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

0:0.08

1:0.07

2:0.07

3:0.07

4:0.08

5:0.08

6:0.10

7:0.08

8:0.08

9:0.08

10:0.08

11:0.08

Negative Logits

TDs

-2.68

trump

-2.65

 "$:/

-2.58

 fascists

-2.51

mson

-2.50

zbollah

-2.47

HUD

-2.45

minus

-2.45

counter

-2.44

FactoryReloaded

-2.42

POSITIVE LOGITS

 Columb

3.42

 Prest

3.07

 Investigative

2.84

rieving

2.78

 rapes

2.76

 investigator

2.76

nes

2.66

educ

2.62

Jol

2.61

 bonding

2.55

Activations Density 0.000%

No Known Activations

This feature has no known activations.