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

3:0.09

4:0.07

5:0.08

6:0.09

7:0.07

8:0.08

9:0.09

10:0.08

11:0.08

Negative Logits

Ng

-3.50

Nat

-3.20

Nan

-3.13

Mek

-3.11

Kin

-3.06

Jen

-2.93

Nat

-2.92

Tan

-2.87

Nau

-2.86

NK

-2.84

POSITIVE LOGITS

ricular

2.97

 hiber

2.85

��

2.64

pload

2.58

rophic

2.53

 vortex

2.52

Downloadha

2.46

 filler

2.45

borgh

2.44

flo

2.43

Activations Density 0.000%

No Known Activations

This feature has no known activations.