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

0:0.09

1:0.07

2:0.08

3:0.07

4:0.08

5:0.08

6:0.09

7:0.08

8:0.07

9:0.08

10:0.08

11:0.08

Negative Logits

lihood

-1.74

DW

-1.69

 Documents

-1.67

 accessing

-1.62

cair

-1.59

ocard

-1.56

FE

-1.55

 Shack

-1.54

 purse

-1.51

 Acquisition

-1.48

POSITIVE LOGITS

 introdu

1.64

使

1.63

iannopoulos

1.61

 juven

1.60

 Canaver

1.57

aunts

1.56

」

1.55

 blat

1.54

 Rudolph

1.54

Û

1.54

Activations Density 0.000%

No Known Activations

This feature has no known activations.