INDEX

Explanations

expressions of agreement and trust

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

0:0.16

1:0.08

2:0.08

3:0.05

4:0.03

5:0.02

6:0.12

7:0.12

8:0.02

9:0.04

10:0.17

11:0.05

Negative Logits

Yang

-3.54

neum

-3.11

 Yang

-3.02

LT

-2.90

Rah

-2.82

CY

-2.82

riter

-2.79

neb

-2.71

NL

-2.70

Sly

-2.68

POSITIVE LOGITS

 Perkins

9.04

Per

3.98

Per

3.80

 Uganda

3.80

Ug

3.62

 Daniels

3.42

 Perry

3.34

 Kamp

3.33

 Pere

3.30

PER

3.29

Activations Density 0.002%

expressions of agreement and trust

No Comments

No Known Activations