INDEX

Explanations

the letter "N"

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Configuration

ckkissane/attn-saes-gpt2-small-all-layers/gpt2-small_L9_Hcat_z_lr1.20e-03_l11.20e+00_ds24576_bs4096_dc1.00e-06_rsanthropic_rie25000_nr4_v9.pt

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

float32

Hook Name

blocks.9.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

1:0.08

2:0.08

3:0.08

4:0.07

5:0.08

6:0.08

7:0.08

8:0.08

9:0.07

10:0.08

11:0.09

Negative Logits

FOX

-3.83

dir

-3.09

ricks

-2.94

anooga

-2.75

fired

-2.71

retty

-2.69

reprene

-2.68

Jen

-2.67

ENTS

-2.66

Hope

-2.61

POSITIVE LOGITS

 relational

3.32

®,

2.94

 accum

2.93

 bearing

2.90

 statistical

2.86

 reproduction

2.84

 correl

2.82

 multipl

2.75

 Parenthood

2.72

 palp

2.71

Activations Density 0.000%

the letter "N"

No Comments

No Known Activations