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

0:0.07

1:0.06

2:0.09

3:0.07

4:0.09

5:0.07

6:0.08

7:0.09

8:0.09

9:0.08

10:0.07

11:0.09

Negative Logits

vot

-1.72

cheat

-1.72

resy

-1.68

 tripod

-1.66

 exponent

-1.64

 silicone

-1.62

quished

-1.61

 ancest

-1.61

 handheld

-1.60

luaj

-1.59

POSITIVE LOGITS

Poll

1.85

覚醒

1.71

escal

1.66

erest

1.63

ylan

1.61

 interns

1.61

rieve

1.60

uel

1.59

rams

1.58

 McMaster

1.56

Activations Density 0.000%

No Known Activations

This feature has no known activations.