INDEX

Explanations

words related to decision-making and evaluation processes

New Auto-Interp

Top Features by Cosine Similarity

Configuration

ckkissane/attn-saes-gpt2-small-all-layers/gpt2-small_L9_Hcat_z_lr1.20e-03_l11.20e+00_ds24576_bs4096_dc1.00e-06_rsanthropic_rie25000_nr4_v9.pt

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

float32

Hook Name

blocks.9.attn.hook_z

Hook Layer

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Activation Function

relu

Embeds

IFrame

Link

Head Attr Weights

0:0.06

1:0.06

2:0.35

3:0.11

4:0.04

5:0.06

6:0.05

7:0.07

8:0.03

9:0.05

10:0.04

11:0.03

Negative Logits

 Activ

-3.18

 Emergency

-3.14

 Shelter

-2.90

 evac

-2.88

Activ

-2.87

activ

-2.81

Emergency

-2.81

 emergency

-2.77

 evacuation

-2.75

 lockdown

-2.71

POSITIVE LOGITS

 differences

5.59

 comparisons

5.54

 similarity

5.49

 similarities

5.46

 comparison

5.37

 Differences

5.27

 comparing

5.19

 Comparison

4.98

 compare

4.97

Compar

4.84

Activations Density 0.375%

words related to decision-making and evaluation processes

No Known Activations