INDEX

Explanations

terminology related to food and culinary descriptions

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

Head Attr Weights

0:0.03

1:0.05

2:0.07

3:0.05

4:0.16

5:0.04

6:0.15

7:0.19

8:0.04

9:0.04

10:0.08

11:0.05

Negative Logits

 unlocks

-1.82

Charges

-1.74

��

-1.65

 destro

-1.65

�

-1.63

 ramps

-1.59

 unlock

-1.59

 abort

-1.57

correct

-1.54

 impuls

-1.53

POSITIVE LOGITS

 Victims

1.85

resa

1.69

 Borders

1.63

 Shirley

1.57

 Dorothy

1.54

Lod

1.51

rag

1.51

 Survivors

1.48

 Britons

1.46

afia

1.46

Activations Density 0.000%

terminology related to food and culinary descriptions

No Known Activations