INDEX

Explanations

phrases indicating potential outcomes or actions related to future events

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_L7_Hcat_z_lr1.20e-03_l11.10e+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.7.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

0:0.09

1:0.04

2:0.00

3:0.19

4:0.09

5:0.17

6:0.05

7:0.02

8:0.10

9:0.16

10:0.01

11:0.02

Negative Logits

 ensures

-2.09

 acknowledges

-1.83

 recognizes

-1.82

 compares

-1.79

maxwell

-1.79

 feels

-1.77

 allows

-1.77

 distinguishes

-1.76

 understands

-1.71

 promotes

-1.70

POSITIVE LOGITS

 someday

2.97

 tomorrow

2.43

morrow

2.24

 hereafter

2.15

venge

2.06

later

2.03

soon

1.95

 soon

1.93

1.89

Eventually

1.84

Activations Density 0.158%

phrases indicating potential outcomes or actions related to future events

No Comments

No Known Activations