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

4:0.09

5:0.08

6:0.06

7:0.08

8:0.08

9:0.08

10:0.11

11:0.07

Negative Logits

igure

-2.67

 incarcer

-2.67

psey

-2.61

Attach

-2.61

EVA

-2.58

lehem

-2.57

 reinforcements

-2.57

'-

-2.57

ework

-2.54

merce

-2.54

POSITIVE LOGITS

 Arts

2.81

 Craig

2.72

Ian

2.64

 Swan

2.63

 Fitzgerald

2.62

 Simon

2.62

Roz

2.59

Nas

2.58

 Santana

2.47

 Cohn

2.43

Activations Density 0.000%

No Known Activations

This feature has no known activations.