Upload folder using huggingface_hub

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+tokenizer.json filter=lfs diff=lfs merge=lfs -text

README.md

@@ -0,0 +1,61 @@
+---
+library_name: transformers
+license: gemma
+base_model: google/gemma-3-270m
+tags:
+- generated_from_trainer
+model-index:
+- name: gemma_noisy_zarma_finetune
+  results: []
+---
+
+<!-- This model card has been generated automatically according to the information the Trainer had access to. You
+should probably proofread and complete it, then remove this comment. -->
+
+# gemma_noisy_zarma_finetune
+
+This model is a fine-tuned version of [google/gemma-3-270m](https://huggingface.co/google/gemma-3-270m) on the None dataset.
+It achieves the following results on the evaluation set:
+- Loss: 2.0806
+
+## Model description
+
+More information needed
+
+## Intended uses & limitations
+
+More information needed
+
+## Training and evaluation data
+
+More information needed
+
+## Training procedure
+
+### Training hyperparameters
+
+The following hyperparameters were used during training:
+- learning_rate: 2e-05
+- train_batch_size: 28
+- eval_batch_size: 28
+- seed: 42
+- optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
+- lr_scheduler_type: linear
+- num_epochs: 3
+- mixed_precision_training: Native AMP
+
+### Training results
+
+| Training Loss | Epoch | Step  | Validation Loss |
+|:-------------:|:-----:|:-----:|:---------------:|
+| 1.2439        | 1.0   | 9643  | 1.6537          |
+| 0.9003        | 2.0   | 19286 | 1.8666          |
+| 0.696         | 3.0   | 28929 | 2.0806          |
+
+
+### Framework versions
+
+- Transformers 4.57.0
+- Pytorch 2.8.0+cu128
+- Datasets 4.1.1
+- Tokenizers 0.22.1

added_tokens.json

@@ -0,0 +1,3 @@
+{
+  "<image_soft_token>": 262144
+}

config.json

@@ -0,0 +1,54 @@
+{
+  "_sliding_window_pattern": 6,
+  "architectures": [
+    "Gemma3ForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "attn_logit_softcapping": null,
+  "bos_token_id": 2,
+  "dtype": "float32",
+  "eos_token_id": 1,
+  "final_logit_softcapping": null,
+  "head_dim": 256,
+  "hidden_activation": "gelu_pytorch_tanh",
+  "hidden_size": 640,
+  "initializer_range": 0.02,
+  "intermediate_size": 2048,
+  "layer_types": [
+    "sliding_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "full_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "full_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "full_attention"
+  ],
+  "max_position_embeddings": 32768,
+  "model_type": "gemma3_text",
+  "num_attention_heads": 4,
+  "num_hidden_layers": 18,
+  "num_key_value_heads": 1,
+  "pad_token_id": 0,
+  "query_pre_attn_scalar": 256,
+  "rms_norm_eps": 1e-06,
+  "rope_local_base_freq": 10000.0,
+  "rope_scaling": null,
+  "rope_theta": 1000000.0,
+  "sliding_window": 512,
+  "transformers_version": "4.57.0",
+  "use_bidirectional_attention": false,
+  "use_cache": true,
+  "vocab_size": 262144
+}

generation_config.json

@@ -0,0 +1,11 @@
+{
+  "bos_token_id": 2,
+  "do_sample": true,
+  "eos_token_id": [
+    1
+  ],
+  "pad_token_id": 0,
+  "top_k": 64,
+  "top_p": 0.95,
+  "transformers_version": "4.57.0"
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:021eaa4dfac2a735365503d2f78e6fd7cbd5da92f64eaac7a5170e70bcbc13bd
+size 1072419256

noisy_zarma_metrics.json

@@ -0,0 +1,68 @@
+{
+  "model": "google/gemma-3-270m",
+  "train_rows": 45000,
+  "eval_rows": 5000,
+  "epochs": 3,
+  "learning_rate": 2e-05,
+  "final_eval_loss": 2.080573,
+  "final_perplexity": 8.009057,
+  "noisy_columns": [
+    "char_swap",
+    "random_char_insertion",
+    "char_delete",
+    "char_substitute",
+    "word_masking",
+    "word_swap"
+  ],
+  "overall": {
+    "BLEU": 46.8891,
+    "ChrF++": 51.4311,
+    "COMET": null,
+    "n_samples": 30000,
+    "COMET_error": "COMET scoring unavailable or failed."
+  },
+  "per_column": {
+    "char_swap": {
+      "BLEU": 50.1065,
+      "ChrF++": 54.3406,
+      "COMET": null,
+      "n_samples": 5000,
+      "COMET_error": "COMET scoring unavailable or failed."
+    },
+    "random_char_insertion": {
+      "BLEU": 49.0539,
+      "ChrF++": 53.9819,
+      "COMET": null,
+      "n_samples": 5000,
+      "COMET_error": "COMET scoring unavailable or failed."
+    },
+    "char_delete": {
+      "BLEU": 46.5467,
+      "ChrF++": 52.1023,
+      "COMET": null,
+      "n_samples": 5000,
+      "COMET_error": "COMET scoring unavailable or failed."
+    },
+    "char_substitute": {
+      "BLEU": 48.4838,
+      "ChrF++": 52.7162,
+      "COMET": null,
+      "n_samples": 5000,
+      "COMET_error": "COMET scoring unavailable or failed."
+    },
+    "word_masking": {
+      "BLEU": 40.6802,
+      "ChrF++": 43.6655,
+      "COMET": null,
+      "n_samples": 5000,
+      "COMET_error": "COMET scoring unavailable or failed."
+    },
+    "word_swap": {
+      "BLEU": 44.524,
+      "ChrF++": 51.7803,
+      "COMET": null,
+      "n_samples": 5000,
+      "COMET_error": "COMET scoring unavailable or failed."
+    }
+  }
+}

special_tokens_map.json

@@ -0,0 +1,33 @@
+{
+  "boi_token": "<start_of_image>",
+  "bos_token": {
+    "content": "<bos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eoi_token": "<end_of_image>",
+  "eos_token": {
+    "content": "<eos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "image_token": "<image_soft_token>",
+  "pad_token": {
+    "content": "<pad>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:81dff017751f3628fd40ceb598dea6e37bce914e3ba98843f65dac72abe71eda
+size 33384819

tokenizer.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1299c11d7cf632ef3b4e11937501358ada021bbdf7c47638d13c0ee982f2e79c
+size 4689074

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0ea4231a12ad03603309e5bb0916d2e7e0e78cdf05a8370e1fd302f34177b107
+size 5841