INDEX

Explanations

terms related to grammatical categories and their usage

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

0:0.03

1:0.04

2:0.07

3:0.14

4:0.03

5:0.06

6:0.09

7:0.20

8:0.06

9:0.04

10:0.08

11:0.10

Negative Logits

DragonMagazine

-1.44

 resil

-1.40

schild

-1.22

riger

-1.21

ideos

-1.19

ngth

-1.16

hiba

-1.16

 recruits

-1.14

ÍÍ

-1.12

gyn

-1.12

POSITIVE LOGITS

š

1.38

 noun

1.25

 Esper

1.23

 adject

1.23

 enumer

1.21

 ital

1.20

('

1.16

heses

1.15

ciation

1.12

č

1.12

Activations Density 0.012%

terms related to grammatical categories and their usage

No Comments

No Known Activations