INDEX

Explanations

specific details and formatting elements within text

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_L3_Hcat_z_lr1.20e-03_l19.00e-01_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.3.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

0:0.02

1:0.04

2:0.12

3:0.06

4:0.03

5:0.06

6:0.11

7:0.26

8:0.06

9:0.03

10:0.08

11:0.08

Negative Logits

��

-1.24

aste

-1.18

URES

-1.09

 diversion

-1.08

Merit

-1.02

 recess

-1.01

URE

-1.00

 Krish

-0.97

NESS

-0.96

forts

-0.96

POSITIVE LOGITS

 Included

1.18

were

1.06

 respectively

1.04

webkit

1.03

ortium

1.02

each

1.02

listed

1.02

cyclopedia

1.02

blogs

0.99

 listed

0.98

Activations Density 0.242%

specific details and formatting elements within text

No Comments

No Known Activations