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

0:0.08

1:0.06

2:0.10

3:0.07

4:0.08

5:0.08

6:0.06

7:0.07

8:0.09

9:0.07

10:0.09

11:0.09

Negative Logits

kun

-1.67

 sailed

-1.63

 Shal

-1.50

Ces

-1.49

BuyableInstoreAndOnline

-1.49

 writes

-1.48

 Nicola

-1.48

arden

-1.48

 travellers

-1.46

 Travels

-1.45

POSITIVE LOGITS

atically

1.64

�

1.52

hibit

1.50

Reason

1.47

UCT

1.46

unity

1.45

icative

1.45

inct

1.41

fighter

1.41

 bount

1.40

Activations Density 0.000%

No Known Activations

This feature has no known activations.