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

1:0.09

2:0.07

3:0.06

4:0.04

5:0.17

6:0.14

7:0.02

8:0.05

9:0.05

10:0.11

11:0.07

Negative Logits

 Rolls

-1.63

AFB

-1.60

 weap

-1.59

 Plans

-1.53

 Presents

-1.45

 Suite

-1.45

Air

-1.44

 Loft

-1.42

 Sporting

-1.41

 Workshop

-1.40

POSITIVE LOGITS

��

1.84

��

1.72

Streamer

1.62

enum

1.61

enos

1.61

魔

1.60

��

1.59

argon

1.59

�

1.59

endif

1.52

Activations Density 0.000%

No Known Activations

This feature has no known activations.