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

0:0.07

1:0.09

2:0.09

3:0.09

4:0.07

5:0.07

6:0.08

7:0.07

8:0.08

9:0.08

10:0.08

11:0.07

Negative Logits

 disarm

-2.74

 Yanuk

-2.73

 revolt

-2.71

 Syrians

-2.69

 Fasc

-2.67

nikov

-2.62

 retrospect

-2.59

 Iraqis

-2.57

 Syri

-2.57

 Protest

-2.55

POSITIVE LOGITS

ゼウス

2.64

nih

2.63

loe

2.58

rely

2.55

Ward

2.54

ogle

2.52

Hig

2.51

uler

2.46

LC

2.41

ウス

2.41

Activations Density 0.000%

No Known Activations

This feature has no known activations.