INDEX

Explanations

terms related to military actions and engagements

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs49152-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

49,152

Data Type

torch.float32

Hook Point

blocks.8.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Pric

-0.85

 autobi

-0.84

 Oprah

-0.82

soDeliveryDate

-0.79

 Guinness

-0.79

 Kardash

-0.77

creator

-0.76

 parenting

-0.75

 Autism

-0.73

 offic

-0.73

POSITIVE LOGITS

 enemy

1.28

 infantry

1.22

 besie

1.20

combat

1.18

 flank

1.18

 ambush

1.17

enemy

1.16

 skirm

1.16

 armored

1.14

 attack

1.14

Activations Density 0.441%

terms related to military actions and engagements

No Comments

No Known Activations