diff --git a/.gitattributes b/.gitattributes
index f38140ea016c103d1623219bc5c77d75efbfd848..4bf4a1dd01ef1c350aed40bb14bf33c275287a27 100644
--- a/.gitattributes
+++ b/.gitattributes
@@ -74,3 +74,20 @@ llama32-1b-hf/harness_eval_8shot/__home__aiops__zhuty__nanotron__checkpoints__ll
 4000_hf/harness_eval_8shot/__home__aiops__zhuty__nanotron__checkpoints__4000_hf/samples_gsm8k_2025-12-29T05-42-32.998053.jsonl filter=lfs diff=lfs merge=lfs -text
 4000_hf/harness_eval_0shot/__home__aiops__zhuty__nanotron__checkpoints__4000_hf/samples_hellaswag_2025-12-29T05-33-51.606131.jsonl filter=lfs diff=lfs merge=lfs -text
 4000_hf/harness_eval_0shot/__home__aiops__zhuty__nanotron__checkpoints__4000_hf/samples_hellaswag_2026-01-07T03-56-26.179191.jsonl filter=lfs diff=lfs merge=lfs -text
+2500_hf/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+2500_hf/harness_eval_8shot/__home__aiops__zhuty__nanotron__checkpoints__2500_hf/samples_gsm8k_2025-12-29T05-42-37.282529.jsonl filter=lfs diff=lfs merge=lfs -text
+2500_hf/harness_eval_8shot/__home__aiops__zhuty__nanotron__checkpoints__2500_hf/samples_gsm8k_2026-01-07T07-15-20.052939.jsonl filter=lfs diff=lfs merge=lfs -text
+2500_hf/harness_eval_0shot/__home__aiops__zhuty__nanotron__checkpoints__2500_hf/samples_hellaswag_2026-01-07T03-51-34.430609.jsonl filter=lfs diff=lfs merge=lfs -text
+2500_hf/harness_eval_0shot/__home__aiops__zhuty__nanotron__checkpoints__2500_hf/samples_hellaswag_2025-12-29T05-34-34.570482.jsonl filter=lfs diff=lfs merge=lfs -text
+10000_hf/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+10000_hf/harness_eval_8shot/__home__aiops__zhuty__nanotron__checkpoints__10000_hf/samples_gsm8k_2026-01-07T07-13-00.293068.jsonl filter=lfs diff=lfs merge=lfs -text
+10000_hf/harness_eval_0shot/__home__aiops__zhuty__nanotron__checkpoints__10000_hf/samples_hellaswag_2026-01-07T03-56-14.560763.jsonl filter=lfs diff=lfs merge=lfs -text
+12500_hf/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+12500_hf/harness_eval_8shot/__home__aiops__zhuty__nanotron__checkpoints__12500_hf/samples_gsm8k_2026-01-07T07-12-17.705915.jsonl filter=lfs diff=lfs merge=lfs -text
+12500_hf/harness_eval_0shot/__home__aiops__zhuty__nanotron__checkpoints__12500_hf/samples_hellaswag_2026-01-07T03-51-35.016035.jsonl filter=lfs diff=lfs merge=lfs -text
+5000_hf/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+5000_hf/harness_eval_8shot/__home__aiops__zhuty__nanotron__checkpoints__5000_hf/samples_gsm8k_2026-01-07T06-58-05.968321.jsonl filter=lfs diff=lfs merge=lfs -text
+5000_hf/harness_eval_0shot/__home__aiops__zhuty__nanotron__checkpoints__5000_hf/samples_hellaswag_2026-01-07T03-53-49.355990.jsonl filter=lfs diff=lfs merge=lfs -text
+7500_hf/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+7500_hf/harness_eval_8shot/__home__aiops__zhuty__nanotron__checkpoints__7500_hf/samples_gsm8k_2026-01-07T07-00-15.435455.jsonl filter=lfs diff=lfs merge=lfs -text
+7500_hf/harness_eval_0shot/__home__aiops__zhuty__nanotron__checkpoints__7500_hf/samples_hellaswag_2026-01-07T03-51-34.104451.jsonl filter=lfs diff=lfs merge=lfs -text
diff --git a/10000_hf/harness_eval_0shot/__home__aiops__zhuty__nanotron__checkpoints__10000_hf/samples_hellaswag_2026-01-07T03-56-14.560763.jsonl b/10000_hf/harness_eval_0shot/__home__aiops__zhuty__nanotron__checkpoints__10000_hf/samples_hellaswag_2026-01-07T03-56-14.560763.jsonl
new file mode 100644
index 0000000000000000000000000000000000000000..99941d4e1a5379b7f2873ae5649b8484653b23b3
--- /dev/null
+++ b/10000_hf/harness_eval_0shot/__home__aiops__zhuty__nanotron__checkpoints__10000_hf/samples_hellaswag_2026-01-07T03-56-14.560763.jsonl
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:019283f25ac52ae31c6cf25385f1c39868a83c840ee685eeaace39a373d64ec1
+size 42645134
diff --git a/10000_hf/harness_eval_8shot/__home__aiops__zhuty__nanotron__checkpoints__10000_hf/samples_gsm8k_2026-01-07T07-13-00.293068.jsonl b/10000_hf/harness_eval_8shot/__home__aiops__zhuty__nanotron__checkpoints__10000_hf/samples_gsm8k_2026-01-07T07-13-00.293068.jsonl
new file mode 100644
index 0000000000000000000000000000000000000000..2e3a3ef3f0e1c1844a20d6c02915abe0e6c3028a
--- /dev/null
+++ b/10000_hf/harness_eval_8shot/__home__aiops__zhuty__nanotron__checkpoints__10000_hf/samples_gsm8k_2026-01-07T07-13-00.293068.jsonl
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:979f960ef99474ceb5be3a2b15c727db71dd0ddd0d62c82e73abff3a58077fc7
+size 16725724
diff --git a/10000_hf/model.safetensors b/10000_hf/model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..25bb9f36d6b781f4e45528fead09205cc13dc33f
--- /dev/null
+++ b/10000_hf/model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5c4f056bd3db4b0663bcc79008263ed8fc6ab3b2061f578ae12e6361cd47e3b0
+size 2471645608
diff --git a/10000_hf/tokenizer.json b/10000_hf/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..1c1d8d5c9024994f1d3b00f9662b8dd89ca13cf2
--- /dev/null
+++ b/10000_hf/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6b9e4e7fb171f92fd137b777cc2714bf87d11576700a1dcd7a399e7bbe39537b
+size 17209920
diff --git a/12500/lr_scheduler/lr_scheduler_pp-0-of-1_tp-0-of-1_exp-0-of-1.pt b/12500/lr_scheduler/lr_scheduler_pp-0-of-1_tp-0-of-1_exp-0-of-1.pt
new file mode 100644
index 0000000000000000000000000000000000000000..5647209edf9921f853dd7c96654442ba9a569024
--- /dev/null
+++ b/12500/lr_scheduler/lr_scheduler_pp-0-of-1_tp-0-of-1_exp-0-of-1.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:374278ee7760e355e5f7d19924fad3e4fca7d2c8c0e8ce04f9497d4c9091bc23
+size 3781
diff --git a/12500/model/model/decoder/0/pp_block/attn/o_proj/model_weight_pp-rank-0-of-1_tp-rank-0-of-1.safetensors b/12500/model/model/decoder/0/pp_block/attn/o_proj/model_weight_pp-rank-0-of-1_tp-rank-0-of-1.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..d04bd0e387ea995c0e166ab3c6f87a3ec933a546
--- /dev/null
+++ b/12500/model/model/decoder/0/pp_block/attn/o_proj/model_weight_pp-rank-0-of-1_tp-rank-0-of-1.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:33f64f9580bc254ac7da5e2a4f85508ee532ba1941a624e1765bc40b32c3b936
+size 8388848
diff --git a/12500/model/model/decoder/0/pp_block/attn/qkv_proj/model_weight_pp-rank-0-of-1_tp-rank-0-of-1.safetensors b/12500/model/model/decoder/0/pp_block/attn/qkv_proj/model_weight_pp-rank-0-of-1_tp-rank-0-of-1.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..afb9dcc85681ed986f313a79d3249b643ccaa718
--- /dev/null
+++ b/12500/model/model/decoder/0/pp_block/attn/qkv_proj/model_weight_pp-rank-0-of-1_tp-rank-0-of-1.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:755297f607e1208065ec25c006290724f27b2a048deb4f2b0447ebb6a97b64a3
+size 12583280
diff --git a/12500/model/model/decoder/0/pp_block/input_layernorm/model_weight.safetensors b/12500/model/model/decoder/0/pp_block/input_layernorm/model_weight.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..03844937616f5ce129c2593fa4a04f47e9ea70ca
--- /dev/null
+++ b/12500/model/model/decoder/0/pp_block/input_layernorm/model_weight.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0b9320d79d9ad3cb9751098131b608a50be6ccd2e6c22d3311f38c3ec6c4adde
+size 4192
diff --git a/12500/model/model/decoder/0/pp_block/mlp/down_proj/model_weight_pp-rank-0-of-1_tp-rank-0-of-1.safetensors b/12500/model/model/decoder/0/pp_block/mlp/down_proj/model_weight_pp-rank-0-of-1_tp-rank-0-of-1.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..cebda5ec40be785e32d8273a4e210818de526116
--- /dev/null
+++ b/12500/model/model/decoder/0/pp_block/mlp/down_proj/model_weight_pp-rank-0-of-1_tp-rank-0-of-1.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:768b30c5809543dcb7e2b3a80acebcccc2386d5951858ce96f659472c87edbbd
+size 33554672
diff --git a/12500/model/model/decoder/0/pp_block/mlp/gate_up_proj/model_weight_pp-rank-0-of-1_tp-rank-0-of-1.safetensors b/12500/model/model/decoder/0/pp_block/mlp/gate_up_proj/model_weight_pp-rank-0-of-1_tp-rank-0-of-1.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..e81658f19f88c6f21292bf3e3397f09457ba6e44
--- /dev/null
+++ b/12500/model/model/decoder/0/pp_block/mlp/gate_up_proj/model_weight_pp-rank-0-of-1_tp-rank-0-of-1.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d475b067243674f9391064be95d0b62095dd499de017a4cf54e4ced3a80a4967
+size 67109176
diff --git a/12500/model/model/decoder/0/pp_block/post_attention_layernorm/model_weight.safetensors b/12500/model/model/decoder/0/pp_block/post_attention_layernorm/model_weight.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..87b3f07935b1e09ae7930ace5ffc89eecc7aa2a2
--- /dev/null
+++ b/12500/model/model/decoder/0/pp_block/post_attention_layernorm/model_weight.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:286b94d6f2348d14553975a37a13790f35b17394a1319ab681d8e2a53d3b4e32
+size 4192
diff --git a/12500/model/model/decoder/1/pp_block/attn/o_proj/model_weight_pp-rank-0-of-1_tp-rank-0-of-1.safetensors b/12500/model/model/decoder/1/pp_block/attn/o_proj/model_weight_pp-rank-0-of-1_tp-rank-0-of-1.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..c9fe22a903f72a01791032aad4febe450da64409
--- /dev/null
+++ b/12500/model/model/decoder/1/pp_block/attn/o_proj/model_weight_pp-rank-0-of-1_tp-rank-0-of-1.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f4a182a893a8659c7ac62516949ac6ec65ca8d8b18b981b45c7591308afe7121
+size 8388848
diff --git a/12500/model/model/decoder/1/pp_block/attn/qkv_proj/model_weight_pp-rank-0-of-1_tp-rank-0-of-1.safetensors b/12500/model/model/decoder/1/pp_block/attn/qkv_proj/model_weight_pp-rank-0-of-1_tp-rank-0-of-1.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..0a2fa31a9f1b981bd0939e40272f139210827b45
--- /dev/null
+++ b/12500/model/model/decoder/1/pp_block/attn/qkv_proj/model_weight_pp-rank-0-of-1_tp-rank-0-of-1.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:69fc59684b34ca689e9369bce889aad4d0a60bc151ff8a8cabf010986280a13a
+size 12583280
diff --git a/12500/model/model/decoder/1/pp_block/input_layernorm/model_weight.safetensors b/12500/model/model/decoder/1/pp_block/input_layernorm/model_weight.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..37495c04e0b6ff0ad1706966a503e737cd158327
--- /dev/null
+++ b/12500/model/model/decoder/1/pp_block/input_layernorm/model_weight.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:01eb90d8439f20c3259d5b2f6b87ac7d30931a2a8a0ac4ea2f066d0544c19e40
+size 4192
diff --git a/12500/model/model/decoder/1/pp_block/mlp/down_proj/model_weight_pp-rank-0-of-1_tp-rank-0-of-1.safetensors b/12500/model/model/decoder/1/pp_block/mlp/down_proj/model_weight_pp-rank-0-of-1_tp-rank-0-of-1.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..124437e4d4d42b8f5465932a1b1937732a71dcd1
--- /dev/null
+++ b/12500/model/model/decoder/1/pp_block/mlp/down_proj/model_weight_pp-rank-0-of-1_tp-rank-0-of-1.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:be64a1076bdfd9651c17ca8a05384ce161af29542848c980bf9ef32fa422058d
+size 33554672
diff --git a/12500/model/model/decoder/1/pp_block/mlp/gate_up_proj/model_weight_pp-rank-0-of-1_tp-rank-0-of-1.safetensors b/12500/model/model/decoder/1/pp_block/mlp/gate_up_proj/model_weight_pp-rank-0-of-1_tp-rank-0-of-1.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..ac3f460835cfb14fb429834b1d7374e49318bd5f
--- /dev/null
+++ b/12500/model/model/decoder/1/pp_block/mlp/gate_up_proj/model_weight_pp-rank-0-of-1_tp-rank-0-of-1.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e27ba8d73e7115d6ae7d8a05bd6201dc5c975d068d207f14cd95f52d619758a2
+size 67109176
diff --git a/12500/model/model/decoder/1/pp_block/post_attention_layernorm/model_weight.safetensors b/12500/model/model/decoder/1/pp_block/post_attention_layernorm/model_weight.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..98aa037d5f57c1a330853526faca493f715394aa
--- /dev/null
+++ b/12500/model/model/decoder/1/pp_block/post_attention_layernorm/model_weight.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:380913079b683588e3514ec46b60d2bd75699a26230effec7a24ac405831b63d
+size 4192
diff --git a/12500/model/model/decoder/12/pp_block/attn/o_proj/model_weight_pp-rank-0-of-1_tp-rank-0-of-1.safetensors b/12500/model/model/decoder/12/pp_block/attn/o_proj/model_weight_pp-rank-0-of-1_tp-rank-0-of-1.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..1e8d3917bf202ff81b8e928a79383b58712682c2
--- /dev/null
+++ b/12500/model/model/decoder/12/pp_block/attn/o_proj/model_weight_pp-rank-0-of-1_tp-rank-0-of-1.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:932cd453fe7d9cb054ecf9dfe50dc39c4e7fac374e847152601dff3896b382c3
+size 8388848
diff --git a/12500/model/model/decoder/12/pp_block/attn/qkv_proj/model_weight_pp-rank-0-of-1_tp-rank-0-of-1.safetensors b/12500/model/model/decoder/12/pp_block/attn/qkv_proj/model_weight_pp-rank-0-of-1_tp-rank-0-of-1.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..5536c291877c9a01a6f2180448d632c738297f34
--- /dev/null
+++ b/12500/model/model/decoder/12/pp_block/attn/qkv_proj/model_weight_pp-rank-0-of-1_tp-rank-0-of-1.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e5bc299ecea6517951175b8fa6fdcc1811ff224448f8658645e51133b175ee10
+size 12583280
diff --git a/12500/model/model/decoder/12/pp_block/input_layernorm/model_weight.safetensors b/12500/model/model/decoder/12/pp_block/input_layernorm/model_weight.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..de4f4885af57c9c16764d099813016df5f890f33
--- /dev/null
+++ b/12500/model/model/decoder/12/pp_block/input_layernorm/model_weight.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f40df8e3ee1734d75dc14a0708ce9f4f47a01635c96f49fac9e8b6c6142e9a1f
+size 4192
diff --git a/12500/model/model/decoder/12/pp_block/mlp/down_proj/model_weight_pp-rank-0-of-1_tp-rank-0-of-1.safetensors b/12500/model/model/decoder/12/pp_block/mlp/down_proj/model_weight_pp-rank-0-of-1_tp-rank-0-of-1.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..691d2e832158e71db533d2f0ca44879d137e38c6
--- /dev/null
+++ b/12500/model/model/decoder/12/pp_block/mlp/down_proj/model_weight_pp-rank-0-of-1_tp-rank-0-of-1.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f2547623d0a135fc578cf37237581e8c40a9035a80ac3cf47c3567c17fa35194
+size 33554672
diff --git a/12500/model/model/decoder/12/pp_block/mlp/gate_up_proj/model_weight_pp-rank-0-of-1_tp-rank-0-of-1.safetensors b/12500/model/model/decoder/12/pp_block/mlp/gate_up_proj/model_weight_pp-rank-0-of-1_tp-rank-0-of-1.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..2161ec0b7ac8533fd5e29e1fcb8e0c5ef1bd6916
--- /dev/null
+++ b/12500/model/model/decoder/12/pp_block/mlp/gate_up_proj/model_weight_pp-rank-0-of-1_tp-rank-0-of-1.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5fadd4d4606f22244fc7141671bc4e41ee2a39fed5bb59dba76988a95d85f355
+size 67109176
diff --git a/12500/model/model/decoder/12/pp_block/post_attention_layernorm/model_weight.safetensors b/12500/model/model/decoder/12/pp_block/post_attention_layernorm/model_weight.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..a7e1fde475322cff5ca4d166d64f9cce151c9e7d
--- /dev/null
+++ b/12500/model/model/decoder/12/pp_block/post_attention_layernorm/model_weight.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:61a8ea473c8821bfbb047c94ee32b7fc508774b0f035130b1a53929866c29526
+size 4192
diff --git a/12500/model/model/decoder/13/pp_block/attn/o_proj/model_weight_pp-rank-0-of-1_tp-rank-0-of-1.safetensors b/12500/model/model/decoder/13/pp_block/attn/o_proj/model_weight_pp-rank-0-of-1_tp-rank-0-of-1.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..ff679908fd31b3e031aa7f688a6294cfa35d70e6
--- /dev/null
+++ b/12500/model/model/decoder/13/pp_block/attn/o_proj/model_weight_pp-rank-0-of-1_tp-rank-0-of-1.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3f82ecb4e76af4b6984f542f1c2d06a922544b1d2c649a5c456c8eb3010a44ee
+size 8388848
diff --git a/12500/model/model/decoder/13/pp_block/attn/qkv_proj/model_weight_pp-rank-0-of-1_tp-rank-0-of-1.safetensors b/12500/model/model/decoder/13/pp_block/attn/qkv_proj/model_weight_pp-rank-0-of-1_tp-rank-0-of-1.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..37609162c8bea050a47de267d5cb941c4c4d1e44
--- /dev/null
+++ b/12500/model/model/decoder/13/pp_block/attn/qkv_proj/model_weight_pp-rank-0-of-1_tp-rank-0-of-1.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ca1919e6460967b14a09695b74b10b4aad5574ab54a8702ac67de8f4f6b6529e
+size 12583280
diff --git a/12500/model/model/decoder/13/pp_block/input_layernorm/model_weight.safetensors b/12500/model/model/decoder/13/pp_block/input_layernorm/model_weight.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..7c7e386b62973bfaa7d6c08cee2d027762a9e4c7
--- /dev/null
+++ b/12500/model/model/decoder/13/pp_block/input_layernorm/model_weight.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:13254580fbbaeeccd0abe755649ee68566304272e594be3d5d8dfc7385d58eeb
+size 4192
diff --git a/12500/model/model/decoder/13/pp_block/mlp/down_proj/model_weight_pp-rank-0-of-1_tp-rank-0-of-1.safetensors b/12500/model/model/decoder/13/pp_block/mlp/down_proj/model_weight_pp-rank-0-of-1_tp-rank-0-of-1.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..261f2909bdfe5702dc51ea0d9d03375985abfa34
--- /dev/null
+++ b/12500/model/model/decoder/13/pp_block/mlp/down_proj/model_weight_pp-rank-0-of-1_tp-rank-0-of-1.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5deea6376637f31c271ce8f09ed5064aa781d3fd3caad7d7f8f27d0f21e54320
+size 33554672
diff --git a/12500/model/model/decoder/13/pp_block/mlp/gate_up_proj/model_weight_pp-rank-0-of-1_tp-rank-0-of-1.safetensors b/12500/model/model/decoder/13/pp_block/mlp/gate_up_proj/model_weight_pp-rank-0-of-1_tp-rank-0-of-1.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..119b532250b8123a151dac5075edd22de6ebda68
--- /dev/null
+++ b/12500/model/model/decoder/13/pp_block/mlp/gate_up_proj/model_weight_pp-rank-0-of-1_tp-rank-0-of-1.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b80e832585909c699d214ac6c9ddbce83dae96cbc10f81a2cb0055aac7d62828
+size 67109176
diff --git a/12500/model/model/decoder/13/pp_block/post_attention_layernorm/model_weight.safetensors b/12500/model/model/decoder/13/pp_block/post_attention_layernorm/model_weight.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..4943692b91081af4f36aee35f4529edf607d2167
--- /dev/null
+++ b/12500/model/model/decoder/13/pp_block/post_attention_layernorm/model_weight.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3101d7f62ad0477c4d4bdb8220bc62c95e80b309367fb39b1a189a392cebc4a4
+size 4192
diff --git a/12500/model/model/decoder/15/pp_block/post_attention_layernorm/model_weight.safetensors b/12500/model/model/decoder/15/pp_block/post_attention_layernorm/model_weight.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..a7e7e39707349ee7d9d542c89c91954cb53a44ab
--- /dev/null
+++ b/12500/model/model/decoder/15/pp_block/post_attention_layernorm/model_weight.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5a1e6f9470546f244298688febfacdcc6fba40ab32e6f5cc58d9f7f85cbbc999
+size 4192
diff --git a/12500/model/model/decoder/4/pp_block/attn/o_proj/model_weight_pp-rank-0-of-1_tp-rank-0-of-1.safetensors b/12500/model/model/decoder/4/pp_block/attn/o_proj/model_weight_pp-rank-0-of-1_tp-rank-0-of-1.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..2211def76f6ecf0835a053a3a53db113e270646e
--- /dev/null
+++ b/12500/model/model/decoder/4/pp_block/attn/o_proj/model_weight_pp-rank-0-of-1_tp-rank-0-of-1.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4b9798d69b2bad5743aab3eb5f6697449ed6c05ee622d4ba5dd9e306ac2ad947
+size 8388848
diff --git a/12500/model/model/decoder/4/pp_block/attn/qkv_proj/model_weight_pp-rank-0-of-1_tp-rank-0-of-1.safetensors b/12500/model/model/decoder/4/pp_block/attn/qkv_proj/model_weight_pp-rank-0-of-1_tp-rank-0-of-1.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..e5873e45a741bf06d4c272981f373b03b70a0985
--- /dev/null
+++ b/12500/model/model/decoder/4/pp_block/attn/qkv_proj/model_weight_pp-rank-0-of-1_tp-rank-0-of-1.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:24104016d3c6d8552f9359d27350ed190bdb33b5bf666e362f50a9fe8f100bfb
+size 12583280
diff --git a/12500/model/model/decoder/4/pp_block/input_layernorm/model_weight.safetensors b/12500/model/model/decoder/4/pp_block/input_layernorm/model_weight.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..7d361e12ab6bb1d749a2d8ce103a7f18c13019d6
--- /dev/null
+++ b/12500/model/model/decoder/4/pp_block/input_layernorm/model_weight.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3d4542ddadef586703537eddfac6766f609e45cf2195a97647dc29bc3acff1a2
+size 4192
diff --git a/12500/model/model/decoder/4/pp_block/mlp/down_proj/model_weight_pp-rank-0-of-1_tp-rank-0-of-1.safetensors b/12500/model/model/decoder/4/pp_block/mlp/down_proj/model_weight_pp-rank-0-of-1_tp-rank-0-of-1.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..ea0d3b336965c8167d663ee7ab4acf237f5b1ea5
--- /dev/null
+++ b/12500/model/model/decoder/4/pp_block/mlp/down_proj/model_weight_pp-rank-0-of-1_tp-rank-0-of-1.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5f40041638bf95fd0c6acc4fc6c617c35952560fd3b3d6f8bb91c005eab9c031
+size 33554672
diff --git a/12500/model/model/decoder/4/pp_block/mlp/gate_up_proj/model_weight_pp-rank-0-of-1_tp-rank-0-of-1.safetensors b/12500/model/model/decoder/4/pp_block/mlp/gate_up_proj/model_weight_pp-rank-0-of-1_tp-rank-0-of-1.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..71df9b01b85721d7ab48444f697dc68af8ba0ab9
--- /dev/null
+++ b/12500/model/model/decoder/4/pp_block/mlp/gate_up_proj/model_weight_pp-rank-0-of-1_tp-rank-0-of-1.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:957438630a769c12814f4973d984c8e9fe80b04fcf2d7dc7561bba204f19da40
+size 67109176
diff --git a/12500/model/model/decoder/4/pp_block/post_attention_layernorm/model_weight.safetensors b/12500/model/model/decoder/4/pp_block/post_attention_layernorm/model_weight.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..1e439004dae19f7653bf5f2a871fdef3c7a98663
--- /dev/null
+++ b/12500/model/model/decoder/4/pp_block/post_attention_layernorm/model_weight.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6a57aa27c2d8e5b206d4d2e995013ddab23daaaf736f762bbc4c9450d20bd19a
+size 4192
diff --git a/12500/model/model/decoder/5/pp_block/attn/o_proj/model_weight_pp-rank-0-of-1_tp-rank-0-of-1.safetensors b/12500/model/model/decoder/5/pp_block/attn/o_proj/model_weight_pp-rank-0-of-1_tp-rank-0-of-1.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..6e1d342ebc56498c6e1291f27b95affb82ff8448
--- /dev/null
+++ b/12500/model/model/decoder/5/pp_block/attn/o_proj/model_weight_pp-rank-0-of-1_tp-rank-0-of-1.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:888b959cf5a5b1f8bb3db748bdcb86a3e72d279de6ab84abf19e8f20252d3de5
+size 8388848
diff --git a/12500/model/model/decoder/5/pp_block/attn/qkv_proj/model_weight_pp-rank-0-of-1_tp-rank-0-of-1.safetensors b/12500/model/model/decoder/5/pp_block/attn/qkv_proj/model_weight_pp-rank-0-of-1_tp-rank-0-of-1.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..d72bae8580b1a806d7af870c88b87425a05ba0bd
--- /dev/null
+++ b/12500/model/model/decoder/5/pp_block/attn/qkv_proj/model_weight_pp-rank-0-of-1_tp-rank-0-of-1.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:232b95e7c6429c17708b9fae7f8ce3e2d91ee9ce811710459062d2866f921dc4
+size 12583280
diff --git a/12500/model/model/decoder/5/pp_block/input_layernorm/model_weight.safetensors b/12500/model/model/decoder/5/pp_block/input_layernorm/model_weight.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..abf8689a6077332d34e9a4e1f6fc3377ebd6ec59
--- /dev/null
+++ b/12500/model/model/decoder/5/pp_block/input_layernorm/model_weight.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9095db1de1d00abdf92a45573dbd0f2501a12f075124f4b8c934e3ce4dc4b5e1
+size 4192
diff --git a/12500/model/model/decoder/5/pp_block/mlp/down_proj/model_weight_pp-rank-0-of-1_tp-rank-0-of-1.safetensors b/12500/model/model/decoder/5/pp_block/mlp/down_proj/model_weight_pp-rank-0-of-1_tp-rank-0-of-1.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..336b80baa78b0a6549c5153878d946fb0d717506
--- /dev/null
+++ b/12500/model/model/decoder/5/pp_block/mlp/down_proj/model_weight_pp-rank-0-of-1_tp-rank-0-of-1.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:635d465cae92399da192b18274bbeb7b88b214e22cc4b87325904e7c68010ff1
+size 33554672
diff --git a/12500/model/model/decoder/5/pp_block/mlp/gate_up_proj/model_weight_pp-rank-0-of-1_tp-rank-0-of-1.safetensors b/12500/model/model/decoder/5/pp_block/mlp/gate_up_proj/model_weight_pp-rank-0-of-1_tp-rank-0-of-1.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..92ed7f1f2a76b613703b3e2de87b7b08c2279835
--- /dev/null
+++ b/12500/model/model/decoder/5/pp_block/mlp/gate_up_proj/model_weight_pp-rank-0-of-1_tp-rank-0-of-1.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:52f13cf5aeadc3783aa55e8c9e9396d1bdf79a1074ea9d0535c13ef41c961658
+size 67109176
diff --git a/12500/model/model/decoder/5/pp_block/post_attention_layernorm/model_weight.safetensors b/12500/model/model/decoder/5/pp_block/post_attention_layernorm/model_weight.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..d17b28a8c59079ed39fe6cd8cd8cd216a4bf2582
--- /dev/null
+++ b/12500/model/model/decoder/5/pp_block/post_attention_layernorm/model_weight.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6f2c1a4139db805a97468a1324b7be8ffc6cb7a4bb3352a596349d2b812ab53a
+size 4192
diff --git a/12500/optimizer/optimizer_pp-0-of-1_tp-0-of-1_exp-0-of-1.pt b/12500/optimizer/optimizer_pp-0-of-1_tp-0-of-1_exp-0-of-1.pt
new file mode 100644
index 0000000000000000000000000000000000000000..dbb9445f168c270277a97ea94fc7b2e80f107343
--- /dev/null
+++ b/12500/optimizer/optimizer_pp-0-of-1_tp-0-of-1_exp-0-of-1.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d0f2317e040a045732db10e89cf99fa3404d9528f51eb267e7409e484e749b92
+size 14829896970
diff --git a/12500/random/tp-0-of-1_dp-0-of-8_pp-0-of-1.pt b/12500/random/tp-0-of-1_dp-0-of-8_pp-0-of-1.pt
new file mode 100644
index 0000000000000000000000000000000000000000..15bb9782bdd1530ae232fc73387a84decc9195ee
--- /dev/null
+++ b/12500/random/tp-0-of-1_dp-0-of-8_pp-0-of-1.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0601ec41ec35ebf4d71aa340e9ad8eec6d97496febf0c82d4b906408c13f4cb3
+size 1521
diff --git a/12500/random/tp-0-of-1_dp-1-of-8_pp-0-of-1.pt b/12500/random/tp-0-of-1_dp-1-of-8_pp-0-of-1.pt
new file mode 100644
index 0000000000000000000000000000000000000000..a9fd73c1278e11ea0cac6f58ca0de36910f9b70d
--- /dev/null
+++ b/12500/random/tp-0-of-1_dp-1-of-8_pp-0-of-1.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:631c9eaddef421b77417ad6de8caa9b73ce2bcba3028ae643ece004e415842b3
+size 1521
diff --git a/12500/random/tp-0-of-1_dp-2-of-8_pp-0-of-1.pt b/12500/random/tp-0-of-1_dp-2-of-8_pp-0-of-1.pt
new file mode 100644
index 0000000000000000000000000000000000000000..0373c058254c042a3252e06381109f6a90a3d1ae
--- /dev/null
+++ b/12500/random/tp-0-of-1_dp-2-of-8_pp-0-of-1.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9154057cc18766552e3716cb2ea909fe597fbf9bf5bbba55f53e4392660fd471
+size 1521
diff --git a/12500/random/tp-0-of-1_dp-3-of-8_pp-0-of-1.pt b/12500/random/tp-0-of-1_dp-3-of-8_pp-0-of-1.pt
new file mode 100644
index 0000000000000000000000000000000000000000..7439ec735c9cd78d6ca4965a30820eeaad5b546d
--- /dev/null
+++ b/12500/random/tp-0-of-1_dp-3-of-8_pp-0-of-1.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:50a75e874575cdf39fe4d6ab9b0eadf99af310b441c93ec648e504f354a32782
+size 1521
diff --git a/12500/random/tp-0-of-1_dp-4-of-8_pp-0-of-1.pt b/12500/random/tp-0-of-1_dp-4-of-8_pp-0-of-1.pt
new file mode 100644
index 0000000000000000000000000000000000000000..e6feb2bf6a2b27e8013b251ce97e802479b35a33
--- /dev/null
+++ b/12500/random/tp-0-of-1_dp-4-of-8_pp-0-of-1.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6f3a3b3b43fbfa9f6f526bfd0438f1beb8312f8bb70ed1916db7b4ec6db049a8
+size 1521
diff --git a/12500/random/tp-0-of-1_dp-5-of-8_pp-0-of-1.pt b/12500/random/tp-0-of-1_dp-5-of-8_pp-0-of-1.pt
new file mode 100644
index 0000000000000000000000000000000000000000..5353410c2720a6537ce7258058ecae9456b63ce2
--- /dev/null
+++ b/12500/random/tp-0-of-1_dp-5-of-8_pp-0-of-1.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5e00f9afebb28b3eebef1362210f10067a02cd411854be30e8d0a459122294ff
+size 1521
diff --git a/12500/random/tp-0-of-1_dp-6-of-8_pp-0-of-1.pt b/12500/random/tp-0-of-1_dp-6-of-8_pp-0-of-1.pt
new file mode 100644
index 0000000000000000000000000000000000000000..383d9e71229d622ee8c8ac18c15e95d4a4a40e77
--- /dev/null
+++ b/12500/random/tp-0-of-1_dp-6-of-8_pp-0-of-1.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cee8d18e682345ace533c1edc313e2e0abe82d4b60dba80152bb4d53e4f4e753
+size 1521
diff --git a/12500/random/tp-0-of-1_dp-7-of-8_pp-0-of-1.pt b/12500/random/tp-0-of-1_dp-7-of-8_pp-0-of-1.pt
new file mode 100644
index 0000000000000000000000000000000000000000..563a558fd1bfb79f3ae69dd9ca9e23cc780317ee
--- /dev/null
+++ b/12500/random/tp-0-of-1_dp-7-of-8_pp-0-of-1.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8151d92acfbdd2ef1aff82693896ddcb2ffc4b1b18ef05fb15f2dfc55b883250
+size 1521
diff --git a/12500_hf/harness_eval_0shot/__home__aiops__zhuty__nanotron__checkpoints__12500_hf/samples_hellaswag_2026-01-07T03-51-35.016035.jsonl b/12500_hf/harness_eval_0shot/__home__aiops__zhuty__nanotron__checkpoints__12500_hf/samples_hellaswag_2026-01-07T03-51-35.016035.jsonl
new file mode 100644
index 0000000000000000000000000000000000000000..e398511405bca31af04e1beaa07e02f7aaa58ccc
--- /dev/null
+++ b/12500_hf/harness_eval_0shot/__home__aiops__zhuty__nanotron__checkpoints__12500_hf/samples_hellaswag_2026-01-07T03-51-35.016035.jsonl
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d4e88b434464d264983d46811848835622f0bf77b41c0fe8db74903d5b7c9fc3
+size 42645324
diff --git a/12500_hf/harness_eval_8shot/__home__aiops__zhuty__nanotron__checkpoints__12500_hf/samples_gsm8k_2026-01-07T07-12-17.705915.jsonl b/12500_hf/harness_eval_8shot/__home__aiops__zhuty__nanotron__checkpoints__12500_hf/samples_gsm8k_2026-01-07T07-12-17.705915.jsonl
new file mode 100644
index 0000000000000000000000000000000000000000..d9808d3584bfca521d2b7c88751e1edb1ea9d638
--- /dev/null
+++ b/12500_hf/harness_eval_8shot/__home__aiops__zhuty__nanotron__checkpoints__12500_hf/samples_gsm8k_2026-01-07T07-12-17.705915.jsonl
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:94cf7ea2bdf5856f6519081bab85948697f36de9c05b0e4f8f18d61503e7cece
+size 16769501
diff --git a/12500_hf/model.safetensors b/12500_hf/model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..0d7088d69af061874275681999af0bc4288c7634
--- /dev/null
+++ b/12500_hf/model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:93f622dca917b44f638938c5dbd4c778f0fea70e6f6a1a9bbecd411613728c6e
+size 2471645608
diff --git a/12500_hf/tokenizer.json b/12500_hf/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..1c1d8d5c9024994f1d3b00f9662b8dd89ca13cf2
--- /dev/null
+++ b/12500_hf/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6b9e4e7fb171f92fd137b777cc2714bf87d11576700a1dcd7a399e7bbe39537b
+size 17209920
diff --git a/2500_hf/harness_eval_0shot/__home__aiops__zhuty__nanotron__checkpoints__2500_hf/samples_hellaswag_2025-12-29T05-34-34.570482.jsonl b/2500_hf/harness_eval_0shot/__home__aiops__zhuty__nanotron__checkpoints__2500_hf/samples_hellaswag_2025-12-29T05-34-34.570482.jsonl
new file mode 100644
index 0000000000000000000000000000000000000000..03df91f7a08be3c959a7398b510617f69e946e0f
--- /dev/null
+++ b/2500_hf/harness_eval_0shot/__home__aiops__zhuty__nanotron__checkpoints__2500_hf/samples_hellaswag_2025-12-29T05-34-34.570482.jsonl
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4ebd80cdba08d784fe951602282d5f79c84bde9280e821a54d817305a75d8ef8
+size 42643896
diff --git a/2500_hf/harness_eval_0shot/__home__aiops__zhuty__nanotron__checkpoints__2500_hf/samples_hellaswag_2026-01-07T03-51-34.430609.jsonl b/2500_hf/harness_eval_0shot/__home__aiops__zhuty__nanotron__checkpoints__2500_hf/samples_hellaswag_2026-01-07T03-51-34.430609.jsonl
new file mode 100644
index 0000000000000000000000000000000000000000..03df91f7a08be3c959a7398b510617f69e946e0f
--- /dev/null
+++ b/2500_hf/harness_eval_0shot/__home__aiops__zhuty__nanotron__checkpoints__2500_hf/samples_hellaswag_2026-01-07T03-51-34.430609.jsonl
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4ebd80cdba08d784fe951602282d5f79c84bde9280e821a54d817305a75d8ef8
+size 42643896
diff --git a/2500_hf/harness_eval_8shot/__home__aiops__zhuty__nanotron__checkpoints__2500_hf/samples_gsm8k_2025-12-29T05-42-37.282529.jsonl b/2500_hf/harness_eval_8shot/__home__aiops__zhuty__nanotron__checkpoints__2500_hf/samples_gsm8k_2025-12-29T05-42-37.282529.jsonl
new file mode 100644
index 0000000000000000000000000000000000000000..313eae0a5c7b2ce5565605b6c81860d3d151755b
--- /dev/null
+++ b/2500_hf/harness_eval_8shot/__home__aiops__zhuty__nanotron__checkpoints__2500_hf/samples_gsm8k_2025-12-29T05-42-37.282529.jsonl
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b704f2e643956447d5959dd8bd02a236c70949f3082c0e9540a7e15452722117
+size 16776776
diff --git a/2500_hf/harness_eval_8shot/__home__aiops__zhuty__nanotron__checkpoints__2500_hf/samples_gsm8k_2026-01-07T07-15-20.052939.jsonl b/2500_hf/harness_eval_8shot/__home__aiops__zhuty__nanotron__checkpoints__2500_hf/samples_gsm8k_2026-01-07T07-15-20.052939.jsonl
new file mode 100644
index 0000000000000000000000000000000000000000..313eae0a5c7b2ce5565605b6c81860d3d151755b
--- /dev/null
+++ b/2500_hf/harness_eval_8shot/__home__aiops__zhuty__nanotron__checkpoints__2500_hf/samples_gsm8k_2026-01-07T07-15-20.052939.jsonl
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b704f2e643956447d5959dd8bd02a236c70949f3082c0e9540a7e15452722117
+size 16776776
diff --git a/2500_hf/model.safetensors b/2500_hf/model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..197d76e2e593e60fee8d75414945a4bac8de5f17
--- /dev/null
+++ b/2500_hf/model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dcb19733f1e9aee21113269f821b2fd759a2767a8968a5a0a108d056fdcc54df
+size 2471645608
diff --git a/2500_hf/tokenizer.json b/2500_hf/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..1c1d8d5c9024994f1d3b00f9662b8dd89ca13cf2
--- /dev/null
+++ b/2500_hf/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6b9e4e7fb171f92fd137b777cc2714bf87d11576700a1dcd7a399e7bbe39537b
+size 17209920
diff --git a/5000_hf/harness_eval_0shot/__home__aiops__zhuty__nanotron__checkpoints__5000_hf/samples_hellaswag_2026-01-07T03-53-49.355990.jsonl b/5000_hf/harness_eval_0shot/__home__aiops__zhuty__nanotron__checkpoints__5000_hf/samples_hellaswag_2026-01-07T03-53-49.355990.jsonl
new file mode 100644
index 0000000000000000000000000000000000000000..15729cc1868b607476eb4a79c1652597a92dca80
--- /dev/null
+++ b/5000_hf/harness_eval_0shot/__home__aiops__zhuty__nanotron__checkpoints__5000_hf/samples_hellaswag_2026-01-07T03-53-49.355990.jsonl
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d6120efbf55a184f73910a7f24467ff6c616fc5dff73d66b92e126bb05fa0b35
+size 42644444
diff --git a/5000_hf/harness_eval_8shot/__home__aiops__zhuty__nanotron__checkpoints__5000_hf/samples_gsm8k_2026-01-07T06-58-05.968321.jsonl b/5000_hf/harness_eval_8shot/__home__aiops__zhuty__nanotron__checkpoints__5000_hf/samples_gsm8k_2026-01-07T06-58-05.968321.jsonl
new file mode 100644
index 0000000000000000000000000000000000000000..30fa6662727e0344ebd86d7192517546223d7aff
--- /dev/null
+++ b/5000_hf/harness_eval_8shot/__home__aiops__zhuty__nanotron__checkpoints__5000_hf/samples_gsm8k_2026-01-07T06-58-05.968321.jsonl
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ef3bd4d6afdbb207d8cc365427601fc41c8399d525b33e5d6032c94d4028c5a8
+size 16800458
diff --git a/5000_hf/model.safetensors b/5000_hf/model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..f742331766fa8c1f6ce774feb128d819068f7cb4
--- /dev/null
+++ b/5000_hf/model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2ac60c304da7b3998646c03845818001165a317b1740916953bba9ed68c6f422
+size 2471645608
diff --git a/5000_hf/tokenizer.json b/5000_hf/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..1c1d8d5c9024994f1d3b00f9662b8dd89ca13cf2
--- /dev/null
+++ b/5000_hf/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6b9e4e7fb171f92fd137b777cc2714bf87d11576700a1dcd7a399e7bbe39537b
+size 17209920
diff --git a/7500_hf/harness_eval_0shot/__home__aiops__zhuty__nanotron__checkpoints__7500_hf/samples_hellaswag_2026-01-07T03-51-34.104451.jsonl b/7500_hf/harness_eval_0shot/__home__aiops__zhuty__nanotron__checkpoints__7500_hf/samples_hellaswag_2026-01-07T03-51-34.104451.jsonl
new file mode 100644
index 0000000000000000000000000000000000000000..4c7c61a97b4e3fc3320ceb53687771071404dcac
--- /dev/null
+++ b/7500_hf/harness_eval_0shot/__home__aiops__zhuty__nanotron__checkpoints__7500_hf/samples_hellaswag_2026-01-07T03-51-34.104451.jsonl
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:229906941384e79d2f01cb582969d9a0675cdb3ee7e738b9367dda4ce8bce18c
+size 42644720
diff --git a/7500_hf/harness_eval_8shot/__home__aiops__zhuty__nanotron__checkpoints__7500_hf/samples_gsm8k_2026-01-07T07-00-15.435455.jsonl b/7500_hf/harness_eval_8shot/__home__aiops__zhuty__nanotron__checkpoints__7500_hf/samples_gsm8k_2026-01-07T07-00-15.435455.jsonl
new file mode 100644
index 0000000000000000000000000000000000000000..71a717c9939436907d5a5c8a87825a32034ae626
--- /dev/null
+++ b/7500_hf/harness_eval_8shot/__home__aiops__zhuty__nanotron__checkpoints__7500_hf/samples_gsm8k_2026-01-07T07-00-15.435455.jsonl
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9b12ceb6cba2121c62b1b4db784e62fd389d9f21793c0f4d9cd6326d6e33a95d
+size 16769968
diff --git a/7500_hf/model.safetensors b/7500_hf/model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..92ebb07df1bb1cb15cc714a3e56a65bd43a9313b
--- /dev/null
+++ b/7500_hf/model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2499f14923b559337885c6aaf4537b4f18cf92239746de5191b54f8cdfa0a3e0
+size 2471645608
diff --git a/7500_hf/tokenizer.json b/7500_hf/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..1c1d8d5c9024994f1d3b00f9662b8dd89ca13cf2
--- /dev/null
+++ b/7500_hf/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6b9e4e7fb171f92fd137b777cc2714bf87d11576700a1dcd7a399e7bbe39537b
+size 17209920