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

Head Attr Weights

0:0.08

1:0.07

2:0.08

3:0.06

4:0.09

5:0.08

6:0.08

7:0.09

8:0.07

9:0.09

10:0.09

11:0.06

Negative Logits

 wrench

-2.87

 chemically

-2.83

 acids

-2.74

dyn

-2.73

 dstg

-2.71

 fused

-2.59

 absorb

-2.56

ゴン

-2.53

 chained

-2.47

iesel

-2.45

POSITIVE LOGITS

 Surveillance

3.03

 parks

3.00

 Atlas

2.82

 Occupy

2.81

 Topic

2.77

 Sunshine

2.69

 Trafford

2.64

 Advertisement

2.63

 Navigation

2.55

 Meadow

2.55

Activations Density 0.000%

No Known Activations

This feature has no known activations.