ViperEkura
c2c18f10d6
fix: replace unicode checkmark with ASCII argmax label
2026-05-07 23:08:16 +08:00
ViperEkura
e594d02d31
feat: merge D,E,F into one batch (gated by F at t=8) to show worse static idle
2026-05-07 22:47:21 +08:00
ViperEkura
29007c7b97
chore: remove T0-T3 demo tokens from position-grouped decode section
2026-05-07 22:25:25 +08:00
ViperEkura
6b0a1dbb5e
refactor: redesign batching FSM as queue pipeline with dynamic task states
...
- Replace 4 vertical system-phase boxes with 3 horizontal lanes
(PENDING queue / RUNNING batch / FINISHED done) for accurate
request lifecycle per scheduler.py:197-200
- System phases (Refill, Prefill, Decode, Cleanup) shown as
transition labels between lanes
- Tokens placed below lanes with dynamic state badge + cumulative
token count, updated each tick via ReplacementTransform
- Fix prefix_cache collective FadeOut using self.mobjects sweep
- Remove weight=BOLD across all scenes to prevent text drift
- Adjust GQA y-coordinates for subtitle clearance
2026-05-07 17:56:17 +08:00
ViperEkura
c05a432e45
chore: use Times New Roman across all scenes, widen Transformer block to 2.4 for \'Transformer Block × 24\' label
2026-05-07 14:59:27 +08:00
ViperEkura
57abefa47f
fix: shift GQA layout down 0.4 to avoid title-input overlap
2026-05-07 14:51:31 +08:00
ViperEkura
6b26ec33ab
chore: switch .gitignore to whitelist mechanism (only .py, .md, .gitignore)
2026-05-07 14:14:15 +08:00
ViperEkura
0018868ee3
refactor: transformer — heatmap two-phase scores+mask, auto-regressive full I/O pipeline with Emb, RMS Norm, LM Head, distribution
2026-05-07 14:00:52 +08:00
ViperEkura
496f964979
refactor: architecture — boxes left, descriptions right, 4-layer layout
2026-05-07 12:18:33 +08:00
ViperEkura
4d96a84fc5
refactor: architecture layer-by-layer introduction with source file refs
2026-05-07 11:37:21 +08:00
ViperEkura
e7d736a3b0
fix: bottom spacing, remove specs card, full formula first with no ellipsis in steps
2026-05-07 09:28:07 +08:00
ViperEkura
eeaf0a5a16
feat: SDPA formula breakdown + attention score heatmap with per-cell causal mask
2026-05-07 00:58:46 +08:00
ViperEkura
ba100c19d1
refactor: Gantt chart — same 5 requests both sides, wait+run bars, batch boxes connect GPU↔reqs
2026-05-07 00:33:42 +08:00
ViperEkura
b2ff243447
fix: call align_to AFTER move_to so bars follow rect positions
2026-05-06 22:13:11 +08:00
ViperEkura
f6ccefb34e
fix: align throughput bars by same-x positioning instead of arrange Left-edge
2026-05-06 22:11:18 +08:00
ViperEkura
ce2a7c165a
fix: throughput bars — proportional widths (1.0x vs 3.4x), proper GrowFromEdge anim
2026-05-06 22:07:37 +08:00
ViperEkura
42c1d5f700
fix: throughput comparison layout — centered rows with proper Left-Right alignment
2026-05-06 22:02:20 +08:00
ViperEkura
4ed16a70b4
fix: new requests enter at Prefill stage, not Refill
2026-05-06 21:59:47 +08:00
ViperEkura
fc68fc9107
rewrite continuous_batching as state-machine with batch token rotation, in/out flow, bitmask anim, and throughput bars
2026-05-06 21:53:15 +08:00
ViperEkura
a7a79eef96
fix: architecture layout and rewrite continuous_batching with dynamic pipeline animation
2026-05-06 21:48:18 +08:00
ViperEkura
c03abd31fe
add project source files
2026-05-06 21:16:57 +08:00
ViperEkura
94aadb3d8f
first commit
2026-05-06 21:15:18 +08:00
ViperEkura