deepseek-ai/DeepSeek-V4-Pro-0813
Text Generation β’ 1.7T β’ Updated β’ 49.6k β’ β’ 702
AF
capybarist
AI & ML interests
None yet
Recent Activity
liked a model 3 days ago
deepseek-ai/DeepSeek-V4-Pro-0813 liked a Space 17 days ago
openbmb/MiniCPM5-1B-Demo liked a Space 3 months ago
ngxson/wllamaOrganizations
None yet
liked a
model 3 days ago
liked a
Space 17 days ago
liked 3
Spaces 3 months ago
liked a
model 3 months ago
liked 2
Spaces 3 months ago
liked a
Space 5 months ago
liked a
Space 10 months ago
liked 2
Spaces 12 months ago
liked 3
Spaces about 1 year ago
liked 4
Spaces over 1 year ago
Chat with DeepSeek Coder 33B
π¬
315
Generate code snippets and programming help via chat
Llama3.1 405B
π
775
Generate responses with Llamaβ―3.1β―405B AI model
Qwen2.5 Coder Artifacts
π’
274
Generate and preview code from your app description
FLUXllama gpt-oss
π
562
mcp_server & FLUX 4-bit Quantization + Enhanced
reacted to anton-l's post with π over 1 year ago
Post
4206
Introducing ππ
π’π§πππππ‘: the best public math pre-training dataset with 50B+ tokens!
HuggingFaceTB/finemath
Math remains challenging for LLMs and by training on FineMath we see considerable gains over other math datasets, especially on GSM8K and MATH.
We build the dataset by:
π οΈ carefully extracting math data from Common Crawl;
π iteratively filtering and recalling high quality math pages using a classifier trained on synthetic annotations to identify math reasoning and deduction.
We conducted a series of ablations comparing the performance of Llama-3.2-3B-Base after continued pre-training on FineMath and observe notable gains compared to the baseline model and other public math datasets.
We hope this helps advance the performance of LLMs on math and reasoning! π
Weβre also releasing all the ablation models as well as the evaluation code.
HuggingFaceTB/finemath-6763fb8f71b6439b653482c2
HuggingFaceTB/finemath
Math remains challenging for LLMs and by training on FineMath we see considerable gains over other math datasets, especially on GSM8K and MATH.
We build the dataset by:
π οΈ carefully extracting math data from Common Crawl;
π iteratively filtering and recalling high quality math pages using a classifier trained on synthetic annotations to identify math reasoning and deduction.
We conducted a series of ablations comparing the performance of Llama-3.2-3B-Base after continued pre-training on FineMath and observe notable gains compared to the baseline model and other public math datasets.
We hope this helps advance the performance of LLMs on math and reasoning! π
Weβre also releasing all the ablation models as well as the evaluation code.
HuggingFaceTB/finemath-6763fb8f71b6439b653482c2