INDEX

Explanations

repeated references to a specific alignment or comparison, often associated with audience reactions or emotional responses

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

0:0.20

1:0.06

2:0.10

3:0.07

4:0.13

5:0.08

6:0.06

7:0.04

8:0.07

9:0.03

10:0.07

11:0.04

Negative Logits

anmar

-1.33

processing

-1.25

cas

-1.23

asus

-1.20

ruary

-1.20

chnology

-1.17

 batch

-1.16

 casing

-1.15

ancial

-1.15

 activ

-1.15

POSITIVE LOGITS

Dialogue

1.20

externalActionCode

1.16

 Certainly

1.15

�

1.13

Whoever

1.12

Sean

1.10

Laughs

1.09

�

1.09

Journal

1.07

ONSORED

1.07

Activations Density 0.103%

repeated references to a specific alignment or comparison, often associated with audience reactions or emotional responses

No Comments

No Known Activations