INDEX

Explanations

instances of ellipses, indicating omitted or trailing thoughts

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Head Attr Weights

0:0.08

1:0.05

2:0.09

3:0.07

4:0.10

5:0.07

6:0.08

7:0.10

8:0.07

9:0.06

10:0.09

11:0.08

Negative Logits

rouse

-1.83

 anecd

-1.83

---------

-1.81

 belie

-1.64

 understandably

-1.59

Developer

-1.59

Develop

-1.58

 readers

-1.54

 acutely

-1.52

 folks

-1.51

POSITIVE LOGITS

daq

1.95

alach

1.69

azo

1.66

��

1.65

ヤ

1.62

veyard

1.59

alias

1.57

bestos

1.57

ailability

1.53

 towed

1.53

Activations Density 0.000%

instances of ellipses, indicating omitted or trailing thoughts

No Comments

No Known Activations