Skip to content

Tighten public evidence framing #34

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
tsotchke merged 1 commit into from
May 21, 2026
Merged

Tighten public evidence framing #34

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
6 changes: 5 additions & 1 deletion README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -704,7 +704,11 @@ until we can repeat the same commands on a physical PC.
```
## ► Technical Innovations

Our implementation includes several innovative techniques that would have been considered groundbreaking optimizations in the 486 era. For complete technical details, see the [core innovations section](gpt2_basic_documentation.md#4-core-innovations) in our technical documentation:
The implementation combines several constrained-system techniques: fixed-point
weights, compact tokenizer files, optional compressed tensor artifacts, KV decode
caching, head shortlists, streaming fallbacks, and local recall indexes. For
complete technical details, see the [core innovations section](gpt2_basic_documentation.md#4-core-innovations)
in the technical documentation:

### ■ Current Fixed-Point Weight Format

Expand Down
3 changes: 2 additions & 1 deletion docs/marketing/video-plan.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -19,7 +19,8 @@ Target length: 4 to 6 minutes.
- Suggested caption: `Real fixed-point inference in DOS`.

2. Project thesis, 30 seconds
- Explain the alternate-history question.
- Explain the constrained-system thesis: local weights, local execution,
local recall, no network service.
- Show README title and source tree.
- Make the honest scale claim: tiny GPT-style model, not modern LLM scale.

Expand Down
Loading