Wow that’s great news! Nice to see the continued development!
Eric PRO
Dorman11
AI & ML interests
Currently training Endeavor, a SLM model! Interests include: Agentic security, Machine Learning, Reinforcement Learning, PyTorch and all around web dev ninja! Level II Specialist working in Cyber Security field.
Recent Activity
repliedto Banaxi-Tech's post about 10 hours ago
We're excited to release BananaMind 2 SLMoE, an experimental sequence-level mixture-of-experts model.
It uses only 8M parameters per message but has 25M total parameters, 13 experts (out of 64) are selected based on the message prefix and reused for the entire response.
We're testing with this sequence-level architecture to find out how big the capability loss actually is and how much of it can be fixed.
The long-term idea is that this could make very large sparse models usable on machines that can't fit them in RAM by putting the entire model (which is big) on disk and only loading the active parts into VRAM.
This architecture is still in research and shouldn't be used for production models.
We trained it on 60B tokens (of FineWeb-HQ, FineWeb-Edu, DCLM ,Cosmopedia v2, FineMath and NPSet-2) on 8 RTX Pro 6000s.
Check it out at https://huggingface.co/BananaMind/BananaMind-2-SLMoE
Follow us for future models:
https://huggingface.co/BananaMind
@vovaRL
@Banaxi-Tech
@DedeProGames
BananaMind 2 Pro in a few days. You've been waiting 22 days for it. reacted to salma-remyx's post with 🚀 about 10 hours ago
Auto mode for coding agents has improved over the last year.
Devs are relying more on them for code review, and less on manual line-by-line checks.
But this leaves coding agents vulnerable to indirect injection attacks.
IssueTrojanBench (arxiv 2607.20759) studies this in the context of malicious issue requests.
I also observed these prompt injection vulnerabilities in my own harness, testing on the issue: https://github.com/smellslikeml/curator/issues/3
But even red teaming your own repos with toy examples like this can trigger AI safeguards and corrupt your session.
Adopting popular mitigation techniques like 'spotlighting' and 'known answer detection' can help but the researchers show it's not enough.
So our latest updates to Outrider implement additional layers for greater defense-in-depth to ensure that by time a human has to review the code, we've filtered this kind noise. updated a Space about 11 hours ago
Dorman11/open-mythos-runOrganizations
replied to Banaxi-Tech's post about 10 hours ago
reacted to salma-remyx's post with 🚀 about 10 hours ago
Post
559
Auto mode for coding agents has improved over the last year.
Devs are relying more on them for code review, and less on manual line-by-line checks.
But this leaves coding agents vulnerable to indirect injection attacks.
IssueTrojanBench (arxiv 2607.20759) studies this in the context of malicious issue requests.
I also observed these prompt injection vulnerabilities in my own harness, testing on the issue: https://github.com/smellslikeml/curator/issues/3
But even red teaming your own repos with toy examples like this can trigger AI safeguards and corrupt your session.
Adopting popular mitigation techniques like 'spotlighting' and 'known answer detection' can help but the researchers show it's not enough.
So our latest updates to Outrider implement additional layers for greater defense-in-depth to ensure that by time a human has to review the code, we've filtered this kind noise.
Devs are relying more on them for code review, and less on manual line-by-line checks.
But this leaves coding agents vulnerable to indirect injection attacks.
IssueTrojanBench (arxiv 2607.20759) studies this in the context of malicious issue requests.
I also observed these prompt injection vulnerabilities in my own harness, testing on the issue: https://github.com/smellslikeml/curator/issues/3
But even red teaming your own repos with toy examples like this can trigger AI safeguards and corrupt your session.
Adopting popular mitigation techniques like 'spotlighting' and 'known answer detection' can help but the researchers show it's not enough.
So our latest updates to Outrider implement additional layers for greater defense-in-depth to ensure that by time a human has to review the code, we've filtered this kind noise.
- 1 reply
updated a
Space about 11 hours ago
updated a
bucket 1 day ago
reacted to maglun's post with 🚀 2 days ago
Post
1151
I work on quantizing models to run efficiently on local hardware, and kept being curious how existing quants spend their "bit budget" during optimization and built a local tool to explore. Many quants apply one setting across all tensors, but some do more interesting things: the model in the screenshot holds attention K at 4.5 bits while Q/V/O get 8.5, and protects layer 0 MLP.
It turned out useful enough that I made it public: https://tensorlens.dev
Explore any HF model in the browser without downloading it. The anatomy map is read from the safetensors header via a range request, and only tensors you click ever stream. Large tensors are sampled rather than streamed in full.
Limitations: safetensors only (no GGUF yet), some exotic variants don't work yet, and gated repos aren't supported yet.
Feedback very welcome, especially models that break it.
It turned out useful enough that I made it public: https://tensorlens.dev
Explore any HF model in the browser without downloading it. The anatomy map is read from the safetensors header via a range request, and only tensors you click ever stream. Large tensors are sampled rather than streamed in full.
Limitations: safetensors only (no GGUF yet), some exotic variants don't work yet, and gated repos aren't supported yet.
Feedback very welcome, especially models that break it.
- 5 replies
This a great to use! I checked Kimi-K3 with it. Very nice man!
liked a
model 2 days ago
reacted to KlondikeDev's post with 🔥 2 days ago
Post
2121
Boris-2 coming soon!
The Models:
Boris-2-75M: Trained on 26B tokens -- estimated to start training on August 12th.
Boris-2-125M: Trained on 90B tokens -- Estimated to start training on August 20th.
Boris-2-250M: Trained on 60B tokens -- Estimated to start training on September 10th.
Why does 125M get more tokens than 250M?
Well, the straight answer is time. It saves time, while still allowing the 250M model to exceed the 125M model.
Furthermore, we are attempting a unique architecture and layering scheme to hopefully end up around the strength of SmolLM2-135M. Fingers crossed!
We hope to end up in the ballpark of AxiomicLabs/GPT-X2.5-135M or BananaMind/BananaMind-2-Pro-Preview
Following this, we will release the Pro, Instruct and Pro-Instruct variants. More info will be coming soon!
The Models:
Boris-2-75M: Trained on 26B tokens -- estimated to start training on August 12th.
Boris-2-125M: Trained on 90B tokens -- Estimated to start training on August 20th.
Boris-2-250M: Trained on 60B tokens -- Estimated to start training on September 10th.
Why does 125M get more tokens than 250M?
Well, the straight answer is time. It saves time, while still allowing the 250M model to exceed the 125M model.
Furthermore, we are attempting a unique architecture and layering scheme to hopefully end up around the strength of SmolLM2-135M. Fingers crossed!
We hope to end up in the ballpark of AxiomicLabs/GPT-X2.5-135M or BananaMind/BananaMind-2-Pro-Preview
Following this, we will release the Pro, Instruct and Pro-Instruct variants. More info will be coming soon!
- 4 replies
reacted to dddyym's post with 🔥 2 days ago
Post
1921
Released mlquantx 0.2.6: a reproducible PyTorch workflow for cross-sectional multi-factor research.
pip install --upgrade mlquantx
mlquant demo
The demo runs synthetic data → 213 factors → MLP → portfolio construction → cost-aware backtest. CI covers Python 3.9–3.11 with 96 tests.
Dataset: dddyym/ml-quant-trading-synthetic
Checkpoint: dddyym/ml-quant-trading-synthetic-mlp
GitHub: https://github.com/initial-d/ml-quant-trading
PyPI: https://pypi.org/project/mlquantx/
Research and teaching use only — not a live-trading system or a profitability claim.
pip install --upgrade mlquantx
mlquant demo
The demo runs synthetic data → 213 factors → MLP → portfolio construction → cost-aware backtest. CI covers Python 3.9–3.11 with 96 tests.
Dataset: dddyym/ml-quant-trading-synthetic
Checkpoint: dddyym/ml-quant-trading-synthetic-mlp
GitHub: https://github.com/initial-d/ml-quant-trading
PyPI: https://pypi.org/project/mlquantx/
Research and teaching use only — not a live-trading system or a profitability claim.
published a
Space 2 days ago
posted an update 2 days ago
Post
111
I've been thinking this for a long time, other than OpenMythos will we see similar type of open source models be released? The more and more top frontier models get good at finding zero day vulnerabilities and security weaknesses how will the open source model community respond. My take is we will see a open source model very soon be as good at finding security holes as the top models, at some point in the future. Is it just me or anyone seeing this?
updated a
Space 2 days ago
published a
Space 2 days ago
replied to Banaxi-Tech's post 3 days ago
BananaOS I assume could run other small models aside from BananaMind models? I’m training a small model and was just curious…either way this is a really cool idea!
liked a
model 3 days ago
published a
bucket 3 days ago
liked a
dataset 3 days ago
A simple text generation model. Would be cool to build a coding model, but that’s a challenge in and of itself.
Yeah and I’m new making models so happy to find a community! I can learn a lot here!
liked a
Space 4 days ago