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

0:0.07

1:0.09

2:0.08

3:0.09

4:0.09

5:0.07

6:0.08

7:0.08

8:0.08

9:0.07

10:0.08

11:0.08

Negative Logits

 Adin

-2.21

 Solitaire

-1.74

entimes

-1.73

 Galile

-1.67

 Seym

-1.67

nces

-1.66

 Compared

-1.64

 fert

-1.62

Magikarp

-1.62

 Ukrain

-1.61

POSITIVE LOGITS

 disclosure

2.04

 commons

1.94

Catalog

1.67

leg

1.63

 Supporters

1.59

},{"

1.58

Film

1.55

 Ground

1.54

 accordingly

1.52

 spoiler

1.51

Activations Density 0.000%

No Known Activations

This feature has no known activations.