INDEX

Explanations

phrases indicating political actions and statements

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_L5_Hcat_z_lr1.20e-03_l11.00e+00_ds49152_bs4096_dc1.00e-06_rsanthropic_rie25000_nr4_v9.pt

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

49,152

Data Type

float32

Hook Name

blocks.5.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

1:0.01

2:0.10

3:0.38

4:0.11

5:0.03

6:0.03

7:0.08

8:0.05

9:0.04

10:0.07

11:0.05

Negative Logits

~~~~

-1.95

 nutshell

-1.83

MAS

-1.79

AMY

-1.66

))

-1.64

REDACTED

-1.59

!!!!!!!!

-1.57

 tonight

-1.57

CRE

-1.57

DIT

-1.49

POSITIVE LOGITS

 unsuccessfully

1.92

FB

1.61

 weddings

1.58

 similarly

1.52

istani

1.44

ecause

1.43

 experimented

1.41

 pregnancies

1.40

 outright

1.38

 independently

1.34

Activations Density 0.089%

phrases indicating political actions and statements

No Comments

No Known Activations