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

0:0.10

1:0.08

2:0.08

3:0.08

4:0.09

5:0.08

6:0.08

7:0.08

8:0.07

9:0.07

10:0.07

11:0.08

Negative Logits

 cooperative

-1.96

 psychedelic

-1.68

IRD

-1.67

 addictive

-1.65

 beginner

-1.62

league

-1.54

 allocations

-1.54

 Lisp

-1.54

 developmental

-1.53

 layoffs

-1.51

POSITIVE LOGITS

1.88

 msec

1.86

デ

1.82

/-

1.81

urn

1.77

swer

1.75

ankind

1.75

ername

1.74

1.72

 Darkness

1.69

Activations Density 0.000%

No Known Activations

This feature has no known activations.