INDEX

Explanations

the word "to"

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

0:0.07

1:0.07

2:0.08

3:0.07

4:0.07

5:0.08

6:0.09

7:0.08

8:0.08

9:0.08

10:0.08

11:0.08

Negative Logits

aukee

-3.11

angelo

-3.00

jun

-2.94

 tremend

-2.79

 unification

-2.78

cells

-2.72

imen

-2.69

holders

-2.67

cule

-2.64

paio

-2.62

POSITIVE LOGITS

RB

2.94

fee

2.90

 Edinburgh

2.88

 Cheong

2.83

EUR

2.82

thro

2.80

NX

2.76

CK

2.73

EEE

2.72

 apologise

2.66

Activations Density 0.000%

the word "to"

No Comments

No Known Activations