Skip to content

Sourced Pythia-160M and commercial-tariff scenario packs (N4)#12

Merged
Cuuper22 merged 1 commit into
mainfrom
claude/eager-cerf-kvmqta-n4
Jun 11, 2026
Merged

Sourced Pythia-160M and commercial-tariff scenario packs (N4)#12
Cuuper22 merged 1 commit into
mainfrom
claude/eager-cerf-kvmqta-n4

Conversation

@Cuuper22

Copy link
Copy Markdown
Owner

N4 of 10: Broaden the cited scenario inventory

New gpu_stack/presets/scenarios_cited_2026.py with two pack families, every numeric carrying a public source string:

  • Pythia-160M on one DGX H100 node: architecture/workload sourced from the EleutherAI Pythia repository and Hugging Face config.json (n_layers=12, d_model=768, n_heads=12, vocab=50304, seq_len=2048, 2M-token batches, ~300B total tokens), reusing the existing sourced DGX H100 hardware facts and EIA industrial tariff. Closure assumptions live in separate, clearly named packs.
  • Pythia-70M commercial-tariff variant: substitutes the EIA 2024 U.S. commercial average retail electricity price for the industrial rate, making tariff sensitivity an explicit scenario axis.

Packs register through SOURCED_SCENARIO_PACKS and SCENARIO_TARGET_SETS (now wrapped in MappingProxyType). Statuses stay honest: open frontiers keep reporting missing inputs rather than forcing closure (closure work is PR N1's).

Also hardens the test helper from marker-based pack matching to exact-name matching so adding variants cannot make selection ambiguous.

Verification

  • Full pytest: 709 passed
  • Audit gate: PASS

Note: agent was interrupted by a session limit after implementation+tests; coordinator validated, committed, and pushed. Known merge overlap with the N1 branch in scenarios.py registration lines (resolved at integration).

https://claude.ai/code/session_01Eu2JVnPFgMQftwYTP3cGQZ


Generated by Claude Code

New gpu_stack/presets/scenarios_cited_2026.py with two pack families,
every numeric carrying a public source string:
- Pythia-160M on one DGX H100 node (EleutherAI Pythia repository and
  Hugging Face config.json for n_layers=12, d_model=768, n_heads=12,
  vocab=50304, seq_len=2048, 2M-token batches, ~300B total tokens),
  reusing the existing sourced DGX H100 hardware and EIA industrial
  tariff presets, with closure assumptions in separate named packs.
- Pythia-70M commercial-tariff variant substituting the EIA 2024 U.S.
  commercial average retail electricity price for the industrial rate,
  making tariff sensitivity explicit.

Packs register through SOURCED_SCENARIO_PACKS and SCENARIO_TARGET_SETS;
statuses stay honest (open frontiers keep reporting missing inputs).
Test helper hardened to exact-name pack matching so adding variants
cannot make marker-based selection ambiguous.

Full pytest: 709 passed. Audit gate: PASS.

https://claude.ai/code/session_01Eu2JVnPFgMQftwYTP3cGQZ
@Cuuper22 Cuuper22 merged commit bb145ec into main Jun 11, 2026
0 of 6 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants