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

0:0.06

1:0.10

2:0.08

3:0.08

4:0.07

5:0.06

6:0.09

7:0.10

8:0.07

9:0.06

10:0.09

11:0.09

Negative Logits

Wiz

-1.91

 Whip

-1.77

 Chair

-1.64

 Leader

-1.59

Dur

-1.54

)))

-1.52

 hers

-1.51

 Chairman

-1.48

"))

-1.48

 outfit

-1.45

POSITIVE LOGITS

fixes

1.89

cens

1.84

inventoryQuantity

1.76

lethal

1.76

hift

1.75

existent

1.71

覚醒

1.71

suff

1.69

netflix

1.66

フォ

1.65

Activations Density 0.000%

No Known Activations

This feature has no known activations.