INDEX

Explanations

references to news organizations and media outlets

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

0:0.03

1:0.01

2:0.07

3:0.06

4:0.06

5:0.19

6:0.02

7:0.04

8:0.03

9:0.34

10:0.03

11:0.07

Negative Logits

 meant

-2.27

:-

-2.14

 resemb

-2.13

 ALWAYS

-2.13

 enables

-2.10

 ideally

-2.08

 halves

-2.07

 signifies

-2.06

 analogous

-2.06

 enlarg

-2.04

POSITIVE LOGITS

 himself

2.43

 Eater

2.32

ayer

2.26

enger

2.12

 Tuesday

2.08

his

2.08

Daily

2.05

AFP

1.97

yer

1.96

aler

1.96

Activations Density 0.053%

references to news organizations and media outlets

No Comments

No Known Activations