS

usp

0.68

cc

0.67

vec

0.67

 Holden

0.64

Pa

0.63

ect

0.62

mosp

0.61

equ

0.61

llular

0.61

sylv

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.