初始化项目，由ModelHub XC社区提供模型

Model: Polygl0t/Tucano2-qwen-0.5B-Think Source: Original Platform
2026-06-04 14:20:49 +08:00
commit d2e30c7b41
22 changed files with 465611 additions and 0 deletions
--- a/.plots/apo_gradient_norm.png
+++ b/.plots/apo_gradient_norm.png
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0729b0ba79b79c3f9ae2becdfc986f9b9e7b6864e845292aaf78dc77e4535d93
+size 543980
--- a/.plots/apo_reward.png
+++ b/.plots/apo_reward.png
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a7fa6778f74ca9cf434eda73696e2cbaa1144d42a6e61827364881ff55c36065
+size 286838
--- a/.plots/model_comparison.png
+++ b/.plots/model_comparison.png
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:caa80f85de1854e5ae25089c2bd1ab0c3991c59e14d5ee290d78710fb42a5cb4
+size 222035
--- a/.plots/sft_gradient_norm.png
+++ b/.plots/sft_gradient_norm.png
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:54f1bfe1d67832bff0989e695765bfb444633bb504aa4602c240e4a00763f3d2
+size 335390
--- a/.plots/sft_loss.png
+++ b/.plots/sft_loss.png
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fb8c563e02344e1a42c0d66771ec2e6c07445b47470565ec3812aff525570d40
+size 387159