INDEX

Explanations

references to specific soccer teams and player statistics

oai_token-act-pair · gpt-4o-mini Triggered by @bot

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

Not in Any Lists

No Comments

Head Attr Weights

0:0.08

1:0.03

2:0.06

3:0.09

4:0.03

5:0.05

6:0.05

7:0.05

8:0.38

9:0.04

10:0.05

11:0.04

Negative Logits

 Kabul

-4.11

Ö

-4.09

 Afghan

-3.82

�

-3.80

 Karachi

-3.76

Gö

-3.69

 Afghans

-3.68

 Islamabad

-3.60

 Kubrick

-3.58

zai

-3.56

POSITIVE LOGITS

USL

5.61

 Louisville

4.87

 Kentucky

4.27

 Tennessee

3.85

Tenn

3.76

 Tulsa

3.60

 Wilmington

3.57

 statewide

3.53

 Missouri

3.51

 Dayton

3.46

Activations Density 0.008%

references to specific soccer teams and player statistics

No Comments

No Known Activations