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

0:0.07

1:0.06

2:0.09

3:0.09

4:0.08

5:0.07

6:0.07

7:0.08

8:0.09

9:0.08

10:0.08

11:0.09

Negative Logits

rification

-1.75

 deprivation

-1.70

vex

-1.69

 guts

-1.61

 populace

-1.61

 whole

-1.60

 plurality

-1.59

 scapego

-1.54

 disproportion

-1.53

 blot

-1.53

POSITIVE LOGITS

ASA

2.34

CLASSIFIED

2.01

itely

1.91

1.78

ONSORED

1.76

Male

1.72

ACTED

1.71

Stud

1.71

Tree

1.69

[|

1.68

Activations Density 0.000%

No Known Activations

This feature has no known activations.