INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Configuration

ckkissane/attn-saes-gpt2-small-all-layers/gpt2-small_L10_Hcat_z_lr1.20e-03_l11.30e+00_ds24576_bs4096_dc1.00e-05_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.10.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

1:0.10

2:0.08

3:0.07

4:0.08

5:0.08

6:0.08

7:0.07

8:0.08

9:0.07

10:0.07

11:0.08

Negative Logits

 dart

-2.78

utenberg

-2.77

abwe

-2.57

 tradem

-2.54

quart

-2.54

eele

-2.51

ersive

-2.47

 Seym

-2.45

aired

-2.40

 bour

-2.40

POSITIVE LOGITS

UGH

2.92

>]

2.78

 inflic

2.61

yrs

2.55

 Liberals

2.45

 accusing

2.44

\'

2.41

cuts

2.40

TRUMP

2.39

ULTS

2.38

Activations Density 0.000%

No Known Activations

This feature has no known activations.