Training in progress, step 500

.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

config.json

@@ -1,35 +1,68 @@
 {
-  "add_cross_attention": false,
+  "_sliding_window_pattern": 6,
   "architectures": [
-    "BertModel"
+    "Gemma3TextModel"
   ],
-  "attention_probs_dropout_prob": 0.1,
-  "bos_token_id": null,
-  "classifier_dropout": null,
-  "directionality": "bidi",
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "attn_logit_softcapping": null,
+  "bos_token_id": 2,
   "dtype": "float32",
-  "eos_token_id": null,
-  "hidden_act": "gelu",
-  "hidden_dropout_prob": 0.1,
+  "eos_token_id": 1,
+  "final_logit_softcapping": null,
+  "head_dim": 256,
+  "hidden_activation": "gelu_pytorch_tanh",
   "hidden_size": 768,
   "initializer_range": 0.02,
-  "intermediate_size": 3072,
-  "is_decoder": false,
-  "layer_norm_eps": 1e-12,
-  "max_position_embeddings": 512,
-  "model_type": "bert",
-  "num_attention_heads": 12,
-  "num_hidden_layers": 12,
+  "intermediate_size": 1152,
+  "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",
+    "sliding_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "full_attention"
+  ],
+  "max_position_embeddings": 2048,
+  "model_type": "gemma3_text",
+  "num_attention_heads": 3,
+  "num_hidden_layers": 24,
+  "num_key_value_heads": 1,
   "pad_token_id": 0,
-  "pooler_fc_size": 768,
-  "pooler_num_attention_heads": 12,
-  "pooler_num_fc_layers": 3,
-  "pooler_size_per_head": 128,
-  "pooler_type": "first_token_transform",
-  "position_embedding_type": "absolute",
+  "query_pre_attn_scalar": 256,
+  "rms_norm_eps": 1e-06,
+  "rope_parameters": {
+    "full_attention": {
+      "rope_theta": 1000000.0,
+      "rope_type": "default"
+    },
+    "sliding_attention": {
+      "rope_theta": 10000.0,
+      "rope_type": "default"
+    }
+  },
+  "sliding_window": 257,
   "tie_word_embeddings": true,
   "transformers_version": "5.0.0",
-  "type_vocab_size": 2,
+  "use_bidirectional_attention": true,
   "use_cache": true,
-  "vocab_size": 119547
+  "vocab_size": 262144
 }

eval/similarity_evaluation_results.csv

@@ -17,3 +17,8 @@ epoch,steps,cosine_pearson,cosine_spearman
 4.3478260869565215,1600,0.013782718975137054,0.0034692683513641394
 4.619565217391305,1700,0.0052551537438136845,-0.003558310926991916
 4.891304347826087,1800,0.009299551559040874,-0.0006707874030625848
+0.2717391304347826,100,nan,nan
+0.5434782608695652,200,nan,nan
+0.8152173913043478,300,0.0010648633490479042,0.003721979667392573
+1.0869565217391304,400,-0.014113401849935371,-0.01959572686603788
+1.358695652173913,500,-0.007971988965058359,-0.005923451367843734

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:23810ed9b5bd2fcdb0c3cdf8e980ab40c70973e8cbf6ea421b0dc929e42e9f26
-size 711436112
+oid sha256:b1f5fb46540a19abc34680b8e525403f0f7ee313ed606ef9956495cddced7d18
+size 1211486072

tokenizer_config.json

@@ -1,16 +1,24 @@
 {
   "backend": "tokenizers",
+  "boi_token": "<start_of_image>",
+  "bos_token": "<bos>",
   "clean_up_tokenization_spaces": false,
-  "cls_token": "[CLS]",
-  "do_lower_case": false,
+  "eoi_token": "<end_of_image>",
+  "eos_token": "<eos>",
+  "image_token": "<image_soft_token>",
   "is_local": false,
-  "mask_token": "[MASK]",
-  "model_max_length": 512,
-  "model_specific_special_tokens": {},
-  "pad_token": "[PAD]",
-  "sep_token": "[SEP]",
-  "strip_accents": null,
-  "tokenize_chinese_chars": true,
-  "tokenizer_class": "BertTokenizer",
-  "unk_token": "[UNK]"
+  "mask_token": "<mask>",
+  "model_max_length": 2048,
+  "model_specific_special_tokens": {
+    "boi_token": "<start_of_image>",
+    "eoi_token": "<end_of_image>",
+    "image_token": "<image_soft_token>"
+  },
+  "pad_token": "<pad>",
+  "padding_side": "right",
+  "sp_model_kwargs": null,
+  "spaces_between_special_tokens": false,
+  "tokenizer_class": "GemmaTokenizer",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false
 }