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

0:0.10

1:0.08

2:0.10

3:0.08

4:0.07

5:0.08

6:0.06

7:0.08

8:0.08

9:0.08

10:0.07

11:0.07

Negative Logits

��

-2.07

 Dolphin

-1.72

 rall

-1.72

Interstitial

-1.69

TextColor

-1.67

 Pastebin

-1.66

Dot

-1.65

Export

-1.62

olphin

-1.61

olphins

-1.57

POSITIVE LOGITS

igen

1.81

heads

1.63

 constitu

1.62

 squads

1.62

axis

1.61

icals

1.57

alogy

1.53

aeda

1.50

 amino

1.47

otypes

1.46

Activations Density 0.000%

No Known Activations

This feature has no known activations.