INDEX

Explanations

phrases indicating a prolonged period or continuity of state or action

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_L3_Hcat_z_lr1.20e-03_l19.00e-01_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.3.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

0:0.01

1:0.02

2:0.07

3:0.06

4:0.01

5:0.03

6:0.10

7:0.03

8:0.13

9:0.12

10:0.08

11:0.28

Negative Logits

ADVERTISEMENT

-1.19

 converter

-1.16

 liner

-1.07

 Scrib

-1.03

��

-0.99

 scanner

-0.96

���

-0.96

dll

-0.96

vote

-0.95

#$#$

-0.94

POSITIVE LOGITS

 misunder

1.16

 underestimated

1.13

odynam

1.13

 surpassed

1.09

uously

1.06

 notably

1.04

 controvers

1.04

ゼウス

1.03

ocally

1.01

UTC

1.01

Activations Density 0.021%

phrases indicating a prolonged period or continuity of state or action

No Comments

No Known Activations