INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Configuration

ckkissane/attn-saes-gpt2-small-all-layers/gpt2-small_L5_Hcat_z_lr1.20e-03_l11.00e+00_ds49152_bs4096_dc1.00e-06_rsanthropic_rie25000_nr4_v9.pt

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

49,152

Data Type

float32

Hook Name

blocks.5.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

1:0.10

2:0.08

3:0.08

4:0.09

5:0.08

6:0.06

7:0.08

8:0.08

9:0.07

10:0.07

11:0.08

Negative Logits

independence

-1.82

inventoryQuantity

-1.82

gency

-1.79

pill

-1.75

ahime

-1.72

aeda

-1.68

MPG

-1.65

 legality

-1.65

 deterrent

-1.62

 flavours

-1.62

POSITIVE LOGITS

Interstitial

1.95

 sharper

1.72

の�

1.71

 Savior

1.70

ventory

1.64

]}

1.63

—"

1.59

 Tatt

1.58

 Opinion

1.57

 collateral

1.55

Activations Density 0.000%

No Known Activations

This feature has no known activations.