INDEX

Explanations

words related to referrals or recommendations

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

1:0.07

2:0.08

3:0.08

4:0.07

5:0.08

6:0.08

7:0.09

8:0.08

9:0.07

10:0.08

11:0.08

Negative Logits

bet

-3.55

 Faust

-3.21

Bet

-3.03

beer

-3.02

 Hermione

-2.96

aur

-2.96

ヴァ

-2.93

Iran

-2.77

Georg

-2.76

arian

-2.72

POSITIVE LOGITS

 Nanto

3.16

 unfocused

3.12

 scrim

3.12

ansk

3.05

 scout

2.92

 rede

2.84

ops

2.82

 evacuated

2.75

 airst

2.71

 crews

2.67

Activations Density 0.000%

words related to referrals or recommendations

No Comments

No Known Activations