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

1:0.07

2:0.07

3:0.09

4:0.08

5:0.09

6:0.08

7:0.09

8:0.08

9:0.07

10:0.08

11:0.08

Negative Logits

iggurat

-2.13

thumbnails

-2.10

 Hide

-1.91

atomic

-1.70

ノ

-1.69

redo

-1.69

ヘラ

-1.67

ILA

-1.67

エル

-1.64

-1.63

POSITIVE LOGITS

 slack

1.89

 reper

1.72

jog

1.70

 nour

1.70

 horm

1.65

 distressed

1.63

 deepen

1.56

Fram

1.56

 misunder

1.56

 vigil

1.55

Activations Density 0.000%

No Known Activations

This feature has no known activations.