INDEX

Explanations

special characters or symbols that indicate formatting errors

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

0:0.08

1:0.09

2:0.07

3:0.09

4:0.07

5:0.07

6:0.08

7:0.08

8:0.07

9:0.08

10:0.07

11:0.08

Negative Logits

 penet

-3.17

 diesel

-3.08

 sensors

-3.05

��

-3.03

 penetrate

-3.01

 sensor

-2.96

 Israelis

-2.79

 Transport

-2.72

IDF

-2.64

 penetrated

-2.63

POSITIVE LOGITS

Hunt

3.22

goo

2.95

luaj

2.91

 Hallow

2.88

Yang

2.88

 Hungry

2.85

quit

2.76

 Fantastic

2.72

Fry

2.66

bies

2.65

Activations Density 0.000%

special characters or symbols that indicate formatting errors

No Comments

No Known Activations