INDEX

Explanations

references to specific objects, locations, or elements in the context of a narrative

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

0:0.02

1:0.01

2:0.11

3:0.08

4:0.27

5:0.02

6:0.15

7:0.11

8:0.05

9:0.03

10:0.05

11:0.05

Negative Logits

enhagen

-1.58

 jointly

-1.50

���

-1.49

 reinstated

-1.48

 declined

-1.48

 withdrew

-1.45

 refused

-1.44

 cancelled

-1.44

 vigorously

-1.40

 voluntarily

-1.40

POSITIVE LOGITS

idth

1.56

center

1.54

 clich

1.51

typ

1.45

ranean

1.41

tis

1.40

 coma

1.40

SPONSORED

1.35

 extremes

1.35

 audiences

1.34

Activations Density 0.001%

references to specific objects, locations, or elements in the context of a narrative

No Comments

No Known Activations