INDEX

Explanations

the word "go"

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

0:0.06

1:0.07

2:0.07

3:0.09

4:0.09

5:0.09

6:0.09

7:0.06

8:0.09

9:0.08

10:0.09

11:0.06

Negative Logits

[/

-2.13

Ware

-2.08

TPS

-1.99

 ledger

-1.99

 twitter

-1.99

 locker

-1.99

 garbage

-1.99

 bulletin

-1.97

 parchment

-1.96

 pudding

-1.94

POSITIVE LOGITS

cles

2.47

inations

2.42

asaki

2.41

elia

2.41

ault

2.41

leground

2.33

 prescribe

2.27

chen

2.25

icity

2.20

arella

2.15

Activations Density 0.000%

the word "go"

No Comments

No Known Activations