INDEX

Explanations

components of narratives or statements that reflect past expectations and assessments

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_L9_Hcat_z_lr1.20e-03_l11.20e+00_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.9.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

0:0.05

1:0.02

2:0.07

3:0.15

4:0.03

5:0.05

6:0.02

7:0.04

8:0.03

9:0.02

10:0.45

11:0.04

Negative Logits

 unmatched

-2.27

nothing

-2.22

 nothing

-2.20

 alongside

-2.13

 invaluable

-2.12

 forever

-2.11

 none

-2.10

 unparalleled

-1.99

 unique

-1.96

 ineligible

-1.89

POSITIVE LOGITS

 Suppose

3.14

 imagined

3.01

 hoped

2.97

 imagining

2.78

 suppose

2.70

 assumed

2.68

 envis

2.61

 expectation

2.56

 argued

2.52

 suggested

2.50

Activations Density 0.225%

components of narratives or statements that reflect past expectations and assessments

No Comments

No Known Activations