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

1:0.05

2:0.09

3:0.09

4:0.09

5:0.09

6:0.08

7:0.07

8:0.07

9:0.08

10:0.07

11:0.09

Negative Logits

 Nope

-1.92

 Played

-1.86

 Supports

-1.80

abs

-1.79

 Believe

-1.78

 Firstly

-1.69

ocaust

-1.65

omics

-1.61

 Pist

-1.59

Registered

-1.58

POSITIVE LOGITS

aird

1.69

 migr

1.69

 sunset

1.61

 creeping

1.58

 twilight

1.53

 tuning

1.51

 dusk

1.51

 circular

1.48

ITCH

1.47

 unch

1.47

Activations Density 0.000%

No Known Activations

This feature has no known activations.