Update config.json

config.json

@@ -7,12 +7,12 @@
     "AutoConfig": "microsoft/Florence-2-base-ft--configuration_florence2.Florence2Config",
     "AutoModelForCausalLM": "microsoft/Florence-2-base-ft--modeling_florence2.Florence2ForConditionalGeneration"
   },
-   "bos_token_id": 0,
-  "eos_token_id": 2,
+   "bos_token_id": 2,
+  "eos_token_id": 1,
   "ignore_index": -100,
   "is_encoder_decoder": true,
   "model_type": "florence2",
-  "pad_token_id": 1,
+  "pad_token_id": 0,
   "projection_dim": 1024,
   "text_config": {
     "_name_or_path": "",
@@ -30,20 +30,20 @@
     "classif_dropout": 0.1,
     "classifier_dropout": 0.0,
     "cross_attention_hidden_size": null,
-    "d_model": 1024,
-    "decoder_attention_heads": 16,
-    "decoder_ffn_dim": 4096,
+    "d_model": 768,
+    "decoder_attention_heads": 12,
+    "decoder_ffn_dim": 3072,
     "decoder_layerdrop": 0.0,
-    "decoder_layers": 12,
+    "decoder_layers": 6,
     "decoder_start_token_id": 2,
     "diversity_penalty": 0.0,
     "do_sample": false,
     "dropout": 0.1,
     "early_stopping": true,
-    "encoder_attention_heads": 16,
-    "encoder_ffn_dim": 4096,
+    "encoder_attention_heads": 12,
+    "encoder_ffn_dim": 3072,
     "encoder_layerdrop": 0.0,
-    "encoder_layers": 12,
+    "encoder_layers": 6,
     "encoder_no_repeat_ngram_size": 0,
     "eos_token_id": 2,
     "exponential_decay_length_penalty": null,
@@ -73,7 +73,7 @@
     "normalize_before": false,
     "num_beam_groups": 1,
     "num_beams": 3,
-    "num_hidden_layers": 12,
+    "num_hidden_layers": 6,
     "num_return_sequences": 1,
     "output_attentions": false,
     "output_hidden_states": false,
@@ -105,7 +105,7 @@
     "vocab_size": 51289
   },
   "torch_dtype": "float32",
-  "transformers_version": "4.41.2",
+  "transformers_version": "4.42.3",
   "vision_config": {
     "_name_or_path": "",
     "add_cross_attention": false,
@@ -123,10 +123,10 @@
       1
     ],
     "dim_embed": [
+      128,
       256,
       512,
-      1024,
-      2048
+      1024
     ],
     "diversity_penalty": 0.0,
     "do_sample": false,
@@ -165,16 +165,16 @@
     "num_beam_groups": 1,
     "num_beams": 1,
     "num_groups": [
+      4,
       8,
       16,
-      32,
-      64
+      32
     ],
     "num_heads": [
+      4,
       8,
       16,
-      32,
-      64
+      32
     ],
     "num_return_sequences": 1,
     "output_attentions": false,
@@ -207,7 +207,7 @@
     ],
     "prefix": null,
     "problem_type": null,
-    "projection_dim": 1024,
+    "projection_dim": 768,
     "pruned_heads": {},
     "remove_invalid_values": false,
     "repetition_penalty": 1.0,