INDEX

Explanations

verbs and phrases indicating intent or purpose

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

1:0.04

2:0.06

3:0.11

4:0.02

5:0.04

6:0.05

7:0.04

8:0.09

9:0.35

10:0.09

11:0.06

Negative Logits

龍契士

-1.37

von

-1.18

Baltimore

-1.18

onday

-1.15

PsyNetMessage

-1.13

 Archangel

-1.12

BD

-1.11

town

-1.11

writers

-1.10

 haunted

-1.09

POSITIVE LOGITS

rehend

1.32

htaking

1.27

 escape

1.19

 CLICK

1.18

ynthesis

1.18

 firsthand

1.16

erial

1.14

 easily

1.13

COMPLE

1.11

 VIEW

1.11

Activations Density 0.012%

verbs and phrases indicating intent or purpose

No Comments

No Known Activations