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

0:0.08

1:0.09

2:0.07

3:0.08

4:0.08

5:0.09

6:0.10

7:0.08

8:0.08

9:0.07

10:0.08

11:0.06

Negative Logits

accompan

-1.82

inventoryQuantity

-1.67

MpServer

-1.66

 belonged

-1.56

boro

-1.55

$$

-1.55

olester

-1.55

caster

-1.54

cause

-1.54

chwitz

-1.50

POSITIVE LOGITS

nod

1.83

 nods

1.68

 ponder

1.64

 nodding

1.59

 corridors

1.58

 gaps

1.52

shy

1.51

 indec

1.47

 curv

1.47

 HuffPost

1.45

Activations Density 0.000%

No Known Activations

This feature has no known activations.