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

0:0.08

1:0.08

2:0.08

3:0.08

4:0.08

5:0.07

6:0.09

7:0.07

8:0.08

9:0.08

10:0.08

11:0.08

Negative Logits

BST

-3.40

CLS

-3.09

CHAT

-2.77

LCS

-2.69

TOR

-2.68

 Cookie

-2.66

 Whats

-2.63

SOM

-2.57

 bots

-2.56

 Sheikh

-2.53

POSITIVE LOGITS

owa

3.90

aukee

3.00

kowski

2.99

rique

2.99

rican

2.90

qua

2.90

owan

2.87

bernatorial

2.83

apons

2.80

quished

2.78

Activations Density 0.000%

No Known Activations

This feature has no known activations.