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

1:0.07

2:0.07

3:0.09

4:0.09

5:0.09

6:0.08

7:0.07

8:0.08

9:0.09

10:0.08

11:0.07

Negative Logits

 Afterwards

-1.92

Liter

-1.78

Discover

-1.75

roma

-1.73

uesday

-1.71

urion

-1.69

Fax

-1.66

urations

-1.62

olitan

-1.62

rarily

-1.61

POSITIVE LOGITS

 horm

1.84

 inability

1.64

 willingness

1.63

 aggravated

1.61

 susceptibility

1.61

 cytok

1.58

pec

1.57

dy

1.53

��

1.51

 misleading

1.49

Activations Density 0.000%

No Known Activations

This feature has no known activations.