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

3:0.08

4:0.07

5:0.09

6:0.07

7:0.07

8:0.08

9:0.07

10:0.08

11:0.08

Negative Logits

married

-1.91

ヘラ

-1.85

places

-1.72

 restrooms

-1.63

women

-1.58

elected

-1.55

 monop

-1.53

 automobiles

-1.52

variable

-1.52

lat

-1.50

POSITIVE LOGITS

 Dhabi

1.79

 Janeiro

1.74

 independ

1.66

 VICE

1.66

 euph

1.63

 Dota

1.59

 Sans

1.58

LEASE

1.57

 reddit

1.56

externalActionCode

1.52

Activations Density 0.000%

No Known Activations

This feature has no known activations.