INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Configuration

ckkissane/attn-saes-gpt2-small-all-layers/gpt2-small_L2_Hcat_z_lr1.20e-03_l11.00e+00_ds24576_bs4096_dc1.00e-06_rsanthropic_rie25000_nr4_v4.pt

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

float32

Hook Name

blocks.2.attn.hook_z

Hook Layer

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Head Attr Weights

0:0.08

1:0.06

2:0.08

3:0.08

4:0.09

5:0.08

6:0.08

7:0.08

8:0.09

9:0.09

10:0.07

11:0.08

Negative Logits

 precincts

-1.50

icago

-1.41

FM

-1.38

 Mori

-1.38

merce

-1.37

 checkpoints

-1.36

adelphia

-1.32

cius

-1.30

 beaches

-1.30

obos

-1.28

POSITIVE LOGITS

GGGG

1.60

iliar

1.46

tymology

1.45

ighty

1.44

NER

1.42

amide

1.40

allo

1.40

rawler

1.39

EEE

1.36

ahime

1.35

Activations Density 0.000%

No Known Activations

This feature has no known activations.