INDEX

Explanations

actions or states that involve change, particularly in contexts involving restrictions or legal issues

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_L9_Hcat_z_lr1.20e-03_l11.20e+00_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.9.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

0:0.07

1:0.02

2:0.07

3:0.11

4:0.05

5:0.06

6:0.03

7:0.44

8:0.03

9:0.02

10:0.04

11:0.02

Negative Logits

ツ

-3.06

Its

-2.89

Its

-2.72

izabeth

-2.39

its

-2.30

anye

-2.23

ffen

-2.23

ngth

-2.17

atoon

-2.16

etter

-2.15

POSITIVE LOGITS

 them

6.13

they

5.67

them

5.58

 THEM

5.50

They

5.36

 ones

5.18

 They

4.94

 they

4.69

their

4.58

 themselves

4.58

Activations Density 1.474%

actions or states that involve change, particularly in contexts involving restrictions or legal issues

No Comments

No Known Activations