INDEX

Explanations

words related to action or participation in contexts

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_L7_Hcat_z_lr1.20e-03_l11.10e+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.7.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

0:0.08

1:0.08

2:0.09

3:0.08

4:0.07

5:0.07

6:0.08

7:0.06

8:0.09

9:0.09

10:0.08

11:0.08

Negative Logits

 Myst

-2.15

 Merrill

-2.04

JM

-2.01

 Euras

-1.98

Mania

-1.95

 Weaver

-1.94

 Maurit

-1.91

 Fiorina

-1.91

†

-1.89

pse

-1.86

POSITIVE LOGITS

berra

2.42

oxicity

2.31

otin

2.16

undown

2.10

ustain

2.04

 patrol

2.00

terday

1.98

ancock

1.97

kamp

1.96

mos

1.95

Activations Density 0.000%

words related to action or participation in contexts

No Comments

No Known Activations