Add files using upload-large-folder tool

.gitattributes

@@ -155,3 +155,13 @@ robust/exp_3_seed_stability/llama31_8b/baseline_wikitext2_ptb/r0p8/seed_41/fulls
 robust/exp_3_seed_stability/llama31_8b/baseline_wikitext2_ptb/r0p8/seed_42/fullseq_actlora/checkpoint_lowrank/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 robust/exp_3_seed_stability/llama31_8b/baseline_wikitext2_ptb/r0p8/seed_43/fullseq_actlora/checkpoint_lowrank/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 robust/exp_1_calibration_dataset/llama31_8b/c4_only/r0p8/whitening_only/checkpoint_lowrank/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+low_rank/qwen3_8b_base_c4only_w512_diverse_high_lowrank/r0p8/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+low_rank/qwen3_8b_c4only_w512_diverse_high_lowrank/r0p4/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+low_rank/qwen3_8b_c4only_w512_diverse_high_lowrank/r0p5/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+low_rank/qwen3_8b_c4only_w512_diverse_high_lowrank/r0p6/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+low_rank/qwen3_8b_c4only_w512_diverse_high_lowrank/r0p7/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+low_rank/qwen3_8b_c4only_w512_diverse_high_lowrank/r0p8/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+low_rank/qwen3_8b_base_c4only_w512_diverse_high_lowrank/r0p4/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+low_rank/qwen3_8b_base_c4only_w512_diverse_high_lowrank/r0p5/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+low_rank/qwen3_8b_base_c4only_w512_diverse_high_lowrank/r0p6/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+low_rank/qwen3_8b_base_c4only_w512_diverse_high_lowrank/r0p7/tokenizer.json filter=lfs diff=lfs merge=lfs -text

low_rank/qwen3_8b_base_c4only_w512_diverse_high_lowrank/r0p4/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e7b4c974ee1e1544d5c78d2e6299fd89636772349308a230600312afec942077
+size 8043058424

low_rank/qwen3_8b_base_c4only_w512_diverse_high_lowrank/r0p4/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:aeb13307a71acd8fe81861d94ad54ab689df773318809eed3cbe794b4492dae4
+size 11422654

low_rank/qwen3_8b_base_c4only_w512_diverse_high_lowrank/r0p5/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:10ff1b23ea2e98c0af6a53f22a1c1f3226bd7ccaf2e8e7cfbab3d92918e6a5de
+size 9435336088

low_rank/qwen3_8b_base_c4only_w512_diverse_high_lowrank/r0p5/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:aeb13307a71acd8fe81861d94ad54ab689df773318809eed3cbe794b4492dae4
+size 11422654

low_rank/qwen3_8b_base_c4only_w512_diverse_high_lowrank/r0p6/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:008d76edeb889ae2e06da627674f622b34b1bc80e7b140a1a4159ba2e3a7efc6
+size 10822901256

low_rank/qwen3_8b_base_c4only_w512_diverse_high_lowrank/r0p6/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:aeb13307a71acd8fe81861d94ad54ab689df773318809eed3cbe794b4492dae4
+size 11422654

low_rank/qwen3_8b_base_c4only_w512_diverse_high_lowrank/r0p7/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:45566f1945c7edd6df4fa28c5f6ee145b97210e089656a85316a5eb7f5b0625c
+size 12211633864

low_rank/qwen3_8b_base_c4only_w512_diverse_high_lowrank/r0p7/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:aeb13307a71acd8fe81861d94ad54ab689df773318809eed3cbe794b4492dae4
+size 11422654

low_rank/qwen3_8b_base_c4only_w512_diverse_high_lowrank/r0p8/configuration_lowrank_qwen.py

@@ -0,0 +1,19 @@
+from __future__ import annotations
+
+from transformers.models.qwen3.configuration_qwen3 import Qwen3Config
+
+
+class LowRankQwen3Config(Qwen3Config):
+    model_type = "lowrank_qwen3"
+
+    def __init__(
+        self,
+        low_rank_method: str = "svdllm",
+        low_rank_modules: dict[str, int] | None = None,
+        **kwargs,
+    ) -> None:
+        super().__init__(**kwargs)
+        self.low_rank_method = str(low_rank_method)
+        self.low_rank_modules = {
+            str(name): int(rank) for name, rank in (low_rank_modules or {}).items()
+        }

low_rank/qwen3_8b_base_c4only_w512_diverse_high_lowrank/r0p8/generation_config.json

@@ -0,0 +1,6 @@
+{
+  "bos_token_id": 151643,
+  "eos_token_id": 151643,
+  "max_new_tokens": 2048,
+  "transformers_version": "4.57.6"
+}

low_rank/qwen3_8b_base_c4only_w512_diverse_high_lowrank/r0p8/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:781129366fa4ee9d064d655beabad1235a45644ea85c03d39ecf8e04bc033f1c
+size 13600378720

low_rank/qwen3_8b_base_c4only_w512_diverse_high_lowrank/r0p8/modeling_lowrank_qwen.py

@@ -0,0 +1,73 @@
+from __future__ import annotations
+
+import torch.nn as nn
+from transformers.models.qwen3.modeling_qwen3 import Qwen3ForCausalLM, Qwen3Model
+
+from .configuration_lowrank_qwen import LowRankQwen3Config
+
+
+class SplitLowRankLinear(nn.Module):
+    def __init__(self, in_features: int, out_features: int, rank: int, *, bias: bool) -> None:
+        super().__init__()
+        self.in_features = int(in_features)
+        self.out_features = int(out_features)
+        self.rank = int(rank)
+        self.BLinear = nn.Linear(self.in_features, self.rank, bias=False)
+        self.ALinear = nn.Linear(self.rank, self.out_features, bias=bool(bias))
+
+    def forward(self, x):
+        return self.ALinear(self.BLinear(x))
+
+
+def _resolve_parent_and_name(root: nn.Module, qualified_name: str) -> tuple[nn.Module, str]:
+    parent = root
+    parts = qualified_name.split(".")
+    if parts and parts[0] == "model" and not hasattr(root, "model"):
+        parts = parts[1:]
+    for part in parts[:-1]:
+        parent = getattr(parent, part)
+    return parent, parts[-1]
+
+
+def _replace_low_rank_modules(root: nn.Module, low_rank_modules: dict[str, int]) -> None:
+    for qualified_name, rank in low_rank_modules.items():
+        parent, leaf_name = _resolve_parent_and_name(root, qualified_name)
+        original = getattr(parent, leaf_name)
+        if not isinstance(original, nn.Linear):
+            raise TypeError(
+                f"Expected `{qualified_name}` to be nn.Linear before replacement, got {type(original)}"
+            )
+        replacement = SplitLowRankLinear(
+            original.in_features,
+            original.out_features,
+            int(rank),
+            bias=original.bias is not None,
+        )
+        setattr(parent, leaf_name, replacement)
+
+
+class _LowRankQwen3Mixin:
+    config_class = LowRankQwen3Config
+
+    def _init_low_rank_metadata(self, config: LowRankQwen3Config) -> None:
+        self.low_rank_method = getattr(config, "low_rank_method", "svdllm")
+        self.low_rank_modules = {
+            str(name): int(rank) for name, rank in getattr(config, "low_rank_modules", {}).items()
+        }
+        _replace_low_rank_modules(self, self.low_rank_modules)
+
+
+class LowRankQwen3Model(_LowRankQwen3Mixin, Qwen3Model):
+    config_class = LowRankQwen3Config
+
+    def __init__(self, config: LowRankQwen3Config) -> None:
+        super().__init__(config)
+        self._init_low_rank_metadata(config)
+
+
+class LowRankQwen3ForCausalLM(_LowRankQwen3Mixin, Qwen3ForCausalLM):
+    config_class = LowRankQwen3Config
+
+    def __init__(self, config: LowRankQwen3Config) -> None:
+        super().__init__(config)
+        self._init_low_rank_metadata(config)

low_rank/qwen3_8b_base_c4only_w512_diverse_high_lowrank/r0p8/special_tokens_map.json

@@ -0,0 +1,31 @@
+{
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "eos_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

low_rank/qwen3_8b_base_c4only_w512_diverse_high_lowrank/r0p8/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:aeb13307a71acd8fe81861d94ad54ab689df773318809eed3cbe794b4492dae4
+size 11422654

low_rank/qwen3_8b_base_c4only_w512_diverse_high_lowrank/r0p8/tokenizer_config.json

@@ -0,0 +1,239 @@
+{
+  "add_bos_token": false,
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "151643": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151644": {
+      "content": "<|im_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151645": {
+      "content": "<|im_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151646": {
+      "content": "<|object_ref_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151647": {
+      "content": "<|object_ref_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151648": {
+      "content": "<|box_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151649": {
+      "content": "<|box_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151650": {
+      "content": "<|quad_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151651": {
+      "content": "<|quad_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151652": {
+      "content": "<|vision_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151653": {
+      "content": "<|vision_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151654": {
+      "content": "<|vision_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151655": {
+      "content": "<|image_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151656": {
+      "content": "<|video_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151657": {
+      "content": "<tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151658": {
+      "content": "</tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151659": {
+      "content": "<|fim_prefix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151660": {
+      "content": "<|fim_middle|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151661": {
+      "content": "<|fim_suffix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151662": {
+      "content": "<|fim_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151663": {
+      "content": "<|repo_name|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151664": {
+      "content": "<|file_sep|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151665": {
+      "content": "<tool_response>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151666": {
+      "content": "</tool_response>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151667": {
+      "content": "<think>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151668": {
+      "content": "</think>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "bos_token": null,
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|endoftext|>",
+  "errors": "replace",
+  "extra_special_tokens": {},
+  "model_max_length": 131072,
+  "pad_token": "<|endoftext|>",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

low_rank/qwen3_8b_c4only_w512_diverse_high_lowrank/r0p4/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a2d350819d3986054c859a6be7a1310c2dee8573d72a36838c7e7452b8d53b5c
+size 8043052280

low_rank/qwen3_8b_c4only_w512_diverse_high_lowrank/r0p4/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:aeb13307a71acd8fe81861d94ad54ab689df773318809eed3cbe794b4492dae4
+size 11422654

low_rank/qwen3_8b_c4only_w512_diverse_high_lowrank/r0p5/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d3bdfe55aeb4e03d80219af8a6b10504849bc860ddc50da6f8ed4aa7890c491e
+size 9435340184

low_rank/qwen3_8b_c4only_w512_diverse_high_lowrank/r0p5/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:aeb13307a71acd8fe81861d94ad54ab689df773318809eed3cbe794b4492dae4
+size 11422654

low_rank/qwen3_8b_c4only_w512_diverse_high_lowrank/r0p6/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f3eb2de29d6ab6ca90dc8601eefb6b1edc8d654f3fe83814f0e32164cac22380
+size 10822899208

low_rank/qwen3_8b_c4only_w512_diverse_high_lowrank/r0p6/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:aeb13307a71acd8fe81861d94ad54ab689df773318809eed3cbe794b4492dae4
+size 11422654

low_rank/qwen3_8b_c4only_w512_diverse_high_lowrank/r0p7/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d6d1b21e3462947a29ef17f339df189b3e44fc46e2ffa12cccc9d200046f76d4
+size 12211640008

low_rank/qwen3_8b_c4only_w512_diverse_high_lowrank/r0p7/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:aeb13307a71acd8fe81861d94ad54ab689df773318809eed3cbe794b4492dae4
+size 11422654

low_rank/qwen3_8b_c4only_w512_diverse_high_lowrank/r0p8/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6f7b4c6dfdbaff407fda9181952d859ab1afcd5657ae1fc5916a4399f71b72ae
+size 13600376672

low_rank/qwen3_8b_c4only_w512_diverse_high_lowrank/r0p8/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:aeb13307a71acd8fe81861d94ad54ab689df773318809eed3cbe794b4492dae4
+size 11422654