INDEX

Explanations

No Explanations Found

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

0:0.09

1:0.07

2:0.06

3:0.07

4:0.08

5:0.09

6:0.09

7:0.08

8:0.07

9:0.08

10:0.08

11:0.08

Negative Logits

 synthesis

-1.66

 emerging

-1.61

 developmental

-1.60

VIDIA

-1.60

Initialized

-1.56

 bout

-1.54

 unsus

-1.53

 diffusion

-1.52

 cyan

-1.52

 Keystone

-1.51

POSITIVE LOGITS

Gov

1.69

mails

1.59

item

1.56

Lua

1.55

://

1.52

items

1.50

Tu

1.48

null

1.47

ributes

1.47

express

1.45

Activations Density 0.000%

No Known Activations

This feature has no known activations.