INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Configuration

ckkissane/attn-saes-gpt2-small-all-layers/gpt2-small_L2_Hcat_z_lr1.20e-03_l11.00e+00_ds24576_bs4096_dc1.00e-06_rsanthropic_rie25000_nr4_v4.pt

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

float32

Hook Name

blocks.2.attn.hook_z

Hook Layer

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Head Attr Weights

0:0.08

1:0.08

2:0.08

3:0.08

4:0.07

5:0.08

6:0.07

7:0.10

8:0.07

9:0.07

10:0.07

11:0.09

Negative Logits

SpaceEngineers

-1.85

ital

-1.67

assemb

-1.62

tube

-1.59

ilage

-1.52

edition

-1.50

oscopic

-1.48

tur

-1.47

amic

-1.45

acet

-1.43

POSITIVE LOGITS

ahime

1.72

 idols

1.61

"]=>

1.53

 endeavour

1.46

 feminists

1.46

 mathemat

1.45

 Awakens

1.40

 nutshell

1.37

 patriarchal

1.36

 Saga

1.35

Activations Density 0.000%

No Known Activations

This feature has no known activations.