INDEX

Explanations

terms related to filtering and categorizing information

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_L5_Hcat_z_lr1.20e-03_l11.00e+00_ds49152_bs4096_dc1.00e-06_rsanthropic_rie25000_nr4_v9.pt

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

49,152

Data Type

float32

Hook Name

blocks.5.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

1:0.01

2:0.05

3:0.07

4:0.09

5:0.02

6:0.04

7:0.41

8:0.02

9:0.04

10:0.09

11:0.07

Negative Logits

覚醒

-1.51

jad

-1.50

INGTON

-1.40

Carl

-1.38

BRE

-1.33

von

-1.33

�

-1.31

rollers

-1.31

ACT

-1.30

 cents

-1.28

POSITIVE LOGITS

selection

1.48

 attrition

1.44

opian

1.41

rients

1.37

tml

1.35

 applicants

1.32

ividual

1.31

 nominations

1.27

tics

1.26

nels

1.25

Activations Density 0.008%

terms related to filtering and categorizing information

No Comments

No Known Activations