INDEX

Explanations

phrases related to public perception or beliefs and their implications

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

0:0.01

1:0.01

2:0.10

3:0.35

4:0.10

5:0.03

6:0.03

7:0.07

8:0.04

9:0.05

10:0.09

11:0.05

Negative Logits

JUST

-2.14

��

-2.06

-1.85

giene

-1.83

uers

-1.82

untarily

-1.74

phrine

-1.72

Tam

-1.69

exit

-1.67

��

-1.66

POSITIVE LOGITS

 nowadays

1.87

 Comet

1.64

risome

1.61

 famously

1.53

 Franch

1.51

 sightings

1.49

 pamph

1.48

 talented

1.47

 fantastic

1.47

 pretty

1.45

Activations Density 0.040%

phrases related to public perception or beliefs and their implications

No Comments

No Known Activations