INDEX

Explanations

references to reports and statements related to military and governmental actions

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_L10_Hcat_z_lr1.20e-03_l11.30e+00_ds24576_bs4096_dc1.00e-05_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.10.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

0:0.05

1:0.02

2:0.06

3:0.17

4:0.19

5:0.21

6:0.04

7:0.06

8:0.02

9:0.08

10:0.03

11:0.04

Negative Logits

�

-3.35

────

-3.29

™:

-3.21

________________

-3.09

————

-3.07

 🙂

-3.05

Think

-2.98

Simply

-2.96

ウス

-2.91

¢

-2.90

POSITIVE LOGITS

 condemned

4.53

 reportedly

4.41

 criticised

4.33

 denounced

4.17

 claimed

4.16

 detained

4.10

 blamed

4.08

 accused

4.00

 disputed

3.90

 reported

3.83

Activations Density 0.622%

references to reports and statements related to military and governmental actions

No Comments

No Known Activations