INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Configuration

ckkissane/attn-saes-gpt2-small-all-layers/gpt2-small_L9_Hcat_z_lr1.20e-03_l11.20e+00_ds24576_bs4096_dc1.00e-06_rsanthropic_rie25000_nr4_v9.pt

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

float32

Hook Name

blocks.9.attn.hook_z

Hook Layer

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Head Attr Weights

0:0.09

1:0.07

2:0.09

3:0.08

4:0.08

5:0.07

6:0.07

7:0.08

8:0.08

9:0.07

10:0.08

11:0.07

Negative Logits

 Kafka

-2.93

artment

-2.68

Ward

-2.63

ts

-2.58

ISC

-2.54

antry

-2.54

uckland

-2.52

idges

-2.51

 Desmond

-2.50

icol

-2.45

POSITIVE LOGITS

rament

3.06

 Gadget

2.79

 Myster

2.75

 laun

2.75

 depreciation

2.54

 mushrooms

2.54

 Manz

2.53

 Coconut

2.51

 mushroom

2.50

 Brus

2.45

Activations Density 0.000%

No Known Activations

This feature has no known activations.