Add files using upload-large-folder tool

added_tokens.json

@@ -0,0 +1,6 @@
+{
+  "<im_end>": 32003,
+  "<im_patch>": 32001,
+  "<im_start>": 32002,
+  "[PAD]": 32000
+}

config.json

@@ -0,0 +1,37 @@
+{
+  "_name_or_path": "/netscratch/duynguyen/Research/Nghiem_LLaVA-Med/LVLM-Med/models/checkpoint_llava_med_instruct_60k_inline_mention_version_1-5_1e0_multi_graph_100_scale_dci_test_bugfix",
+  "architectures": [
+    "LlavaLlamaForCausalLM"
+  ],
+  "bos_token_id": 0,
+  "eos_token_id": 1,
+  "freeze_mm_mlp_adapter": false,
+  "graph_num_features": 4096,
+  "hidden_act": "silu",
+  "hidden_size": 4096,
+  "initializer_range": 0.02,
+  "intermediate_size": 11008,
+  "max_sequence_length": 2048,
+  "mm_hidden_size": 3072,
+  "mm_projector_type": "mlp2x_gelu",
+  "mm_use_im_start_end": true,
+  "mm_vision_select_layer": -2,
+  "mm_vision_tower": "openai/clip-vit-large-patch14",
+  "model_type": "llava",
+  "more_mlp": false,
+  "multi_graph": true,
+  "num_attention_heads": 32,
+  "num_hidden_layers": 32,
+  "pad_token_id": -1,
+  "qformer_path": "/netscratch/trnguyen/instructBLIP_checkpoint/blip2_pretrained_vitL.pth",
+  "remove_graph": false,
+  "rms_norm_eps": 1e-06,
+  "tie_word_embeddings": false,
+  "torch_dtype": "float32",
+  "transformers_version": "4.28.0.dev0",
+  "tune_mm_mlp_adapter": false,
+  "unify": true,
+  "use_cache": false,
+  "use_mm_proj": true,
+  "vocab_size": 32004
+}

generation_config.json

@@ -0,0 +1,8 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 0,
+  "eos_token_id": 1,
+  "pad_token_id": -1,
+  "transformers_version": "4.28.0.dev0",
+  "use_cache": false
+}

pytorch_model-00001-of-00003.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:51bdfb88bfd1266693bbd421888040f1b1daa1b173c07b1680cb79efc6475a47
+size 9878055609

pytorch_model-00002-of-00003.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7eb5a32e749b567c63eed72cef30595c9045a5861d0d51045241d1d00699251d
+size 9894801501

pytorch_model-00003-of-00003.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a276345aef3cce558f936e66af401702f1cefbf3f3173f9c68aa8fce6d8f02b7
+size 7298531288

pytorch_model.bin.index.json

@@ -0,0 +1,334 @@
+{
+  "metadata": {
+    "total_size": 27071275008
+  },
+  "weight_map": {
+    "lm_head.weight": "pytorch_model-00003-of-00003.bin",
+    "model.embed_tokens.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.0.input_layernorm.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.0.mlp.down_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.0.mlp.gate_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.0.mlp.up_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.0.post_attention_layernorm.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.0.self_attn.k_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.0.self_attn.o_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.0.self_attn.q_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.0.self_attn.rotary_emb.inv_freq": "pytorch_model-00001-of-00003.bin",
+    "model.layers.0.self_attn.v_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.1.input_layernorm.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.1.mlp.down_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.1.mlp.gate_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.1.mlp.up_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.1.post_attention_layernorm.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.1.self_attn.k_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.1.self_attn.o_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.1.self_attn.q_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.1.self_attn.rotary_emb.inv_freq": "pytorch_model-00001-of-00003.bin",
+    "model.layers.1.self_attn.v_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.10.input_layernorm.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.10.mlp.down_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.10.mlp.gate_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.10.mlp.up_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.10.post_attention_layernorm.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.10.self_attn.k_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.10.self_attn.o_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.10.self_attn.q_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.10.self_attn.rotary_emb.inv_freq": "pytorch_model-00001-of-00003.bin",
+    "model.layers.10.self_attn.v_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.11.input_layernorm.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.11.mlp.down_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.11.mlp.gate_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.11.mlp.up_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.11.post_attention_layernorm.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.11.self_attn.k_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.11.self_attn.o_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.11.self_attn.q_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.11.self_attn.rotary_emb.inv_freq": "pytorch_model-00001-of-00003.bin",
+    "model.layers.11.self_attn.v_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.12.input_layernorm.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.12.mlp.down_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.12.mlp.gate_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.12.mlp.up_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.12.post_attention_layernorm.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.12.self_attn.k_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.12.self_attn.o_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.12.self_attn.q_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.12.self_attn.rotary_emb.inv_freq": "pytorch_model-00002-of-00003.bin",
+    "model.layers.12.self_attn.v_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.13.input_layernorm.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.13.mlp.down_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.13.mlp.gate_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.13.mlp.up_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.13.post_attention_layernorm.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.13.self_attn.k_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.13.self_attn.o_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.13.self_attn.q_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.13.self_attn.rotary_emb.inv_freq": "pytorch_model-00002-of-00003.bin",
+    "model.layers.13.self_attn.v_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.14.input_layernorm.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.14.mlp.down_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.14.mlp.gate_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.14.mlp.up_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.14.post_attention_layernorm.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.14.self_attn.k_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.14.self_attn.o_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.14.self_attn.q_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.14.self_attn.rotary_emb.inv_freq": "pytorch_model-00002-of-00003.bin",
+    "model.layers.14.self_attn.v_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.15.input_layernorm.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.15.mlp.down_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.15.mlp.gate_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.15.mlp.up_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.15.post_attention_layernorm.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.15.self_attn.k_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.15.self_attn.o_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.15.self_attn.q_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.15.self_attn.rotary_emb.inv_freq": "pytorch_model-00002-of-00003.bin",
+    "model.layers.15.self_attn.v_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.16.input_layernorm.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.16.mlp.down_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.16.mlp.gate_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.16.mlp.up_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.16.post_attention_layernorm.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.16.self_attn.k_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.16.self_attn.o_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.16.self_attn.q_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.16.self_attn.rotary_emb.inv_freq": "pytorch_model-00002-of-00003.bin",
+    "model.layers.16.self_attn.v_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.17.input_layernorm.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.17.mlp.down_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.17.mlp.gate_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.17.mlp.up_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.17.post_attention_layernorm.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.17.self_attn.k_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.17.self_attn.o_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.17.self_attn.q_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.17.self_attn.rotary_emb.inv_freq": "pytorch_model-00002-of-00003.bin",
+    "model.layers.17.self_attn.v_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.18.input_layernorm.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.18.mlp.down_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.18.mlp.gate_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.18.mlp.up_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.18.post_attention_layernorm.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.18.self_attn.k_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.18.self_attn.o_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.18.self_attn.q_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.18.self_attn.rotary_emb.inv_freq": "pytorch_model-00002-of-00003.bin",
+    "model.layers.18.self_attn.v_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.19.input_layernorm.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.19.mlp.down_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.19.mlp.gate_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.19.mlp.up_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.19.post_attention_layernorm.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.19.self_attn.k_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.19.self_attn.o_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.19.self_attn.q_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.19.self_attn.rotary_emb.inv_freq": "pytorch_model-00002-of-00003.bin",
+    "model.layers.19.self_attn.v_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.2.input_layernorm.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.2.mlp.down_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.2.mlp.gate_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.2.mlp.up_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.2.post_attention_layernorm.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.2.self_attn.k_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.2.self_attn.o_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.2.self_attn.q_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.2.self_attn.rotary_emb.inv_freq": "pytorch_model-00001-of-00003.bin",
+    "model.layers.2.self_attn.v_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.20.input_layernorm.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.20.mlp.down_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.20.mlp.gate_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.20.mlp.up_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.20.post_attention_layernorm.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.20.self_attn.k_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.20.self_attn.o_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.20.self_attn.q_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.20.self_attn.rotary_emb.inv_freq": "pytorch_model-00002-of-00003.bin",
+    "model.layers.20.self_attn.v_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.21.input_layernorm.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.21.mlp.down_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.21.mlp.gate_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.21.mlp.up_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.21.post_attention_layernorm.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.21.self_attn.k_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.21.self_attn.o_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.21.self_attn.q_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.21.self_attn.rotary_emb.inv_freq": "pytorch_model-00002-of-00003.bin",
+    "model.layers.21.self_attn.v_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.22.input_layernorm.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.22.mlp.down_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.22.mlp.gate_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.22.mlp.up_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.22.post_attention_layernorm.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.22.self_attn.k_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.22.self_attn.o_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.22.self_attn.q_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.22.self_attn.rotary_emb.inv_freq": "pytorch_model-00002-of-00003.bin",
+    "model.layers.22.self_attn.v_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.23.input_layernorm.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.23.mlp.down_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.23.mlp.gate_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.23.mlp.up_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.23.post_attention_layernorm.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.23.self_attn.k_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.23.self_attn.o_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.23.self_attn.q_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.23.self_attn.rotary_emb.inv_freq": "pytorch_model-00002-of-00003.bin",
+    "model.layers.23.self_attn.v_proj.weight": "pytorch_model-00002-of-00003.bin",
+    "model.layers.24.input_layernorm.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.24.mlp.down_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.24.mlp.gate_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.24.mlp.up_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.24.post_attention_layernorm.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.24.self_attn.k_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.24.self_attn.o_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.24.self_attn.q_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.24.self_attn.rotary_emb.inv_freq": "pytorch_model-00003-of-00003.bin",
+    "model.layers.24.self_attn.v_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.25.input_layernorm.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.25.mlp.down_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.25.mlp.gate_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.25.mlp.up_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.25.post_attention_layernorm.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.25.self_attn.k_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.25.self_attn.o_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.25.self_attn.q_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.25.self_attn.rotary_emb.inv_freq": "pytorch_model-00003-of-00003.bin",
+    "model.layers.25.self_attn.v_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.26.input_layernorm.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.26.mlp.down_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.26.mlp.gate_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.26.mlp.up_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.26.post_attention_layernorm.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.26.self_attn.k_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.26.self_attn.o_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.26.self_attn.q_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.26.self_attn.rotary_emb.inv_freq": "pytorch_model-00003-of-00003.bin",
+    "model.layers.26.self_attn.v_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.27.input_layernorm.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.27.mlp.down_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.27.mlp.gate_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.27.mlp.up_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.27.post_attention_layernorm.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.27.self_attn.k_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.27.self_attn.o_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.27.self_attn.q_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.27.self_attn.rotary_emb.inv_freq": "pytorch_model-00003-of-00003.bin",
+    "model.layers.27.self_attn.v_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.28.input_layernorm.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.28.mlp.down_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.28.mlp.gate_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.28.mlp.up_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.28.post_attention_layernorm.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.28.self_attn.k_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.28.self_attn.o_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.28.self_attn.q_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.28.self_attn.rotary_emb.inv_freq": "pytorch_model-00003-of-00003.bin",
+    "model.layers.28.self_attn.v_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.29.input_layernorm.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.29.mlp.down_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.29.mlp.gate_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.29.mlp.up_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.29.post_attention_layernorm.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.29.self_attn.k_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.29.self_attn.o_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.29.self_attn.q_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.29.self_attn.rotary_emb.inv_freq": "pytorch_model-00003-of-00003.bin",
+    "model.layers.29.self_attn.v_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.3.input_layernorm.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.3.mlp.down_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.3.mlp.gate_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.3.mlp.up_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.3.post_attention_layernorm.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.3.self_attn.k_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.3.self_attn.o_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.3.self_attn.q_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.3.self_attn.rotary_emb.inv_freq": "pytorch_model-00001-of-00003.bin",
+    "model.layers.3.self_attn.v_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.30.input_layernorm.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.30.mlp.down_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.30.mlp.gate_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.30.mlp.up_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.30.post_attention_layernorm.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.30.self_attn.k_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.30.self_attn.o_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.30.self_attn.q_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.30.self_attn.rotary_emb.inv_freq": "pytorch_model-00003-of-00003.bin",
+    "model.layers.30.self_attn.v_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.31.input_layernorm.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.31.mlp.down_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.31.mlp.gate_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.31.mlp.up_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.31.post_attention_layernorm.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.31.self_attn.k_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.31.self_attn.o_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.31.self_attn.q_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.31.self_attn.rotary_emb.inv_freq": "pytorch_model-00003-of-00003.bin",
+    "model.layers.31.self_attn.v_proj.weight": "pytorch_model-00003-of-00003.bin",
+    "model.layers.4.input_layernorm.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.4.mlp.down_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.4.mlp.gate_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.4.mlp.up_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.4.post_attention_layernorm.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.4.self_attn.k_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.4.self_attn.o_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.4.self_attn.q_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.4.self_attn.rotary_emb.inv_freq": "pytorch_model-00001-of-00003.bin",
+    "model.layers.4.self_attn.v_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.5.input_layernorm.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.5.mlp.down_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.5.mlp.gate_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.5.mlp.up_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.5.post_attention_layernorm.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.5.self_attn.k_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.5.self_attn.o_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.5.self_attn.q_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.5.self_attn.rotary_emb.inv_freq": "pytorch_model-00001-of-00003.bin",
+    "model.layers.5.self_attn.v_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.6.input_layernorm.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.6.mlp.down_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.6.mlp.gate_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.6.mlp.up_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.6.post_attention_layernorm.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.6.self_attn.k_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.6.self_attn.o_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.6.self_attn.q_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.6.self_attn.rotary_emb.inv_freq": "pytorch_model-00001-of-00003.bin",
+    "model.layers.6.self_attn.v_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.7.input_layernorm.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.7.mlp.down_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.7.mlp.gate_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.7.mlp.up_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.7.post_attention_layernorm.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.7.self_attn.k_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.7.self_attn.o_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.7.self_attn.q_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.7.self_attn.rotary_emb.inv_freq": "pytorch_model-00001-of-00003.bin",
+    "model.layers.7.self_attn.v_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.8.input_layernorm.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.8.mlp.down_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.8.mlp.gate_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.8.mlp.up_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.8.post_attention_layernorm.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.8.self_attn.k_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.8.self_attn.o_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.8.self_attn.q_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.8.self_attn.rotary_emb.inv_freq": "pytorch_model-00001-of-00003.bin",
+    "model.layers.8.self_attn.v_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.9.input_layernorm.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.9.mlp.down_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.9.mlp.gate_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.9.mlp.up_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.9.post_attention_layernorm.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.9.self_attn.k_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.9.self_attn.o_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.9.self_attn.q_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.layers.9.self_attn.rotary_emb.inv_freq": "pytorch_model-00001-of-00003.bin",
+    "model.layers.9.self_attn.v_proj.weight": "pytorch_model-00001-of-00003.bin",
+    "model.mm_projector.0.bias": "pytorch_model-00003-of-00003.bin",
+    "model.mm_projector.0.weight": "pytorch_model-00003-of-00003.bin",
+    "model.mm_projector.2.bias": "pytorch_model-00003-of-00003.bin",
+    "model.mm_projector.2.weight": "pytorch_model-00003-of-00003.bin",
+    "model.norm.weight": "pytorch_model-00003-of-00003.bin"
+  }
+}

special_tokens_map.json

@@ -0,0 +1,6 @@
+{
+  "bos_token": "</s>",
+  "eos_token": "</s>",
+  "pad_token": "[PAD]",
+  "unk_token": "<unk>"
+}

tokenizer.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9e556afd44213b6bd1be2b850ebbbd98f5481437a8021afaf58ee7fb1818d347
+size 499723

tokenizer_config.json

@@ -0,0 +1,32 @@
+{
+  "bos_token": {
+    "__type": "AddedToken",
+    "content": "<s>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "clean_up_tokenization_spaces": false,
+  "eos_token": {
+    "__type": "AddedToken",
+    "content": "</s>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "model_max_length": 2048,
+  "pad_token": null,
+  "padding_side": "right",
+  "special_tokens_map_file": "/nobackup/haotian/.cache/huggingface/hub/models--lmsys--vicuna-7b-delta-v0/snapshots/6d2e645f62e734821be8c392c08c8b2098b35434/special_tokens_map.json",
+  "tokenizer_class": "LlamaTokenizer",
+  "unk_token": {
+    "__type": "AddedToken",
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  }
+}

trainer_state.json

@@ -0,0 +1,4369 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 18.0,
+  "global_step": 702,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.03,
+      "learning_rate": 1.8181818181818183e-06,
+      "loss": 2.5841,
+      "step": 1
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 3.6363636363636366e-06,
+      "loss": 2.5898,
+      "step": 2
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 5.4545454545454545e-06,
+      "loss": 1.6886,
+      "step": 3
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 7.272727272727273e-06,
+      "loss": 0.999,
+      "step": 4
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 9.090909090909091e-06,
+      "loss": 0.5869,
+      "step": 5
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 1.0909090909090909e-05,
+      "loss": 0.3544,
+      "step": 6
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 1.2727272727272728e-05,
+      "loss": 0.3183,
+      "step": 7
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 1.4545454545454546e-05,
+      "loss": 0.2377,
+      "step": 8
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 1.6363636363636366e-05,
+      "loss": 0.289,
+      "step": 9
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 1.8181818181818182e-05,
+      "loss": 0.1791,
+      "step": 10
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 2e-05,
+      "loss": 0.152,
+      "step": 11
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 2.1818181818181818e-05,
+      "loss": 0.1551,
+      "step": 12
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 2.363636363636364e-05,
+      "loss": 0.1501,
+      "step": 13
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 2.5454545454545457e-05,
+      "loss": 0.1434,
+      "step": 14
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 2.7272727272727273e-05,
+      "loss": 0.1221,
+      "step": 15
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 2.9090909090909093e-05,
+      "loss": 0.1988,
+      "step": 16
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 3.090909090909091e-05,
+      "loss": 0.154,
+      "step": 17
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 3.272727272727273e-05,
+      "loss": 0.1034,
+      "step": 18
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 3.454545454545455e-05,
+      "loss": 0.1239,
+      "step": 19
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 3.6363636363636364e-05,
+      "loss": 0.1626,
+      "step": 20
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 3.818181818181819e-05,
+      "loss": 0.1231,
+      "step": 21
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 4e-05,
+      "loss": 0.1245,
+      "step": 22
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 3.99997865573779e-05,
+      "loss": 0.1269,
+      "step": 23
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 3.999914623406736e-05,
+      "loss": 0.1271,
+      "step": 24
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 3.999807904373562e-05,
+      "loss": 0.1462,
+      "step": 25
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 3.9996585009161056e-05,
+      "loss": 0.1237,
+      "step": 26
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 3.999466416223275e-05,
+      "loss": 0.1354,
+      "step": 27
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 3.999231654394975e-05,
+      "loss": 0.0988,
+      "step": 28
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 3.9989542204420243e-05,
+      "loss": 0.1454,
+      "step": 29
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 3.9986341202860467e-05,
+      "loss": 0.0983,
+      "step": 30
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 3.998271360759342e-05,
+      "loss": 0.1008,
+      "step": 31
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 3.9978659496047456e-05,
+      "loss": 0.0895,
+      "step": 32
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 3.997417895475459e-05,
+      "loss": 0.1208,
+      "step": 33
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 3.9969272079348685e-05,
+      "loss": 0.087,
+      "step": 34
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 3.9963938974563355e-05,
+      "loss": 0.1024,
+      "step": 35
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 3.995817975422981e-05,
+      "loss": 0.0935,
+      "step": 36
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 3.9951994541274345e-05,
+      "loss": 0.0888,
+      "step": 37
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 3.994538346771576e-05,
+      "loss": 0.1204,
+      "step": 38
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.9938346674662565e-05,
+      "loss": 0.069,
+      "step": 39
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 3.9930884312309894e-05,
+      "loss": 0.1135,
+      "step": 40
+    },
+    {
+      "epoch": 1.05,
+      "learning_rate": 3.9922996539936374e-05,
+      "loss": 0.0879,
+      "step": 41
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 3.991468352590069e-05,
+      "loss": 0.0729,
+      "step": 42
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 3.9905945447638e-05,
+      "loss": 0.0738,
+      "step": 43
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 3.989678249165612e-05,
+      "loss": 0.0896,
+      "step": 44
+    },
+    {
+      "epoch": 1.15,
+      "learning_rate": 3.9887194853531584e-05,
+      "loss": 0.0783,
+      "step": 45
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 3.987718273790548e-05,
+      "loss": 0.0761,
+      "step": 46
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 3.9866746358479e-05,
+      "loss": 0.0994,
+      "step": 47
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 3.9855885938008986e-05,
+      "loss": 0.0954,
+      "step": 48
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 3.984460170830308e-05,
+      "loss": 0.0829,
+      "step": 49
+    },
+    {
+      "epoch": 1.28,
+      "learning_rate": 3.983289391021486e-05,
+      "loss": 0.0796,
+      "step": 50
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 3.9820762793638626e-05,
+      "loss": 0.0887,
+      "step": 51
+    },
+    {
+      "epoch": 1.33,
+      "learning_rate": 3.9808208617504106e-05,
+      "loss": 0.075,
+      "step": 52
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 3.979523164977094e-05,
+      "loss": 0.0528,
+      "step": 53
+    },
+    {
+      "epoch": 1.38,
+      "learning_rate": 3.9781832167422926e-05,
+      "loss": 0.0804,
+      "step": 54
+    },
+    {
+      "epoch": 1.41,
+      "learning_rate": 3.976801045646212e-05,
+      "loss": 0.1193,
+      "step": 55
+    },
+    {
+      "epoch": 1.44,
+      "learning_rate": 3.9753766811902756e-05,
+      "loss": 0.0737,
+      "step": 56
+    },
+    {
+      "epoch": 1.46,
+      "learning_rate": 3.973910153776492e-05,
+      "loss": 0.0952,
+      "step": 57
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 3.972401494706805e-05,
+      "loss": 0.0574,
+      "step": 58
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 3.970850736182432e-05,
+      "loss": 0.0857,
+      "step": 59
+    },
+    {
+      "epoch": 1.54,
+      "learning_rate": 3.969257911303167e-05,
+      "loss": 0.0682,
+      "step": 60
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 3.967623054066684e-05,
+      "loss": 0.0765,
+      "step": 61
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 3.965946199367804e-05,
+      "loss": 0.0616,
+      "step": 62
+    },
+    {
+      "epoch": 1.62,
+      "learning_rate": 3.964227382997752e-05,
+      "loss": 0.0806,
+      "step": 63
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 3.962466641643398e-05,
+      "loss": 0.0929,
+      "step": 64
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 3.9606640128864635e-05,
+      "loss": 0.0548,
+      "step": 65
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 3.958819535202732e-05,
+      "loss": 0.0671,
+      "step": 66
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 3.956933247961218e-05,
+      "loss": 0.0853,
+      "step": 67
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 3.9550051914233314e-05,
+      "loss": 0.0558,
+      "step": 68
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 3.953035406742016e-05,
+      "loss": 0.0512,
+      "step": 69
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 3.951023935960874e-05,
+      "loss": 0.0845,
+      "step": 70
+    },
+    {
+      "epoch": 1.82,
+      "learning_rate": 3.9489708220132626e-05,
+      "loss": 0.0885,
+      "step": 71
+    },
+    {
+      "epoch": 1.85,
+      "learning_rate": 3.9468761087213864e-05,
+      "loss": 0.0814,
+      "step": 72
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 3.9447398407953536e-05,
+      "loss": 0.0918,
+      "step": 73
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 3.942562063832228e-05,
+      "loss": 0.0769,
+      "step": 74
+    },
+    {
+      "epoch": 1.92,
+      "learning_rate": 3.940342824315052e-05,
+      "loss": 0.0647,
+      "step": 75
+    },
+    {
+      "epoch": 1.95,
+      "learning_rate": 3.9380821696118556e-05,
+      "loss": 0.0956,
+      "step": 76
+    },
+    {
+      "epoch": 1.97,
+      "learning_rate": 3.935780147974646e-05,
+      "loss": 0.4721,
+      "step": 77
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.933436808538375e-05,
+      "loss": 0.0697,
+      "step": 78
+    },
+    {
+      "epoch": 2.03,
+      "learning_rate": 3.9310522013198965e-05,
+      "loss": 0.0723,
+      "step": 79
+    },
+    {
+      "epoch": 2.05,
+      "learning_rate": 3.92862637721689e-05,
+      "loss": 0.0685,
+      "step": 80
+    },
+    {
+      "epoch": 2.08,
+      "learning_rate": 3.9261593880067826e-05,
+      "loss": 0.0635,
+      "step": 81
+    },
+    {
+      "epoch": 2.1,
+      "learning_rate": 3.923651286345638e-05,
+      "loss": 0.0457,
+      "step": 82
+    },
+    {
+      "epoch": 2.13,
+      "learning_rate": 3.921102125767037e-05,
+      "loss": 0.0721,
+      "step": 83
+    },
+    {
+      "epoch": 2.15,
+      "learning_rate": 3.9185119606809305e-05,
+      "loss": 0.0543,
+      "step": 84
+    },
+    {
+      "epoch": 2.18,
+      "learning_rate": 3.9158808463724806e-05,
+      "loss": 0.0575,
+      "step": 85
+    },
+    {
+      "epoch": 2.21,
+      "learning_rate": 3.913208839000882e-05,
+      "loss": 0.0598,
+      "step": 86
+    },
+    {
+      "epoch": 2.23,
+      "learning_rate": 3.9104959955981605e-05,
+      "loss": 0.0556,
+      "step": 87
+    },
+    {
+      "epoch": 2.26,
+      "learning_rate": 3.907742374067956e-05,
+      "loss": 0.0557,
+      "step": 88
+    },
+    {
+      "epoch": 2.28,
+      "learning_rate": 3.904948033184291e-05,
+      "loss": 0.0618,
+      "step": 89
+    },
+    {
+      "epoch": 2.31,
+      "learning_rate": 3.9021130325903076e-05,
+      "loss": 0.0549,
+      "step": 90
+    },
+    {
+      "epoch": 2.33,
+      "learning_rate": 3.8992374327970024e-05,
+      "loss": 0.0481,
+      "step": 91
+    },
+    {
+      "epoch": 2.36,
+      "learning_rate": 3.896321295181932e-05,
+      "loss": 0.0556,
+      "step": 92
+    },
+    {
+      "epoch": 2.38,
+      "learning_rate": 3.893364681987902e-05,
+      "loss": 0.0447,
+      "step": 93
+    },
+    {
+      "epoch": 2.41,
+      "learning_rate": 3.89036765632164e-05,
+      "loss": 0.0563,
+      "step": 94
+    },
+    {
+      "epoch": 2.44,
+      "learning_rate": 3.887330282152446e-05,
+      "loss": 0.0553,
+      "step": 95
+    },
+    {
+      "epoch": 2.46,
+      "learning_rate": 3.8842526243108326e-05,
+      "loss": 0.0533,
+      "step": 96
+    },
+    {
+      "epoch": 2.49,
+      "learning_rate": 3.8811347484871353e-05,
+      "loss": 0.0411,
+      "step": 97
+    },
+    {
+      "epoch": 2.51,
+      "learning_rate": 3.877976721230114e-05,
+      "loss": 0.0698,
+      "step": 98
+    },
+    {
+      "epoch": 2.54,
+      "learning_rate": 3.8747786099455285e-05,
+      "loss": 0.0684,
+      "step": 99
+    },
+    {
+      "epoch": 2.56,
+      "learning_rate": 3.8715404828947055e-05,
+      "loss": 0.0486,
+      "step": 100
+    },
+    {
+      "epoch": 2.59,
+      "learning_rate": 3.868262409193078e-05,
+      "loss": 0.0632,
+      "step": 101
+    },
+    {
+      "epoch": 2.62,
+      "learning_rate": 3.864944458808712e-05,
+      "loss": 0.0753,
+      "step": 102
+    },
+    {
+      "epoch": 2.64,
+      "learning_rate": 3.861586702560808e-05,
+      "loss": 0.052,
+      "step": 103
+    },
+    {
+      "epoch": 2.67,
+      "learning_rate": 3.8581892121181984e-05,
+      "loss": 0.0611,
+      "step": 104
+    },
+    {
+      "epoch": 2.69,
+      "learning_rate": 3.854752059997807e-05,
+      "loss": 0.0531,
+      "step": 105
+    },
+    {
+      "epoch": 2.72,
+      "learning_rate": 3.851275319563113e-05,
+      "loss": 0.079,
+      "step": 106
+    },
+    {
+      "epoch": 2.74,
+      "learning_rate": 3.8477590650225735e-05,
+      "loss": 0.066,
+      "step": 107
+    },
+    {
+      "epoch": 2.77,
+      "learning_rate": 3.844203371428049e-05,
+      "loss": 0.0497,
+      "step": 108
+    },
+    {
+      "epoch": 2.79,
+      "learning_rate": 3.8406083146731956e-05,
+      "loss": 0.0701,
+      "step": 109
+    },
+    {
+      "epoch": 2.82,
+      "learning_rate": 3.836973971491847e-05,
+      "loss": 0.0663,
+      "step": 110
+    },
+    {
+      "epoch": 2.85,
+      "learning_rate": 3.8333004194563764e-05,
+      "loss": 0.0569,
+      "step": 111
+    },
+    {
+      "epoch": 2.87,
+      "learning_rate": 3.8295877369760426e-05,
+      "loss": 0.054,
+      "step": 112
+    },
+    {
+      "epoch": 2.9,
+      "learning_rate": 3.8258360032953136e-05,
+      "loss": 0.0642,
+      "step": 113
+    },
+    {
+      "epoch": 2.92,
+      "learning_rate": 3.822045298492177e-05,
+      "loss": 0.0547,
+      "step": 114
+    },
+    {
+      "epoch": 2.95,
+      "learning_rate": 3.81821570347643e-05,
+      "loss": 0.0511,
+      "step": 115
+    },
+    {
+      "epoch": 2.97,
+      "learning_rate": 3.814347299987953e-05,
+      "loss": 0.0566,
+      "step": 116
+    },
+    {
+      "epoch": 2.97,
+      "eval_test_accuracy": 74.75961538461539,
+      "eval_test_average": 78.50293262938611,
+      "eval_test_loss": 1.1910189390182495,
+      "eval_test_recall": 82.24624987415683,
+      "eval_test_runtime": 774.1404,
+      "eval_test_samples_per_second": 1.371,
+      "eval_test_steps_per_second": 0.006,
+      "step": 116
+    },
+    {
+      "epoch": 2.97,
+      "eval_val_accuracy": 74.40758293838863,
+      "eval_val_average": 78.0483275019237,
+      "eval_val_loss": 1.202035903930664,
+      "eval_val_recall": 81.68907206545876,
+      "eval_val_runtime": 781.9322,
+      "eval_val_samples_per_second": 1.347,
+      "eval_val_steps_per_second": 0.006,
+      "step": 116
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 3.810440170594964e-05,
+      "loss": 0.0973,
+      "step": 117
+    },
+    {
+      "epoch": 3.03,
+      "learning_rate": 3.806494398692258e-05,
+      "loss": 0.0425,
+      "step": 118
+    },
+    {
+      "epoch": 3.05,
+      "learning_rate": 3.802510068499424e-05,
+      "loss": 0.0441,
+      "step": 119
+    },
+    {
+      "epoch": 3.08,
+      "learning_rate": 3.7984872650590516e-05,
+      "loss": 0.0477,
+      "step": 120
+    },
+    {
+      "epoch": 3.1,
+      "learning_rate": 3.7944260742349113e-05,
+      "loss": 0.0435,
+      "step": 121
+    },
+    {
+      "epoch": 3.13,
+      "learning_rate": 3.790326582710125e-05,
+      "loss": 0.0417,
+      "step": 122
+    },
+    {
+      "epoch": 3.15,
+      "learning_rate": 3.786188877985315e-05,
+      "loss": 0.0507,
+      "step": 123
+    },
+    {
+      "epoch": 3.18,
+      "learning_rate": 3.782013048376736e-05,
+      "loss": 0.0381,
+      "step": 124
+    },
+    {
+      "epoch": 3.21,
+      "learning_rate": 3.77779918301439e-05,
+      "loss": 0.0349,
+      "step": 125
+    },
+    {
+      "epoch": 3.23,
+      "learning_rate": 3.773547371840124e-05,
+      "loss": 0.0492,
+      "step": 126
+    },
+    {
+      "epoch": 3.26,
+      "learning_rate": 3.769257705605711e-05,
+      "loss": 0.0391,
+      "step": 127
+    },
+    {
+      "epoch": 3.28,
+      "learning_rate": 3.764930275870912e-05,
+      "loss": 0.0228,
+      "step": 128
+    },
+    {
+      "epoch": 3.31,
+      "learning_rate": 3.760565175001521e-05,
+      "loss": 0.0274,
+      "step": 129
+    },
+    {
+      "epoch": 3.33,
+      "learning_rate": 3.756162496167396e-05,
+      "loss": 0.0364,
+      "step": 130
+    },
+    {
+      "epoch": 3.36,
+      "learning_rate": 3.7517223333404694e-05,
+      "loss": 0.0431,
+      "step": 131
+    },
+    {
+      "epoch": 3.38,
+      "learning_rate": 3.7472447812927395e-05,
+      "loss": 0.0524,
+      "step": 132
+    },
+    {
+      "epoch": 3.41,
+      "learning_rate": 3.742729935594252e-05,
+      "loss": 0.047,
+      "step": 133
+    },
+    {
+      "epoch": 3.44,
+      "learning_rate": 3.738177892611057e-05,
+      "loss": 0.0465,
+      "step": 134
+    },
+    {
+      "epoch": 3.46,
+      "learning_rate": 3.733588749503154e-05,
+      "loss": 0.0908,
+      "step": 135
+    },
+    {
+      "epoch": 3.49,
+      "learning_rate": 3.728962604222416e-05,
+      "loss": 0.05,
+      "step": 136
+    },
+    {
+      "epoch": 3.51,
+      "learning_rate": 3.7242995555105016e-05,
+      "loss": 0.0418,
+      "step": 137
+    },
+    {
+      "epoch": 3.54,
+      "learning_rate": 3.719599702896745e-05,
+      "loss": 0.0466,
+      "step": 138
+    },
+    {
+      "epoch": 3.56,
+      "learning_rate": 3.714863146696033e-05,
+      "loss": 0.0418,
+      "step": 139
+    },
+    {
+      "epoch": 3.59,
+      "learning_rate": 3.710089988006662e-05,
+      "loss": 0.0331,
+      "step": 140
+    },
+    {
+      "epoch": 3.62,
+      "learning_rate": 3.705280328708185e-05,
+      "loss": 0.0483,
+      "step": 141
+    },
+    {
+      "epoch": 3.64,
+      "learning_rate": 3.700434271459229e-05,
+      "loss": 0.0525,
+      "step": 142
+    },
+    {
+      "epoch": 3.67,
+      "learning_rate": 3.695551919695311e-05,
+      "loss": 0.0498,
+      "step": 143
+    },
+    {
+      "epoch": 3.69,
+      "learning_rate": 3.690633377626628e-05,
+      "loss": 0.0835,
+      "step": 144
+    },
+    {
+      "epoch": 3.72,
+      "learning_rate": 3.685678750235831e-05,
+      "loss": 0.0489,
+      "step": 145
+    },
+    {
+      "epoch": 3.74,
+      "learning_rate": 3.680688143275786e-05,
+      "loss": 0.0289,
+      "step": 146
+    },
+    {
+      "epoch": 3.77,
+      "learning_rate": 3.675661663267317e-05,
+      "loss": 0.0465,
+      "step": 147
+    },
+    {
+      "epoch": 3.79,
+      "learning_rate": 3.670599417496931e-05,
+      "loss": 0.0521,
+      "step": 148
+    },
+    {
+      "epoch": 3.82,
+      "learning_rate": 3.6655015140145296e-05,
+      "loss": 0.046,
+      "step": 149
+    },
+    {
+      "epoch": 3.85,
+      "learning_rate": 3.6603680616311013e-05,
+      "loss": 0.0359,
+      "step": 150
+    },
+    {
+      "epoch": 3.87,
+      "learning_rate": 3.6551991699164e-05,
+      "loss": 0.036,
+      "step": 151
+    },
+    {
+      "epoch": 3.9,
+      "learning_rate": 3.6499949491966046e-05,
+      "loss": 0.0476,
+      "step": 152
+    },
+    {
+      "epoch": 3.92,
+      "learning_rate": 3.644755510551968e-05,
+      "loss": 0.0399,
+      "step": 153
+    },
+    {
+      "epoch": 3.95,
+      "learning_rate": 3.639480965814443e-05,
+      "loss": 0.0475,
+      "step": 154
+    },
+    {
+      "epoch": 3.97,
+      "learning_rate": 3.634171427565293e-05,
+      "loss": 0.061,
+      "step": 155
+    },
+    {
+      "epoch": 4.0,
+      "learning_rate": 3.628827009132697e-05,
+      "loss": 0.0382,
+      "step": 156
+    },
+    {
+      "epoch": 4.03,
+      "learning_rate": 3.623447824589323e-05,
+      "loss": 0.0386,
+      "step": 157
+    },
+    {
+      "epoch": 4.05,
+      "learning_rate": 3.6180339887498953e-05,
+      "loss": 0.0314,
+      "step": 158
+    },
+    {
+      "epoch": 4.08,
+      "learning_rate": 3.6125856171687465e-05,
+      "loss": 0.0253,
+      "step": 159
+    },
+    {
+      "epoch": 4.1,
+      "learning_rate": 3.6071028261373474e-05,
+      "loss": 0.0275,
+      "step": 160
+    },
+    {
+      "epoch": 4.13,
+      "learning_rate": 3.6015857326818295e-05,
+      "loss": 0.0379,
+      "step": 161
+    },
+    {
+      "epoch": 4.15,
+      "learning_rate": 3.5960344545604796e-05,
+      "loss": 0.0684,
+      "step": 162
+    },
+    {
+      "epoch": 4.18,
+      "learning_rate": 3.5904491102612346e-05,
+      "loss": 0.0297,
+      "step": 163
+    },
+    {
+      "epoch": 4.21,
+      "learning_rate": 3.584829818999148e-05,
+      "loss": 0.0251,
+      "step": 164
+    },
+    {
+      "epoch": 4.23,
+      "learning_rate": 3.5791767007138456e-05,
+      "loss": 0.0353,
+      "step": 165
+    },
+    {
+      "epoch": 4.26,
+      "learning_rate": 3.573489876066967e-05,
+      "loss": 0.0297,
+      "step": 166
+    },
+    {
+      "epoch": 4.28,
+      "learning_rate": 3.567769466439588e-05,
+      "loss": 0.0381,
+      "step": 167
+    },
+    {
+      "epoch": 4.31,
+      "learning_rate": 3.5620155939296314e-05,
+      "loss": 0.0381,
+      "step": 168
+    },
+    {
+      "epoch": 4.33,
+      "learning_rate": 3.556228381349261e-05,
+      "loss": 0.0439,
+      "step": 169
+    },
+    {
+      "epoch": 4.36,
+      "learning_rate": 3.55040795222226e-05,
+      "loss": 0.0311,
+      "step": 170
+    },
+    {
+      "epoch": 4.38,
+      "learning_rate": 3.544554430781394e-05,
+      "loss": 0.0312,
+      "step": 171
+    },
+    {
+      "epoch": 4.41,
+      "learning_rate": 3.538667941965758e-05,
+      "loss": 0.0249,
+      "step": 172
+    },
+    {
+      "epoch": 4.44,
+      "learning_rate": 3.5327486114181144e-05,
+      "loss": 0.0377,
+      "step": 173
+    },
+    {
+      "epoch": 4.46,
+      "learning_rate": 3.526796565482206e-05,
+      "loss": 0.0352,
+      "step": 174
+    },
+    {
+      "epoch": 4.49,
+      "learning_rate": 3.520811931200063e-05,
+      "loss": 0.0336,
+      "step": 175
+    },
+    {
+      "epoch": 4.51,
+      "learning_rate": 3.514794836309286e-05,
+      "loss": 0.0335,
+      "step": 176
+    },
+    {
+      "epoch": 4.54,
+      "learning_rate": 3.5087454092403285e-05,
+      "loss": 0.0328,
+      "step": 177
+    },
+    {
+      "epoch": 4.56,
+      "learning_rate": 3.502663779113747e-05,
+      "loss": 0.0435,
+      "step": 178
+    },
+    {
+      "epoch": 4.59,
+      "learning_rate": 3.49655007573745e-05,
+      "loss": 0.0301,
+      "step": 179
+    },
+    {
+      "epoch": 4.62,
+      "learning_rate": 3.490404429603925e-05,
+      "loss": 0.032,
+      "step": 180
+    },
+    {
+      "epoch": 4.64,
+      "learning_rate": 3.484226971887456e-05,
+      "loss": 0.0459,
+      "step": 181
+    },
+    {
+      "epoch": 4.67,
+      "learning_rate": 3.478017834441319e-05,
+      "loss": 0.0393,
+      "step": 182
+    },
+    {
+      "epoch": 4.69,
+      "learning_rate": 3.4717771497949706e-05,
+      "loss": 0.0295,
+      "step": 183
+    },
+    {
+      "epoch": 4.72,
+      "learning_rate": 3.4655050511512236e-05,
+      "loss": 0.0337,
+      "step": 184
+    },
+    {
+      "epoch": 4.74,
+      "learning_rate": 3.459201672383392e-05,
+      "loss": 0.0331,
+      "step": 185
+    },
+    {
+      "epoch": 4.77,
+      "learning_rate": 3.452867148032449e-05,
+      "loss": 0.0197,
+      "step": 186
+    },
+    {
+      "epoch": 4.79,
+      "learning_rate": 3.4465016133041405e-05,
+      "loss": 0.0245,
+      "step": 187
+    },
+    {
+      "epoch": 4.82,
+      "learning_rate": 3.44010520406611e-05,
+      "loss": 0.0272,
+      "step": 188
+    },
+    {
+      "epoch": 4.85,
+      "learning_rate": 3.433678056844993e-05,
+      "loss": 0.0427,
+      "step": 189
+    },
+    {
+      "epoch": 4.87,
+      "learning_rate": 3.427220308823505e-05,
+      "loss": 0.0439,
+      "step": 190
+    },
+    {
+      "epoch": 4.9,
+      "learning_rate": 3.420732097837514e-05,
+      "loss": 0.0313,
+      "step": 191
+    },
+    {
+      "epoch": 4.92,
+      "learning_rate": 3.4142135623730954e-05,
+      "loss": 0.0299,
+      "step": 192
+    },
+    {
+      "epoch": 4.95,
+      "learning_rate": 3.4076648415635804e-05,
+      "loss": 0.0363,
+      "step": 193
+    },
+    {
+      "epoch": 4.97,
+      "learning_rate": 3.401086075186582e-05,
+      "loss": 0.0394,
+      "step": 194
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 3.394477403661016e-05,
+      "loss": 0.0246,
+      "step": 195
+    },
+    {
+      "epoch": 5.03,
+      "learning_rate": 3.3878389680440995e-05,
+      "loss": 0.0124,
+      "step": 196
+    },
+    {
+      "epoch": 5.05,
+      "learning_rate": 3.3811709100283434e-05,
+      "loss": 0.0155,
+      "step": 197
+    },
+    {
+      "epoch": 5.08,
+      "learning_rate": 3.374473371938526e-05,
+      "loss": 0.0143,
+      "step": 198
+    },
+    {
+      "epoch": 5.1,
+      "learning_rate": 3.367746496728656e-05,
+      "loss": 0.0179,
+      "step": 199
+    },
+    {
+      "epoch": 5.13,
+      "learning_rate": 3.3609904279789235e-05,
+      "loss": 0.0174,
+      "step": 200
+    },
+    {
+      "epoch": 5.15,
+      "learning_rate": 3.3542053098926296e-05,
+      "loss": 0.03,
+      "step": 201
+    },
+    {
+      "epoch": 5.18,
+      "learning_rate": 3.347391287293115e-05,
+      "loss": 0.0258,
+      "step": 202
+    },
+    {
+      "epoch": 5.21,
+      "learning_rate": 3.3405485056206636e-05,
+      "loss": 0.0297,
+      "step": 203
+    },
+    {
+      "epoch": 5.23,
+      "learning_rate": 3.333677110929403e-05,
+      "loss": 0.0498,
+      "step": 204
+    },
+    {
+      "epoch": 5.26,
+      "learning_rate": 3.326777249884183e-05,
+      "loss": 0.0205,
+      "step": 205
+    },
+    {
+      "epoch": 5.28,
+      "learning_rate": 3.319849069757446e-05,
+      "loss": 0.0185,
+      "step": 206
+    },
+    {
+      "epoch": 5.31,
+      "learning_rate": 3.312892718426086e-05,
+      "loss": 0.0164,
+      "step": 207
+    },
+    {
+      "epoch": 5.33,
+      "learning_rate": 3.305908344368289e-05,
+      "loss": 0.014,
+      "step": 208
+    },
+    {
+      "epoch": 5.36,
+      "learning_rate": 3.298896096660367e-05,
+      "loss": 0.0182,
+      "step": 209
+    },
+    {
+      "epoch": 5.38,
+      "learning_rate": 3.291856124973575e-05,
+      "loss": 0.0271,
+      "step": 210
+    },
+    {
+      "epoch": 5.41,
+      "learning_rate": 3.284788579570912e-05,
+      "loss": 0.0203,
+      "step": 211
+    },
+    {
+      "epoch": 5.44,
+      "learning_rate": 3.277693611303922e-05,
+      "loss": 0.0306,
+      "step": 212
+    },
+    {
+      "epoch": 5.46,
+      "learning_rate": 3.2705713716094694e-05,
+      "loss": 0.0174,
+      "step": 213
+    },
+    {
+      "epoch": 5.49,
+      "learning_rate": 3.263422012506502e-05,
+      "loss": 0.0152,
+      "step": 214
+    },
+    {
+      "epoch": 5.51,
+      "learning_rate": 3.2562456865928184e-05,
+      "loss": 0.0343,
+      "step": 215
+    },
+    {
+      "epoch": 5.54,
+      "learning_rate": 3.249042547041799e-05,
+      "loss": 0.0117,
+      "step": 216
+    },
+    {
+      "epoch": 5.56,
+      "learning_rate": 3.241812747599143e-05,
+      "loss": 0.025,
+      "step": 217
+    },
+    {
+      "epoch": 5.59,
+      "learning_rate": 3.234556442579586e-05,
+      "loss": 0.0264,
+      "step": 218
+    },
+    {
+      "epoch": 5.62,
+      "learning_rate": 3.2272737868636056e-05,
+      "loss": 0.0212,
+      "step": 219
+    },
+    {
+      "epoch": 5.64,
+      "learning_rate": 3.219964935894114e-05,
+      "loss": 0.0187,
+      "step": 220
+    },
+    {
+      "epoch": 5.67,
+      "learning_rate": 3.2126300456731425e-05,
+      "loss": 0.0292,
+      "step": 221
+    },
+    {
+      "epoch": 5.69,
+      "learning_rate": 3.205269272758513e-05,
+      "loss": 0.0223,
+      "step": 222
+    },
+    {
+      "epoch": 5.72,
+      "learning_rate": 3.197882774260491e-05,
+      "loss": 0.0192,
+      "step": 223
+    },
+    {
+      "epoch": 5.74,
+      "learning_rate": 3.190470707838438e-05,
+      "loss": 0.0209,
+      "step": 224
+    },
+    {
+      "epoch": 5.77,
+      "learning_rate": 3.1830332316974427e-05,
+      "loss": 0.0167,
+      "step": 225
+    },
+    {
+      "epoch": 5.79,
+      "learning_rate": 3.1755705045849465e-05,
+      "loss": 0.0223,
+      "step": 226
+    },
+    {
+      "epoch": 5.82,
+      "learning_rate": 3.1680826857873534e-05,
+      "loss": 0.0232,
+      "step": 227
+    },
+    {
+      "epoch": 5.85,
+      "learning_rate": 3.160569935126632e-05,
+      "loss": 0.0322,
+      "step": 228
+    },
+    {
+      "epoch": 5.87,
+      "learning_rate": 3.153032412956901e-05,
+      "loss": 0.0291,
+      "step": 229
+    },
+    {
+      "epoch": 5.9,
+      "learning_rate": 3.145470280161011e-05,
+      "loss": 0.0208,
+      "step": 230
+    },
+    {
+      "epoch": 5.92,
+      "learning_rate": 3.1378836981471066e-05,
+      "loss": 0.016,
+      "step": 231
+    },
+    {
+      "epoch": 5.95,
+      "learning_rate": 3.130272828845184e-05,
+      "loss": 0.0286,
+      "step": 232
+    },
+    {
+      "epoch": 5.95,
+      "eval_test_accuracy": 80.76923076923077,
+      "eval_test_average": 81.63979173281498,
+      "eval_test_loss": 0.8692004680633545,
+      "eval_test_recall": 82.51035269639921,
+      "eval_test_runtime": 794.04,
+      "eval_test_samples_per_second": 1.336,
+      "eval_test_steps_per_second": 0.006,
+      "step": 232
+    },
+    {
+      "epoch": 5.95,
+      "eval_val_accuracy": 82.46445497630332,
+      "eval_val_average": 83.1463368864132,
+      "eval_val_loss": 0.8709683418273926,
+      "eval_val_recall": 83.82821879652307,
+      "eval_val_runtime": 781.0898,
+      "eval_val_samples_per_second": 1.348,
+      "eval_val_steps_per_second": 0.006,
+      "step": 232
+    },
+    {
+      "epoch": 5.97,
+      "learning_rate": 3.1226378347036334e-05,
+      "loss": 0.0181,
+      "step": 233
+    },
+    {
+      "epoch": 6.0,
+      "learning_rate": 3.114978878685771e-05,
+      "loss": 0.0241,
+      "step": 234
+    },
+    {
+      "epoch": 6.03,
+      "learning_rate": 3.1072961242663635e-05,
+      "loss": 0.0094,
+      "step": 235
+    },
+    {
+      "epoch": 6.05,
+      "learning_rate": 3.0995897354281347e-05,
+      "loss": 0.0085,
+      "step": 236
+    },
+    {
+      "epoch": 6.08,
+      "learning_rate": 3.091859876658269e-05,
+      "loss": 0.0178,
+      "step": 237
+    },
+    {
+      "epoch": 6.1,
+      "learning_rate": 3.084106712944899e-05,
+      "loss": 0.013,
+      "step": 238
+    },
+    {
+      "epoch": 6.13,
+      "learning_rate": 3.076330409773584e-05,
+      "loss": 0.0132,
+      "step": 239
+    },
+    {
+      "epoch": 6.15,
+      "learning_rate": 3.068531133123777e-05,
+      "loss": 0.0136,
+      "step": 240
+    },
+    {
+      "epoch": 6.18,
+      "learning_rate": 3.060709049465285e-05,
+      "loss": 0.0089,
+      "step": 241
+    },
+    {
+      "epoch": 6.21,
+      "learning_rate": 3.052864325754712e-05,
+      "loss": 0.0155,
+      "step": 242
+    },
+    {
+      "epoch": 6.23,
+      "learning_rate": 3.0449971294318977e-05,
+      "loss": 0.0156,
+      "step": 243
+    },
+    {
+      "epoch": 6.26,
+      "learning_rate": 3.0371076284163442e-05,
+      "loss": 0.0181,
+      "step": 244
+    },
+    {
+      "epoch": 6.28,
+      "learning_rate": 3.0291959911036293e-05,
+      "loss": 0.0136,
+      "step": 245
+    },
+    {
+      "epoch": 6.31,
+      "learning_rate": 3.021262386361814e-05,
+      "loss": 0.0094,
+      "step": 246
+    },
+    {
+      "epoch": 6.33,
+      "learning_rate": 3.013306983527839e-05,
+      "loss": 0.0117,
+      "step": 247
+    },
+    {
+      "epoch": 6.36,
+      "learning_rate": 3.0053299524039077e-05,
+      "loss": 0.0126,
+      "step": 248
+    },
+    {
+      "epoch": 6.38,
+      "learning_rate": 2.997331463253864e-05,
+      "loss": 0.01,
+      "step": 249
+    },
+    {
+      "epoch": 6.41,
+      "learning_rate": 2.9893116867995583e-05,
+      "loss": 0.0129,
+      "step": 250
+    },
+    {
+      "epoch": 6.44,
+      "learning_rate": 2.981270794217201e-05,
+      "loss": 0.0068,
+      "step": 251
+    },
+    {
+      "epoch": 6.46,
+      "learning_rate": 2.9732089571337126e-05,
+      "loss": 0.0077,
+      "step": 252
+    },
+    {
+      "epoch": 6.49,
+      "learning_rate": 2.9651263476230577e-05,
+      "loss": 0.0066,
+      "step": 253
+    },
+    {
+      "epoch": 6.51,
+      "learning_rate": 2.9570231382025732e-05,
+      "loss": 0.0044,
+      "step": 254
+    },
+    {
+      "epoch": 6.54,
+      "learning_rate": 2.9488995018292854e-05,
+      "loss": 0.0123,
+      "step": 255
+    },
+    {
+      "epoch": 6.56,
+      "learning_rate": 2.9407556118962192e-05,
+      "loss": 0.0119,
+      "step": 256
+    },
+    {
+      "epoch": 6.59,
+      "learning_rate": 2.932591642228696e-05,
+      "loss": 0.0141,
+      "step": 257
+    },
+    {
+      "epoch": 6.62,
+      "learning_rate": 2.924407767080627e-05,
+      "loss": 0.0075,
+      "step": 258
+    },
+    {
+      "epoch": 6.64,
+      "learning_rate": 2.9162041611307868e-05,
+      "loss": 0.0106,
+      "step": 259
+    },
+    {
+      "epoch": 6.67,
+      "learning_rate": 2.9079809994790937e-05,
+      "loss": 0.0152,
+      "step": 260
+    },
+    {
+      "epoch": 6.69,
+      "learning_rate": 2.899738457642866e-05,
+      "loss": 0.0167,
+      "step": 261
+    },
+    {
+      "epoch": 6.72,
+      "learning_rate": 2.891476711553077e-05,
+      "loss": 0.0072,
+      "step": 262
+    },
+    {
+      "epoch": 6.74,
+      "learning_rate": 2.883195937550602e-05,
+      "loss": 0.0107,
+      "step": 263
+    },
+    {
+      "epoch": 6.77,
+      "learning_rate": 2.8748963123824532e-05,
+      "loss": 0.0088,
+      "step": 264
+    },
+    {
+      "epoch": 6.79,
+      "learning_rate": 2.8665780131980054e-05,
+      "loss": 0.0103,
+      "step": 265
+    },
+    {
+      "epoch": 6.82,
+      "learning_rate": 2.858241217545218e-05,
+      "loss": 0.0219,
+      "step": 266
+    },
+    {
+      "epoch": 6.85,
+      "learning_rate": 2.8498861033668444e-05,
+      "loss": 0.0107,
+      "step": 267
+    },
+    {
+      "epoch": 6.87,
+      "learning_rate": 2.8415128489966308e-05,
+      "loss": 0.0088,
+      "step": 268
+    },
+    {
+      "epoch": 6.9,
+      "learning_rate": 2.8331216331555148e-05,
+      "loss": 0.0171,
+      "step": 269
+    },
+    {
+      "epoch": 6.92,
+      "learning_rate": 2.8247126349478073e-05,
+      "loss": 0.0102,
+      "step": 270
+    },
+    {
+      "epoch": 6.95,
+      "learning_rate": 2.8162860338573718e-05,
+      "loss": 0.0172,
+      "step": 271
+    },
+    {
+      "epoch": 6.97,
+      "learning_rate": 2.80784200974379e-05,
+      "loss": 0.0089,
+      "step": 272
+    },
+    {
+      "epoch": 7.0,
+      "learning_rate": 2.7993807428385285e-05,
+      "loss": 0.0048,
+      "step": 273
+    },
+    {
+      "epoch": 7.03,
+      "learning_rate": 2.790902413741085e-05,
+      "loss": 0.0098,
+      "step": 274
+    },
+    {
+      "epoch": 7.05,
+      "learning_rate": 2.7824072034151406e-05,
+      "loss": 0.0252,
+      "step": 275
+    },
+    {
+      "epoch": 7.08,
+      "learning_rate": 2.773895293184691e-05,
+      "loss": 0.0053,
+      "step": 276
+    },
+    {
+      "epoch": 7.1,
+      "learning_rate": 2.7653668647301797e-05,
+      "loss": 0.0202,
+      "step": 277
+    },
+    {
+      "epoch": 7.13,
+      "learning_rate": 2.756822100084621e-05,
+      "loss": 0.012,
+      "step": 278
+    },
+    {
+      "epoch": 7.15,
+      "learning_rate": 2.748261181629711e-05,
+      "loss": 0.0082,
+      "step": 279
+    },
+    {
+      "epoch": 7.18,
+      "learning_rate": 2.7396842920919384e-05,
+      "loss": 0.0163,
+      "step": 280
+    },
+    {
+      "epoch": 7.21,
+      "learning_rate": 2.7310916145386826e-05,
+      "loss": 0.0186,
+      "step": 281
+    },
+    {
+      "epoch": 7.23,
+      "learning_rate": 2.7224833323743064e-05,
+      "loss": 0.0088,
+      "step": 282
+    },
+    {
+      "epoch": 7.26,
+      "learning_rate": 2.7138596293362404e-05,
+      "loss": 0.0137,
+      "step": 283
+    },
+    {
+      "epoch": 7.28,
+      "learning_rate": 2.7052206894910653e-05,
+      "loss": 0.0086,
+      "step": 284
+    },
+    {
+      "epoch": 7.31,
+      "learning_rate": 2.696566697230577e-05,
+      "loss": 0.0063,
+      "step": 285
+    },
+    {
+      "epoch": 7.33,
+      "learning_rate": 2.6878978372678567e-05,
+      "loss": 0.0114,
+      "step": 286
+    },
+    {
+      "epoch": 7.36,
+      "learning_rate": 2.6792142946333227e-05,
+      "loss": 0.0104,
+      "step": 287
+    },
+    {
+      "epoch": 7.38,
+      "learning_rate": 2.670516254670788e-05,
+      "loss": 0.0073,
+      "step": 288
+    },
+    {
+      "epoch": 7.41,
+      "learning_rate": 2.6618039030334962e-05,
+      "loss": 0.0112,
+      "step": 289
+    },
+    {
+      "epoch": 7.44,
+      "learning_rate": 2.6530774256801666e-05,
+      "loss": 0.004,
+      "step": 290
+    },
+    {
+      "epoch": 7.46,
+      "learning_rate": 2.6443370088710197e-05,
+      "loss": 0.0045,
+      "step": 291
+    },
+    {
+      "epoch": 7.49,
+      "learning_rate": 2.6355828391638036e-05,
+      "loss": 0.008,
+      "step": 292
+    },
+    {
+      "epoch": 7.51,
+      "learning_rate": 2.6268151034098117e-05,
+      "loss": 0.0069,
+      "step": 293
+    },
+    {
+      "epoch": 7.54,
+      "learning_rate": 2.618033988749895e-05,
+      "loss": 0.0086,
+      "step": 294
+    },
+    {
+      "epoch": 7.56,
+      "learning_rate": 2.6092396826104674e-05,
+      "loss": 0.0062,
+      "step": 295
+    },
+    {
+      "epoch": 7.59,
+      "learning_rate": 2.6004323726995057e-05,
+      "loss": 0.0085,
+      "step": 296
+    },
+    {
+      "epoch": 7.62,
+      "learning_rate": 2.5916122470025414e-05,
+      "loss": 0.0078,
+      "step": 297
+    },
+    {
+      "epoch": 7.64,
+      "learning_rate": 2.5827794937786497e-05,
+      "loss": 0.0051,
+      "step": 298
+    },
+    {
+      "epoch": 7.67,
+      "learning_rate": 2.573934301556432e-05,
+      "loss": 0.009,
+      "step": 299
+    },
+    {
+      "epoch": 7.69,
+      "learning_rate": 2.5650768591299905e-05,
+      "loss": 0.0063,
+      "step": 300
+    },
+    {
+      "epoch": 7.72,
+      "learning_rate": 2.5562073555548984e-05,
+      "loss": 0.0052,
+      "step": 301
+    },
+    {
+      "epoch": 7.74,
+      "learning_rate": 2.5473259801441663e-05,
+      "loss": 0.0064,
+      "step": 302
+    },
+    {
+      "epoch": 7.77,
+      "learning_rate": 2.5384329224641994e-05,
+      "loss": 0.0069,
+      "step": 303
+    },
+    {
+      "epoch": 7.79,
+      "learning_rate": 2.5295283723307517e-05,
+      "loss": 0.0129,
+      "step": 304
+    },
+    {
+      "epoch": 7.82,
+      "learning_rate": 2.520612519804878e-05,
+      "loss": 0.0032,
+      "step": 305
+    },
+    {
+      "epoch": 7.85,
+      "learning_rate": 2.5116855551888715e-05,
+      "loss": 0.0026,
+      "step": 306
+    },
+    {
+      "epoch": 7.87,
+      "learning_rate": 2.5027476690222058e-05,
+      "loss": 0.0063,
+      "step": 307
+    },
+    {
+      "epoch": 7.9,
+      "learning_rate": 2.4937990520774664e-05,
+      "loss": 0.0089,
+      "step": 308
+    },
+    {
+      "epoch": 7.92,
+      "learning_rate": 2.4848398953562806e-05,
+      "loss": 0.004,
+      "step": 309
+    },
+    {
+      "epoch": 7.95,
+      "learning_rate": 2.4758703900852376e-05,
+      "loss": 0.019,
+      "step": 310
+    },
+    {
+      "epoch": 7.97,
+      "learning_rate": 2.4668907277118114e-05,
+      "loss": 0.0107,
+      "step": 311
+    },
+    {
+      "epoch": 8.0,
+      "learning_rate": 2.4579010999002683e-05,
+      "loss": 0.015,
+      "step": 312
+    },
+    {
+      "epoch": 8.03,
+      "learning_rate": 2.448901698527583e-05,
+      "loss": 0.0033,
+      "step": 313
+    },
+    {
+      "epoch": 8.05,
+      "learning_rate": 2.4398927156793376e-05,
+      "loss": 0.0044,
+      "step": 314
+    },
+    {
+      "epoch": 8.08,
+      "learning_rate": 2.4308743436456238e-05,
+      "loss": 0.0022,
+      "step": 315
+    },
+    {
+      "epoch": 8.1,
+      "learning_rate": 2.42184677491694e-05,
+      "loss": 0.0025,
+      "step": 316
+    },
+    {
+      "epoch": 8.13,
+      "learning_rate": 2.4128102021800794e-05,
+      "loss": 0.0037,
+      "step": 317
+    },
+    {
+      "epoch": 8.15,
+      "learning_rate": 2.4037648183140205e-05,
+      "loss": 0.003,
+      "step": 318
+    },
+    {
+      "epoch": 8.18,
+      "learning_rate": 2.3947108163858085e-05,
+      "loss": 0.0044,
+      "step": 319
+    },
+    {
+      "epoch": 8.21,
+      "learning_rate": 2.385648389646434e-05,
+      "loss": 0.0094,
+      "step": 320
+    },
+    {
+      "epoch": 8.23,
+      "learning_rate": 2.37657773152671e-05,
+      "loss": 0.0162,
+      "step": 321
+    },
+    {
+      "epoch": 8.26,
+      "learning_rate": 2.367499035633141e-05,
+      "loss": 0.007,
+      "step": 322
+    },
+    {
+      "epoch": 8.28,
+      "learning_rate": 2.358412495743794e-05,
+      "loss": 0.0039,
+      "step": 323
+    },
+    {
+      "epoch": 8.31,
+      "learning_rate": 2.3493183058041578e-05,
+      "loss": 0.006,
+      "step": 324
+    },
+    {
+      "epoch": 8.33,
+      "learning_rate": 2.340216659923008e-05,
+      "loss": 0.0022,
+      "step": 325
+    },
+    {
+      "epoch": 8.36,
+      "learning_rate": 2.33110775236826e-05,
+      "loss": 0.0083,
+      "step": 326
+    },
+    {
+      "epoch": 8.38,
+      "learning_rate": 2.321991777562826e-05,
+      "loss": 0.0064,
+      "step": 327
+    },
+    {
+      "epoch": 8.41,
+      "learning_rate": 2.312868930080462e-05,
+      "loss": 0.0038,
+      "step": 328
+    },
+    {
+      "epoch": 8.44,
+      "learning_rate": 2.303739404641617e-05,
+      "loss": 0.0039,
+      "step": 329
+    },
+    {
+      "epoch": 8.46,
+      "learning_rate": 2.2946033961092754e-05,
+      "loss": 0.0039,
+      "step": 330
+    },
+    {
+      "epoch": 8.49,
+      "learning_rate": 2.285461099484799e-05,
+      "loss": 0.009,
+      "step": 331
+    },
+    {
+      "epoch": 8.51,
+      "learning_rate": 2.2763127099037646e-05,
+      "loss": 0.0046,
+      "step": 332
+    },
+    {
+      "epoch": 8.54,
+      "learning_rate": 2.267158422631798e-05,
+      "loss": 0.0055,
+      "step": 333
+    },
+    {
+      "epoch": 8.56,
+      "learning_rate": 2.257998433060407e-05,
+      "loss": 0.0048,
+      "step": 334
+    },
+    {
+      "epoch": 8.59,
+      "learning_rate": 2.248832936702811e-05,
+      "loss": 0.0032,
+      "step": 335
+    },
+    {
+      "epoch": 8.62,
+      "learning_rate": 2.2396621291897666e-05,
+      "loss": 0.0031,
+      "step": 336
+    },
+    {
+      "epoch": 8.64,
+      "learning_rate": 2.2304862062653956e-05,
+      "loss": 0.0036,
+      "step": 337
+    },
+    {
+      "epoch": 8.67,
+      "learning_rate": 2.2213053637830016e-05,
+      "loss": 0.0018,
+      "step": 338
+    },
+    {
+      "epoch": 8.69,
+      "learning_rate": 2.2121197977008953e-05,
+      "loss": 0.0057,
+      "step": 339
+    },
+    {
+      "epoch": 8.72,
+      "learning_rate": 2.2029297040782063e-05,
+      "loss": 0.0043,
+      "step": 340
+    },
+    {
+      "epoch": 8.74,
+      "learning_rate": 2.1937352790707028e-05,
+      "loss": 0.0063,
+      "step": 341
+    },
+    {
+      "epoch": 8.77,
+      "learning_rate": 2.184536718926604e-05,
+      "loss": 0.0068,
+      "step": 342
+    },
+    {
+      "epoch": 8.79,
+      "learning_rate": 2.1753342199823894e-05,
+      "loss": 0.0032,
+      "step": 343
+    },
+    {
+      "epoch": 8.82,
+      "learning_rate": 2.166127978658608e-05,
+      "loss": 0.003,
+      "step": 344
+    },
+    {
+      "epoch": 8.85,
+      "learning_rate": 2.1569181914556904e-05,
+      "loss": 0.0091,
+      "step": 345
+    },
+    {
+      "epoch": 8.87,
+      "learning_rate": 2.147705054949748e-05,
+      "loss": 0.0047,
+      "step": 346
+    },
+    {
+      "epoch": 8.9,
+      "learning_rate": 2.1384887657883836e-05,
+      "loss": 0.0019,
+      "step": 347
+    },
+    {
+      "epoch": 8.92,
+      "learning_rate": 2.1292695206864887e-05,
+      "loss": 0.0036,
+      "step": 348
+    },
+    {
+      "epoch": 8.92,
+      "eval_test_accuracy": 83.89423076923077,
+      "eval_test_average": 83.77504585208074,
+      "eval_test_loss": 1.025055170059204,
+      "eval_test_recall": 83.6558609349307,
+      "eval_test_runtime": 778.8006,
+      "eval_test_samples_per_second": 1.362,
+      "eval_test_steps_per_second": 0.006,
+      "step": 348
+    },
+    {
+      "epoch": 8.92,
+      "eval_val_accuracy": 83.88625592417061,
+      "eval_val_average": 84.50143402474953,
+      "eval_val_loss": 1.0246869325637817,
+      "eval_val_recall": 85.11661212532843,
+      "eval_val_runtime": 801.1127,
+      "eval_val_samples_per_second": 1.314,
+      "eval_val_steps_per_second": 0.006,
+      "step": 348
+    },
+    {
+      "epoch": 8.95,
+      "learning_rate": 2.1200475164220488e-05,
+      "loss": 0.0042,
+      "step": 349
+    },
+    {
+      "epoch": 8.97,
+      "learning_rate": 2.11082294983194e-05,
+      "loss": 0.0038,
+      "step": 350
+    },
+    {
+      "epoch": 9.0,
+      "learning_rate": 2.1015960178077317e-05,
+      "loss": 0.0019,
+      "step": 351
+    },
+    {
+      "epoch": 9.03,
+      "learning_rate": 2.0923669172914796e-05,
+      "loss": 0.0005,
+      "step": 352
+    },
+    {
+      "epoch": 9.05,
+      "learning_rate": 2.0831358452715255e-05,
+      "loss": 0.0036,
+      "step": 353
+    },
+    {
+      "epoch": 9.08,
+      "learning_rate": 2.0739029987782903e-05,
+      "loss": 0.001,
+      "step": 354
+    },
+    {
+      "epoch": 9.1,
+      "learning_rate": 2.0646685748800716e-05,
+      "loss": 0.0015,
+      "step": 355
+    },
+    {
+      "epoch": 9.13,
+      "learning_rate": 2.055432770678833e-05,
+      "loss": 0.0061,
+      "step": 356
+    },
+    {
+      "epoch": 9.15,
+      "learning_rate": 2.0461957833060025e-05,
+      "loss": 0.0025,
+      "step": 357
+    },
+    {
+      "epoch": 9.18,
+      "learning_rate": 2.03695780991826e-05,
+      "loss": 0.001,
+      "step": 358
+    },
+    {
+      "epoch": 9.21,
+      "learning_rate": 2.0277190476933318e-05,
+      "loss": 0.0037,
+      "step": 359
+    },
+    {
+      "epoch": 9.23,
+      "learning_rate": 2.018479693825782e-05,
+      "loss": 0.0011,
+      "step": 360
+    },
+    {
+      "epoch": 9.26,
+      "learning_rate": 2.009239945522801e-05,
+      "loss": 0.005,
+      "step": 361
+    },
+    {
+      "epoch": 9.28,
+      "learning_rate": 2e-05,
+      "loss": 0.0045,
+      "step": 362
+    },
+    {
+      "epoch": 9.31,
+      "learning_rate": 1.9907600544771994e-05,
+      "loss": 0.003,
+      "step": 363
+    },
+    {
+      "epoch": 9.33,
+      "learning_rate": 1.9815203061742188e-05,
+      "loss": 0.0032,
+      "step": 364
+    },
+    {
+      "epoch": 9.36,
+      "learning_rate": 1.9722809523066685e-05,
+      "loss": 0.0088,
+      "step": 365
+    },
+    {
+      "epoch": 9.38,
+      "learning_rate": 1.9630421900817407e-05,
+      "loss": 0.0009,
+      "step": 366
+    },
+    {
+      "epoch": 9.41,
+      "learning_rate": 1.9538042166939982e-05,
+      "loss": 0.0043,
+      "step": 367
+    },
+    {
+      "epoch": 9.44,
+      "learning_rate": 1.9445672293211675e-05,
+      "loss": 0.0029,
+      "step": 368
+    },
+    {
+      "epoch": 9.46,
+      "learning_rate": 1.9353314251199298e-05,
+      "loss": 0.0022,
+      "step": 369
+    },
+    {
+      "epoch": 9.49,
+      "learning_rate": 1.9260970012217107e-05,
+      "loss": 0.0012,
+      "step": 370
+    },
+    {
+      "epoch": 9.51,
+      "learning_rate": 1.916864154728476e-05,
+      "loss": 0.0037,
+      "step": 371
+    },
+    {
+      "epoch": 9.54,
+      "learning_rate": 1.9076330827085214e-05,
+      "loss": 0.006,
+      "step": 372
+    },
+    {
+      "epoch": 9.56,
+      "learning_rate": 1.898403982192269e-05,
+      "loss": 0.0051,
+      "step": 373
+    },
+    {
+      "epoch": 9.59,
+      "learning_rate": 1.8891770501680602e-05,
+      "loss": 0.0006,
+      "step": 374
+    },
+    {
+      "epoch": 9.62,
+      "learning_rate": 1.879952483577952e-05,
+      "loss": 0.002,
+      "step": 375
+    },
+    {
+      "epoch": 9.64,
+      "learning_rate": 1.8707304793135117e-05,
+      "loss": 0.0023,
+      "step": 376
+    },
+    {
+      "epoch": 9.67,
+      "learning_rate": 1.861511234211617e-05,
+      "loss": 0.0031,
+      "step": 377
+    },
+    {
+      "epoch": 9.69,
+      "learning_rate": 1.8522949450502522e-05,
+      "loss": 0.0057,
+      "step": 378
+    },
+    {
+      "epoch": 9.72,
+      "learning_rate": 1.8430818085443106e-05,
+      "loss": 0.0008,
+      "step": 379
+    },
+    {
+      "epoch": 9.74,
+      "learning_rate": 1.8338720213413924e-05,
+      "loss": 0.0014,
+      "step": 380
+    },
+    {
+      "epoch": 9.77,
+      "learning_rate": 1.8246657800176116e-05,
+      "loss": 0.001,
+      "step": 381
+    },
+    {
+      "epoch": 9.79,
+      "learning_rate": 1.815463281073396e-05,
+      "loss": 0.0013,
+      "step": 382
+    },
+    {
+      "epoch": 9.82,
+      "learning_rate": 1.8062647209292975e-05,
+      "loss": 0.0014,
+      "step": 383
+    },
+    {
+      "epoch": 9.85,
+      "learning_rate": 1.7970702959217944e-05,
+      "loss": 0.009,
+      "step": 384
+    },
+    {
+      "epoch": 9.87,
+      "learning_rate": 1.7878802022991054e-05,
+      "loss": 0.0038,
+      "step": 385
+    },
+    {
+      "epoch": 9.9,
+      "learning_rate": 1.7786946362169987e-05,
+      "loss": 0.0041,
+      "step": 386
+    },
+    {
+      "epoch": 9.92,
+      "learning_rate": 1.769513793734605e-05,
+      "loss": 0.0014,
+      "step": 387
+    },
+    {
+      "epoch": 9.95,
+      "learning_rate": 1.760337870810234e-05,
+      "loss": 0.0006,
+      "step": 388
+    },
+    {
+      "epoch": 9.97,
+      "learning_rate": 1.75116706329719e-05,
+      "loss": 0.0009,
+      "step": 389
+    },
+    {
+      "epoch": 10.0,
+      "learning_rate": 1.742001566939594e-05,
+      "loss": 0.0011,
+      "step": 390
+    },
+    {
+      "epoch": 10.03,
+      "learning_rate": 1.7328415773682028e-05,
+      "loss": 0.0002,
+      "step": 391
+    },
+    {
+      "epoch": 10.05,
+      "learning_rate": 1.7236872900962364e-05,
+      "loss": 0.0008,
+      "step": 392
+    },
+    {
+      "epoch": 10.08,
+      "learning_rate": 1.7145389005152017e-05,
+      "loss": 0.0005,
+      "step": 393
+    },
+    {
+      "epoch": 10.1,
+      "learning_rate": 1.705396603890725e-05,
+      "loss": 0.0007,
+      "step": 394
+    },
+    {
+      "epoch": 10.13,
+      "learning_rate": 1.6962605953583833e-05,
+      "loss": 0.0011,
+      "step": 395
+    },
+    {
+      "epoch": 10.15,
+      "learning_rate": 1.687131069919538e-05,
+      "loss": 0.0014,
+      "step": 396
+    },
+    {
+      "epoch": 10.18,
+      "learning_rate": 1.678008222437174e-05,
+      "loss": 0.0006,
+      "step": 397
+    },
+    {
+      "epoch": 10.21,
+      "learning_rate": 1.66889224763174e-05,
+      "loss": 0.0004,
+      "step": 398
+    },
+    {
+      "epoch": 10.23,
+      "learning_rate": 1.659783340076992e-05,
+      "loss": 0.002,
+      "step": 399
+    },
+    {
+      "epoch": 10.26,
+      "learning_rate": 1.6506816941958425e-05,
+      "loss": 0.0017,
+      "step": 400
+    },
+    {
+      "epoch": 10.28,
+      "learning_rate": 1.6415875042562065e-05,
+      "loss": 0.001,
+      "step": 401
+    },
+    {
+      "epoch": 10.31,
+      "learning_rate": 1.6325009643668592e-05,
+      "loss": 0.0005,
+      "step": 402
+    },
+    {
+      "epoch": 10.33,
+      "learning_rate": 1.6234222684732908e-05,
+      "loss": 0.0002,
+      "step": 403
+    },
+    {
+      "epoch": 10.36,
+      "learning_rate": 1.6143516103535666e-05,
+      "loss": 0.0002,
+      "step": 404
+    },
+    {
+      "epoch": 10.38,
+      "learning_rate": 1.6052891836141925e-05,
+      "loss": 0.0003,
+      "step": 405
+    },
+    {
+      "epoch": 10.41,
+      "learning_rate": 1.59623518168598e-05,
+      "loss": 0.0003,
+      "step": 406
+    },
+    {
+      "epoch": 10.44,
+      "learning_rate": 1.5871897978199213e-05,
+      "loss": 0.002,
+      "step": 407
+    },
+    {
+      "epoch": 10.46,
+      "learning_rate": 1.578153225083061e-05,
+      "loss": 0.0002,
+      "step": 408
+    },
+    {
+      "epoch": 10.49,
+      "learning_rate": 1.569125656354377e-05,
+      "loss": 0.003,
+      "step": 409
+    },
+    {
+      "epoch": 10.51,
+      "learning_rate": 1.5601072843206634e-05,
+      "loss": 0.0003,
+      "step": 410
+    },
+    {
+      "epoch": 10.54,
+      "learning_rate": 1.5510983014724178e-05,
+      "loss": 0.0005,
+      "step": 411
+    },
+    {
+      "epoch": 10.56,
+      "learning_rate": 1.5420989000997324e-05,
+      "loss": 0.0005,
+      "step": 412
+    },
+    {
+      "epoch": 10.59,
+      "learning_rate": 1.53310927228819e-05,
+      "loss": 0.0011,
+      "step": 413
+    },
+    {
+      "epoch": 10.62,
+      "learning_rate": 1.524129609914763e-05,
+      "loss": 0.0025,
+      "step": 414
+    },
+    {
+      "epoch": 10.64,
+      "learning_rate": 1.5151601046437206e-05,
+      "loss": 0.0005,
+      "step": 415
+    },
+    {
+      "epoch": 10.67,
+      "learning_rate": 1.5062009479225336e-05,
+      "loss": 0.0003,
+      "step": 416
+    },
+    {
+      "epoch": 10.69,
+      "learning_rate": 1.4972523309777947e-05,
+      "loss": 0.0006,
+      "step": 417
+    },
+    {
+      "epoch": 10.72,
+      "learning_rate": 1.4883144448111288e-05,
+      "loss": 0.0008,
+      "step": 418
+    },
+    {
+      "epoch": 10.74,
+      "learning_rate": 1.4793874801951221e-05,
+      "loss": 0.0007,
+      "step": 419
+    },
+    {
+      "epoch": 10.77,
+      "learning_rate": 1.4704716276692483e-05,
+      "loss": 0.0002,
+      "step": 420
+    },
+    {
+      "epoch": 10.79,
+      "learning_rate": 1.4615670775358015e-05,
+      "loss": 0.0013,
+      "step": 421
+    },
+    {
+      "epoch": 10.82,
+      "learning_rate": 1.4526740198558345e-05,
+      "loss": 0.0002,
+      "step": 422
+    },
+    {
+      "epoch": 10.85,
+      "learning_rate": 1.443792644445102e-05,
+      "loss": 0.0009,
+      "step": 423
+    },
+    {
+      "epoch": 10.87,
+      "learning_rate": 1.43492314087001e-05,
+      "loss": 0.002,
+      "step": 424
+    },
+    {
+      "epoch": 10.9,
+      "learning_rate": 1.4260656984435683e-05,
+      "loss": 0.0003,
+      "step": 425
+    },
+    {
+      "epoch": 10.92,
+      "learning_rate": 1.417220506221351e-05,
+      "loss": 0.0009,
+      "step": 426
+    },
+    {
+      "epoch": 10.95,
+      "learning_rate": 1.4083877529974594e-05,
+      "loss": 0.001,
+      "step": 427
+    },
+    {
+      "epoch": 10.97,
+      "learning_rate": 1.3995676273004948e-05,
+      "loss": 0.0002,
+      "step": 428
+    },
+    {
+      "epoch": 11.0,
+      "learning_rate": 1.3907603173895327e-05,
+      "loss": 0.0019,
+      "step": 429
+    },
+    {
+      "epoch": 11.03,
+      "learning_rate": 1.3819660112501054e-05,
+      "loss": 0.0002,
+      "step": 430
+    },
+    {
+      "epoch": 11.05,
+      "learning_rate": 1.3731848965901891e-05,
+      "loss": 0.0001,
+      "step": 431
+    },
+    {
+      "epoch": 11.08,
+      "learning_rate": 1.364417160836197e-05,
+      "loss": 0.0006,
+      "step": 432
+    },
+    {
+      "epoch": 11.1,
+      "learning_rate": 1.355662991128981e-05,
+      "loss": 0.0006,
+      "step": 433
+    },
+    {
+      "epoch": 11.13,
+      "learning_rate": 1.3469225743198337e-05,
+      "loss": 0.0001,
+      "step": 434
+    },
+    {
+      "epoch": 11.15,
+      "learning_rate": 1.3381960969665041e-05,
+      "loss": 0.0038,
+      "step": 435
+    },
+    {
+      "epoch": 11.18,
+      "learning_rate": 1.329483745329213e-05,
+      "loss": 0.0001,
+      "step": 436
+    },
+    {
+      "epoch": 11.21,
+      "learning_rate": 1.3207857053666773e-05,
+      "loss": 0.0004,
+      "step": 437
+    },
+    {
+      "epoch": 11.23,
+      "learning_rate": 1.3121021627321438e-05,
+      "loss": 0.0019,
+      "step": 438
+    },
+    {
+      "epoch": 11.26,
+      "learning_rate": 1.303433302769423e-05,
+      "loss": 0.0008,
+      "step": 439
+    },
+    {
+      "epoch": 11.28,
+      "learning_rate": 1.2947793105089347e-05,
+      "loss": 0.0002,
+      "step": 440
+    },
+    {
+      "epoch": 11.31,
+      "learning_rate": 1.2861403706637594e-05,
+      "loss": 0.0012,
+      "step": 441
+    },
+    {
+      "epoch": 11.33,
+      "learning_rate": 1.2775166676256942e-05,
+      "loss": 0.0007,
+      "step": 442
+    },
+    {
+      "epoch": 11.36,
+      "learning_rate": 1.2689083854613178e-05,
+      "loss": 0.0029,
+      "step": 443
+    },
+    {
+      "epoch": 11.38,
+      "learning_rate": 1.260315707908062e-05,
+      "loss": 0.0001,
+      "step": 444
+    },
+    {
+      "epoch": 11.41,
+      "learning_rate": 1.2517388183702895e-05,
+      "loss": 0.0001,
+      "step": 445
+    },
+    {
+      "epoch": 11.44,
+      "learning_rate": 1.2431778999153796e-05,
+      "loss": 0.0003,
+      "step": 446
+    },
+    {
+      "epoch": 11.46,
+      "learning_rate": 1.2346331352698206e-05,
+      "loss": 0.0002,
+      "step": 447
+    },
+    {
+      "epoch": 11.49,
+      "learning_rate": 1.2261047068153098e-05,
+      "loss": 0.0002,
+      "step": 448
+    },
+    {
+      "epoch": 11.51,
+      "learning_rate": 1.21759279658486e-05,
+      "loss": 0.0005,
+      "step": 449
+    },
+    {
+      "epoch": 11.54,
+      "learning_rate": 1.2090975862589151e-05,
+      "loss": 0.0001,
+      "step": 450
+    },
+    {
+      "epoch": 11.56,
+      "learning_rate": 1.2006192571614723e-05,
+      "loss": 0.0001,
+      "step": 451
+    },
+    {
+      "epoch": 11.59,
+      "learning_rate": 1.1921579902562103e-05,
+      "loss": 0.0006,
+      "step": 452
+    },
+    {
+      "epoch": 11.62,
+      "learning_rate": 1.183713966142629e-05,
+      "loss": 0.0001,
+      "step": 453
+    },
+    {
+      "epoch": 11.64,
+      "learning_rate": 1.1752873650521934e-05,
+      "loss": 0.0001,
+      "step": 454
+    },
+    {
+      "epoch": 11.67,
+      "learning_rate": 1.166878366844486e-05,
+      "loss": 0.0001,
+      "step": 455
+    },
+    {
+      "epoch": 11.69,
+      "learning_rate": 1.1584871510033707e-05,
+      "loss": 0.0001,
+      "step": 456
+    },
+    {
+      "epoch": 11.72,
+      "learning_rate": 1.150113896633157e-05,
+      "loss": 0.0001,
+      "step": 457
+    },
+    {
+      "epoch": 11.74,
+      "learning_rate": 1.1417587824547822e-05,
+      "loss": 0.0001,
+      "step": 458
+    },
+    {
+      "epoch": 11.77,
+      "learning_rate": 1.1334219868019946e-05,
+      "loss": 0.0004,
+      "step": 459
+    },
+    {
+      "epoch": 11.79,
+      "learning_rate": 1.1251036876175476e-05,
+      "loss": 0.0001,
+      "step": 460
+    },
+    {
+      "epoch": 11.82,
+      "learning_rate": 1.1168040624493982e-05,
+      "loss": 0.0012,
+      "step": 461
+    },
+    {
+      "epoch": 11.85,
+      "learning_rate": 1.1085232884469236e-05,
+      "loss": 0.0004,
+      "step": 462
+    },
+    {
+      "epoch": 11.87,
+      "learning_rate": 1.1002615423571344e-05,
+      "loss": 0.0006,
+      "step": 463
+    },
+    {
+      "epoch": 11.9,
+      "learning_rate": 1.0920190005209066e-05,
+      "loss": 0.0001,
+      "step": 464
+    },
+    {
+      "epoch": 11.9,
+      "eval_test_accuracy": 84.85576923076923,
+      "eval_test_average": 84.64341198207478,
+      "eval_test_loss": 1.012276291847229,
+      "eval_test_recall": 84.43105473338032,
+      "eval_test_runtime": 790.3907,
+      "eval_test_samples_per_second": 1.342,
+      "eval_test_steps_per_second": 0.006,
+      "step": 464
+    },
+    {
+      "epoch": 11.9,
+      "eval_val_accuracy": 85.30805687203792,
+      "eval_val_average": 85.0317735118623,
+      "eval_val_loss": 1.010198950767517,
+      "eval_val_recall": 84.75549015168666,
+      "eval_val_runtime": 789.112,
+      "eval_val_samples_per_second": 1.334,
+      "eval_val_steps_per_second": 0.006,
+      "step": 464
+    },
+    {
+      "epoch": 11.92,
+      "learning_rate": 1.083795838869213e-05,
+      "loss": 0.0001,
+      "step": 465
+    },
+    {
+      "epoch": 11.95,
+      "learning_rate": 1.0755922329193739e-05,
+      "loss": 0.0005,
+      "step": 466
+    },
+    {
+      "epoch": 11.97,
+      "learning_rate": 1.0674083577713037e-05,
+      "loss": 0.0001,
+      "step": 467
+    },
+    {
+      "epoch": 12.0,
+      "learning_rate": 1.0592443881037816e-05,
+      "loss": 0.0002,
+      "step": 468
+    },
+    {
+      "epoch": 12.03,
+      "learning_rate": 1.051100498170715e-05,
+      "loss": 0.0001,
+      "step": 469
+    },
+    {
+      "epoch": 12.05,
+      "learning_rate": 1.0429768617974271e-05,
+      "loss": 0.0001,
+      "step": 470
+    },
+    {
+      "epoch": 12.08,
+      "learning_rate": 1.0348736523769423e-05,
+      "loss": 0.0001,
+      "step": 471
+    },
+    {
+      "epoch": 12.1,
+      "learning_rate": 1.0267910428662878e-05,
+      "loss": 0.0001,
+      "step": 472
+    },
+    {
+      "epoch": 12.13,
+      "learning_rate": 1.0187292057827993e-05,
+      "loss": 0.0001,
+      "step": 473
+    },
+    {
+      "epoch": 12.15,
+      "learning_rate": 1.0106883132004428e-05,
+      "loss": 0.0001,
+      "step": 474
+    },
+    {
+      "epoch": 12.18,
+      "learning_rate": 1.0026685367461364e-05,
+      "loss": 0.0002,
+      "step": 475
+    },
+    {
+      "epoch": 12.21,
+      "learning_rate": 9.946700475960933e-06,
+      "loss": 0.0001,
+      "step": 476
+    },
+    {
+      "epoch": 12.23,
+      "learning_rate": 9.866930164721615e-06,
+      "loss": 0.0001,
+      "step": 477
+    },
+    {
+      "epoch": 12.26,
+      "learning_rate": 9.787376136381866e-06,
+      "loss": 0.0001,
+      "step": 478
+    },
+    {
+      "epoch": 12.28,
+      "learning_rate": 9.70804008896371e-06,
+      "loss": 0.0003,
+      "step": 479
+    },
+    {
+      "epoch": 12.31,
+      "learning_rate": 9.628923715836558e-06,
+      "loss": 0.0,
+      "step": 480
+    },
+    {
+      "epoch": 12.33,
+      "learning_rate": 9.550028705681024e-06,
+      "loss": 0.0,
+      "step": 481
+    },
+    {
+      "epoch": 12.36,
+      "learning_rate": 9.471356742452881e-06,
+      "loss": 0.0,
+      "step": 482
+    },
+    {
+      "epoch": 12.38,
+      "learning_rate": 9.392909505347157e-06,
+      "loss": 0.0005,
+      "step": 483
+    },
+    {
+      "epoch": 12.41,
+      "learning_rate": 9.314688668762232e-06,
+      "loss": 0.0007,
+      "step": 484
+    },
+    {
+      "epoch": 12.44,
+      "learning_rate": 9.23669590226417e-06,
+      "loss": 0.0001,
+      "step": 485
+    },
+    {
+      "epoch": 12.46,
+      "learning_rate": 9.158932870551012e-06,
+      "loss": 0.0001,
+      "step": 486
+    },
+    {
+      "epoch": 12.49,
+      "learning_rate": 9.081401233417315e-06,
+      "loss": 0.0001,
+      "step": 487
+    },
+    {
+      "epoch": 12.51,
+      "learning_rate": 9.004102645718655e-06,
+      "loss": 0.0001,
+      "step": 488
+    },
+    {
+      "epoch": 12.54,
+      "learning_rate": 8.92703875733637e-06,
+      "loss": 0.0001,
+      "step": 489
+    },
+    {
+      "epoch": 12.56,
+      "learning_rate": 8.85021121314229e-06,
+      "loss": 0.0001,
+      "step": 490
+    },
+    {
+      "epoch": 12.59,
+      "learning_rate": 8.773621652963673e-06,
+      "loss": 0.0,
+      "step": 491
+    },
+    {
+      "epoch": 12.62,
+      "learning_rate": 8.697271711548163e-06,
+      "loss": 0.0001,
+      "step": 492
+    },
+    {
+      "epoch": 12.64,
+      "learning_rate": 8.621163018528942e-06,
+      "loss": 0.0001,
+      "step": 493
+    },
+    {
+      "epoch": 12.67,
+      "learning_rate": 8.545297198389896e-06,
+      "loss": 0.0,
+      "step": 494
+    },
+    {
+      "epoch": 12.69,
+      "learning_rate": 8.469675870431e-06,
+      "loss": 0.0001,
+      "step": 495
+    },
+    {
+      "epoch": 12.72,
+      "learning_rate": 8.394300648733688e-06,
+      "loss": 0.0001,
+      "step": 496
+    },
+    {
+      "epoch": 12.74,
+      "learning_rate": 8.319173142126473e-06,
+      "loss": 0.0001,
+      "step": 497
+    },
+    {
+      "epoch": 12.77,
+      "learning_rate": 8.24429495415054e-06,
+      "loss": 0.0,
+      "step": 498
+    },
+    {
+      "epoch": 12.79,
+      "learning_rate": 8.169667683025582e-06,
+      "loss": 0.0001,
+      "step": 499
+    },
+    {
+      "epoch": 12.82,
+      "learning_rate": 8.095292921615628e-06,
+      "loss": 0.0,
+      "step": 500
+    },
+    {
+      "epoch": 12.85,
+      "learning_rate": 8.021172257395092e-06,
+      "loss": 0.0001,
+      "step": 501
+    },
+    {
+      "epoch": 12.87,
+      "learning_rate": 7.947307272414874e-06,
+      "loss": 0.0001,
+      "step": 502
+    },
+    {
+      "epoch": 12.9,
+      "learning_rate": 7.873699543268572e-06,
+      "loss": 0.0001,
+      "step": 503
+    },
+    {
+      "epoch": 12.92,
+      "learning_rate": 7.800350641058867e-06,
+      "loss": 0.0001,
+      "step": 504
+    },
+    {
+      "epoch": 12.95,
+      "learning_rate": 7.727262131363949e-06,
+      "loss": 0.0001,
+      "step": 505
+    },
+    {
+      "epoch": 12.97,
+      "learning_rate": 7.654435574204145e-06,
+      "loss": 0.0001,
+      "step": 506
+    },
+    {
+      "epoch": 13.0,
+      "learning_rate": 7.581872524008574e-06,
+      "loss": 0.0001,
+      "step": 507
+    },
+    {
+      "epoch": 13.03,
+      "learning_rate": 7.509574529582022e-06,
+      "loss": 0.0001,
+      "step": 508
+    },
+    {
+      "epoch": 13.05,
+      "learning_rate": 7.437543134071823e-06,
+      "loss": 0.0001,
+      "step": 509
+    },
+    {
+      "epoch": 13.08,
+      "learning_rate": 7.365779874934987e-06,
+      "loss": 0.0001,
+      "step": 510
+    },
+    {
+      "epoch": 13.1,
+      "learning_rate": 7.294286283905314e-06,
+      "loss": 0.0,
+      "step": 511
+    },
+    {
+      "epoch": 13.13,
+      "learning_rate": 7.223063886960779e-06,
+      "loss": 0.0,
+      "step": 512
+    },
+    {
+      "epoch": 13.15,
+      "learning_rate": 7.152114204290879e-06,
+      "loss": 0.0001,
+      "step": 513
+    },
+    {
+      "epoch": 13.18,
+      "learning_rate": 7.081438750264258e-06,
+      "loss": 0.0,
+      "step": 514
+    },
+    {
+      "epoch": 13.21,
+      "learning_rate": 7.01103903339633e-06,
+      "loss": 0.0,
+      "step": 515
+    },
+    {
+      "epoch": 13.23,
+      "learning_rate": 6.940916556317119e-06,
+      "loss": 0.0001,
+      "step": 516
+    },
+    {
+      "epoch": 13.26,
+      "learning_rate": 6.87107281573915e-06,
+      "loss": 0.0001,
+      "step": 517
+    },
+    {
+      "epoch": 13.28,
+      "learning_rate": 6.801509302425553e-06,
+      "loss": 0.0,
+      "step": 518
+    },
+    {
+      "epoch": 13.31,
+      "learning_rate": 6.732227501158182e-06,
+      "loss": 0.0,
+      "step": 519
+    },
+    {
+      "epoch": 13.33,
+      "learning_rate": 6.6632288907059795e-06,
+      "loss": 0.0003,
+      "step": 520
+    },
+    {
+      "epoch": 13.36,
+      "learning_rate": 6.594514943793369e-06,
+      "loss": 0.0,
+      "step": 521
+    },
+    {
+      "epoch": 13.38,
+      "learning_rate": 6.526087127068857e-06,
+      "loss": 0.0001,
+      "step": 522
+    },
+    {
+      "epoch": 13.41,
+      "learning_rate": 6.457946901073704e-06,
+      "loss": 0.0,
+      "step": 523
+    },
+    {
+      "epoch": 13.44,
+      "learning_rate": 6.3900957202107695e-06,
+      "loss": 0.0,
+      "step": 524
+    },
+    {
+      "epoch": 13.46,
+      "learning_rate": 6.322535032713437e-06,
+      "loss": 0.0,
+      "step": 525
+    },
+    {
+      "epoch": 13.49,
+      "learning_rate": 6.255266280614747e-06,
+      "loss": 0.0001,
+      "step": 526
+    },
+    {
+      "epoch": 13.51,
+      "learning_rate": 6.188290899716569e-06,
+      "loss": 0.0001,
+      "step": 527
+    },
+    {
+      "epoch": 13.54,
+      "learning_rate": 6.1216103195590085e-06,
+      "loss": 0.0001,
+      "step": 528
+    },
+    {
+      "epoch": 13.56,
+      "learning_rate": 6.055225963389841e-06,
+      "loss": 0.0001,
+      "step": 529
+    },
+    {
+      "epoch": 13.59,
+      "learning_rate": 5.989139248134181e-06,
+      "loss": 0.0,
+      "step": 530
+    },
+    {
+      "epoch": 13.62,
+      "learning_rate": 5.923351584364201e-06,
+      "loss": 0.0001,
+      "step": 531
+    },
+    {
+      "epoch": 13.64,
+      "learning_rate": 5.857864376269051e-06,
+      "loss": 0.0,
+      "step": 532
+    },
+    {
+      "epoch": 13.67,
+      "learning_rate": 5.792679021624872e-06,
+      "loss": 0.0001,
+      "step": 533
+    },
+    {
+      "epoch": 13.69,
+      "learning_rate": 5.727796911764955e-06,
+      "loss": 0.0,
+      "step": 534
+    },
+    {
+      "epoch": 13.72,
+      "learning_rate": 5.6632194315500845e-06,
+      "loss": 0.0001,
+      "step": 535
+    },
+    {
+      "epoch": 13.74,
+      "learning_rate": 5.598947959338912e-06,
+      "loss": 0.0001,
+      "step": 536
+    },
+    {
+      "epoch": 13.77,
+      "learning_rate": 5.534983866958608e-06,
+      "loss": 0.0,
+      "step": 537
+    },
+    {
+      "epoch": 13.79,
+      "learning_rate": 5.471328519675521e-06,
+      "loss": 0.0,
+      "step": 538
+    },
+    {
+      "epoch": 13.82,
+      "learning_rate": 5.407983276166084e-06,
+      "loss": 0.0,
+      "step": 539
+    },
+    {
+      "epoch": 13.85,
+      "learning_rate": 5.344949488487776e-06,
+      "loss": 0.0001,
+      "step": 540
+    },
+    {
+      "epoch": 13.87,
+      "learning_rate": 5.2822285020503e-06,
+      "loss": 0.0001,
+      "step": 541
+    },
+    {
+      "epoch": 13.9,
+      "learning_rate": 5.219821655586821e-06,
+      "loss": 0.0,
+      "step": 542
+    },
+    {
+      "epoch": 13.92,
+      "learning_rate": 5.157730281125444e-06,
+      "loss": 0.0,
+      "step": 543
+    },
+    {
+      "epoch": 13.95,
+      "learning_rate": 5.095955703960746e-06,
+      "loss": 0.0001,
+      "step": 544
+    },
+    {
+      "epoch": 13.97,
+      "learning_rate": 5.034499242625504e-06,
+      "loss": 0.0,
+      "step": 545
+    },
+    {
+      "epoch": 14.0,
+      "learning_rate": 4.9733622088625335e-06,
+      "loss": 0.0,
+      "step": 546
+    },
+    {
+      "epoch": 14.03,
+      "learning_rate": 4.912545907596722e-06,
+      "loss": 0.0,
+      "step": 547
+    },
+    {
+      "epoch": 14.05,
+      "learning_rate": 4.852051636907144e-06,
+      "loss": 0.0,
+      "step": 548
+    },
+    {
+      "epoch": 14.08,
+      "learning_rate": 4.791880687999382e-06,
+      "loss": 0.0,
+      "step": 549
+    },
+    {
+      "epoch": 14.1,
+      "learning_rate": 4.732034345177941e-06,
+      "loss": 0.0,
+      "step": 550
+    },
+    {
+      "epoch": 14.13,
+      "learning_rate": 4.672513885818859e-06,
+      "loss": 0.0,
+      "step": 551
+    },
+    {
+      "epoch": 14.15,
+      "learning_rate": 4.613320580342422e-06,
+      "loss": 0.0,
+      "step": 552
+    },
+    {
+      "epoch": 14.18,
+      "learning_rate": 4.55445569218607e-06,
+      "loss": 0.0001,
+      "step": 553
+    },
+    {
+      "epoch": 14.21,
+      "learning_rate": 4.495920477777403e-06,
+      "loss": 0.0001,
+      "step": 554
+    },
+    {
+      "epoch": 14.23,
+      "learning_rate": 4.437716186507397e-06,
+      "loss": 0.0,
+      "step": 555
+    },
+    {
+      "epoch": 14.26,
+      "learning_rate": 4.379844060703693e-06,
+      "loss": 0.0,
+      "step": 556
+    },
+    {
+      "epoch": 14.28,
+      "learning_rate": 4.3223053356041315e-06,
+      "loss": 0.0001,
+      "step": 557
+    },
+    {
+      "epoch": 14.31,
+      "learning_rate": 4.265101239330336e-06,
+      "loss": 0.0,
+      "step": 558
+    },
+    {
+      "epoch": 14.33,
+      "learning_rate": 4.2082329928615494e-06,
+      "loss": 0.0001,
+      "step": 559
+    },
+    {
+      "epoch": 14.36,
+      "learning_rate": 4.151701810008524e-06,
+      "loss": 0.0,
+      "step": 560
+    },
+    {
+      "epoch": 14.38,
+      "learning_rate": 4.095508897387661e-06,
+      "loss": 0.0001,
+      "step": 561
+    },
+    {
+      "epoch": 14.41,
+      "learning_rate": 4.03965545439521e-06,
+      "loss": 0.0001,
+      "step": 562
+    },
+    {
+      "epoch": 14.44,
+      "learning_rate": 3.984142673181717e-06,
+      "loss": 0.0,
+      "step": 563
+    },
+    {
+      "epoch": 14.46,
+      "learning_rate": 3.9289717386265255e-06,
+      "loss": 0.0,
+      "step": 564
+    },
+    {
+      "epoch": 14.49,
+      "learning_rate": 3.8741438283125374e-06,
+      "loss": 0.0001,
+      "step": 565
+    },
+    {
+      "epoch": 14.51,
+      "learning_rate": 3.819660112501053e-06,
+      "loss": 0.0001,
+      "step": 566
+    },
+    {
+      "epoch": 14.54,
+      "learning_rate": 3.765521754106776e-06,
+      "loss": 0.0,
+      "step": 567
+    },
+    {
+      "epoch": 14.56,
+      "learning_rate": 3.711729908673034e-06,
+      "loss": 0.0,
+      "step": 568
+    },
+    {
+      "epoch": 14.59,
+      "learning_rate": 3.6582857243470706e-06,
+      "loss": 0.0,
+      "step": 569
+    },
+    {
+      "epoch": 14.62,
+      "learning_rate": 3.60519034185558e-06,
+      "loss": 0.0,
+      "step": 570
+    },
+    {
+      "epoch": 14.64,
+      "learning_rate": 3.552444894480318e-06,
+      "loss": 0.0,
+      "step": 571
+    },
+    {
+      "epoch": 14.67,
+      "learning_rate": 3.5000505080339565e-06,
+      "loss": 0.0001,
+      "step": 572
+    },
+    {
+      "epoch": 14.69,
+      "learning_rate": 3.448008300836003e-06,
+      "loss": 0.0001,
+      "step": 573
+    },
+    {
+      "epoch": 14.72,
+      "learning_rate": 3.3963193836889907e-06,
+      "loss": 0.0,
+      "step": 574
+    },
+    {
+      "epoch": 14.74,
+      "learning_rate": 3.344984859854703e-06,
+      "loss": 0.0,
+      "step": 575
+    },
+    {
+      "epoch": 14.77,
+      "learning_rate": 3.2940058250306927e-06,
+      "loss": 0.0,
+      "step": 576
+    },
+    {
+      "epoch": 14.79,
+      "learning_rate": 3.2433833673268358e-06,
+      "loss": 0.0,
+      "step": 577
+    },
+    {
+      "epoch": 14.82,
+      "learning_rate": 3.193118567242148e-06,
+      "loss": 0.0,
+      "step": 578
+    },
+    {
+      "epoch": 14.85,
+      "learning_rate": 3.1432124976416988e-06,
+      "loss": 0.0001,
+      "step": 579
+    },
+    {
+      "epoch": 14.87,
+      "learning_rate": 3.093666223733731e-06,
+      "loss": 0.0,
+      "step": 580
+    },
+    {
+      "epoch": 14.87,
+      "eval_test_accuracy": 85.57692307692307,
+      "eval_test_average": 85.24300699300699,
+      "eval_test_loss": 1.0223864316940308,
+      "eval_test_recall": 84.9090909090909,
+      "eval_test_runtime": 785.4649,
+      "eval_test_samples_per_second": 1.351,
+      "eval_test_steps_per_second": 0.006,
+      "step": 580
+    },
+    {
+      "epoch": 14.87,
+      "eval_val_accuracy": 85.07109004739335,
+      "eval_val_average": 84.80103442072331,
+      "eval_val_loss": 1.0219820737838745,
+      "eval_val_recall": 84.53097879405325,
+      "eval_val_runtime": 794.4161,
+      "eval_val_samples_per_second": 1.326,
+      "eval_val_steps_per_second": 0.006,
+      "step": 580
+    },
+    {
+      "epoch": 14.9,
+      "learning_rate": 3.0444808030468966e-06,
+      "loss": 0.0001,
+      "step": 581
+    },
+    {
+      "epoch": 14.92,
+      "learning_rate": 2.9956572854077205e-06,
+      "loss": 0.0,
+      "step": 582
+    },
+    {
+      "epoch": 14.95,
+      "learning_rate": 2.947196712918157e-06,
+      "loss": 0.0,
+      "step": 583
+    },
+    {
+      "epoch": 14.97,
+      "learning_rate": 2.89910011993338e-06,
+      "loss": 0.0001,
+      "step": 584
+    },
+    {
+      "epoch": 15.0,
+      "learning_rate": 2.8513685330396755e-06,
+      "loss": 0.0,
+      "step": 585
+    },
+    {
+      "epoch": 15.03,
+      "learning_rate": 2.804002971032551e-06,
+      "loss": 0.0,
+      "step": 586
+    },
+    {
+      "epoch": 15.05,
+      "learning_rate": 2.7570044448949886e-06,
+      "loss": 0.0,
+      "step": 587
+    },
+    {
+      "epoch": 15.08,
+      "learning_rate": 2.7103739577758426e-06,
+      "loss": 0.0,
+      "step": 588
+    },
+    {
+      "epoch": 15.1,
+      "learning_rate": 2.664112504968468e-06,
+      "loss": 0.0001,
+      "step": 589
+    },
+    {
+      "epoch": 15.13,
+      "learning_rate": 2.618221073889433e-06,
+      "loss": 0.0,
+      "step": 590
+    },
+    {
+      "epoch": 15.15,
+      "learning_rate": 2.5727006440574866e-06,
+      "loss": 0.0001,
+      "step": 591
+    },
+    {
+      "epoch": 15.18,
+      "learning_rate": 2.5275521870726107e-06,
+      "loss": 0.0,
+      "step": 592
+    },
+    {
+      "epoch": 15.21,
+      "learning_rate": 2.4827766665953147e-06,
+      "loss": 0.0,
+      "step": 593
+    },
+    {
+      "epoch": 15.23,
+      "learning_rate": 2.4383750383260417e-06,
+      "loss": 0.0,
+      "step": 594
+    },
+    {
+      "epoch": 15.26,
+      "learning_rate": 2.394348249984797e-06,
+      "loss": 0.0,
+      "step": 595
+    },
+    {
+      "epoch": 15.28,
+      "learning_rate": 2.3506972412908866e-06,
+      "loss": 0.0,
+      "step": 596
+    },
+    {
+      "epoch": 15.31,
+      "learning_rate": 2.3074229439428964e-06,
+      "loss": 0.0001,
+      "step": 597
+    },
+    {
+      "epoch": 15.33,
+      "learning_rate": 2.264526281598762e-06,
+      "loss": 0.0,
+      "step": 598
+    },
+    {
+      "epoch": 15.36,
+      "learning_rate": 2.2220081698561067e-06,
+      "loss": 0.0001,
+      "step": 599
+    },
+    {
+      "epoch": 15.38,
+      "learning_rate": 2.1798695162326444e-06,
+      "loss": 0.0,
+      "step": 600
+    },
+    {
+      "epoch": 15.41,
+      "learning_rate": 2.138111220146857e-06,
+      "loss": 0.0,
+      "step": 601
+    },
+    {
+      "epoch": 15.44,
+      "learning_rate": 2.0967341728987554e-06,
+      "loss": 0.0,
+      "step": 602
+    },
+    {
+      "epoch": 15.46,
+      "learning_rate": 2.055739257650895e-06,
+      "loss": 0.0,
+      "step": 603
+    },
+    {
+      "epoch": 15.49,
+      "learning_rate": 2.015127349409489e-06,
+      "loss": 0.0001,
+      "step": 604
+    },
+    {
+      "epoch": 15.51,
+      "learning_rate": 1.974899315005763e-06,
+      "loss": 0.0001,
+      "step": 605
+    },
+    {
+      "epoch": 15.54,
+      "learning_rate": 1.9350560130774234e-06,
+      "loss": 0.0,
+      "step": 606
+    },
+    {
+      "epoch": 15.56,
+      "learning_rate": 1.895598294050358e-06,
+      "loss": 0.0001,
+      "step": 607
+    },
+    {
+      "epoch": 15.59,
+      "learning_rate": 1.8565270001204693e-06,
+      "loss": 0.0,
+      "step": 608
+    },
+    {
+      "epoch": 15.62,
+      "learning_rate": 1.817842965235701e-06,
+      "loss": 0.0001,
+      "step": 609
+    },
+    {
+      "epoch": 15.64,
+      "learning_rate": 1.7795470150782312e-06,
+      "loss": 0.0,
+      "step": 610
+    },
+    {
+      "epoch": 15.67,
+      "learning_rate": 1.7416399670468687e-06,
+      "loss": 0.0001,
+      "step": 611
+    },
+    {
+      "epoch": 15.69,
+      "learning_rate": 1.7041226302395797e-06,
+      "loss": 0.0,
+      "step": 612
+    },
+    {
+      "epoch": 15.72,
+      "learning_rate": 1.6669958054362444e-06,
+      "loss": 0.0,
+      "step": 613
+    },
+    {
+      "epoch": 15.74,
+      "learning_rate": 1.6302602850815397e-06,
+      "loss": 0.0,
+      "step": 614
+    },
+    {
+      "epoch": 15.77,
+      "learning_rate": 1.5939168532680516e-06,
+      "loss": 0.0,
+      "step": 615
+    },
+    {
+      "epoch": 15.79,
+      "learning_rate": 1.55796628571951e-06,
+      "loss": 0.0001,
+      "step": 616
+    },
+    {
+      "epoch": 15.82,
+      "learning_rate": 1.5224093497742654e-06,
+      "loss": 0.0,
+      "step": 617
+    },
+    {
+      "epoch": 15.85,
+      "learning_rate": 1.487246804368876e-06,
+      "loss": 0.0,
+      "step": 618
+    },
+    {
+      "epoch": 15.87,
+      "learning_rate": 1.452479400021929e-06,
+      "loss": 0.0001,
+      "step": 619
+    },
+    {
+      "epoch": 15.9,
+      "learning_rate": 1.418107878818027e-06,
+      "loss": 0.0,
+      "step": 620
+    },
+    {
+      "epoch": 15.92,
+      "learning_rate": 1.3841329743919206e-06,
+      "loss": 0.0,
+      "step": 621
+    },
+    {
+      "epoch": 15.95,
+      "learning_rate": 1.3505554119128861e-06,
+      "loss": 0.0,
+      "step": 622
+    },
+    {
+      "epoch": 15.97,
+      "learning_rate": 1.3173759080692184e-06,
+      "loss": 0.0,
+      "step": 623
+    },
+    {
+      "epoch": 16.0,
+      "learning_rate": 1.2845951710529513e-06,
+      "loss": 0.0,
+      "step": 624
+    },
+    {
+      "epoch": 16.03,
+      "learning_rate": 1.2522139005447232e-06,
+      "loss": 0.0,
+      "step": 625
+    },
+    {
+      "epoch": 16.05,
+      "learning_rate": 1.2202327876988719e-06,
+      "loss": 0.0,
+      "step": 626
+    },
+    {
+      "epoch": 16.08,
+      "learning_rate": 1.1886525151286477e-06,
+      "loss": 0.0,
+      "step": 627
+    },
+    {
+      "epoch": 16.1,
+      "learning_rate": 1.157473756891674e-06,
+      "loss": 0.0001,
+      "step": 628
+    },
+    {
+      "epoch": 16.13,
+      "learning_rate": 1.1266971784755398e-06,
+      "loss": 0.0,
+      "step": 629
+    },
+    {
+      "epoch": 16.15,
+      "learning_rate": 1.0963234367836106e-06,
+      "loss": 0.0001,
+      "step": 630
+    },
+    {
+      "epoch": 16.18,
+      "learning_rate": 1.0663531801209826e-06,
+      "loss": 0.0,
+      "step": 631
+    },
+    {
+      "epoch": 16.21,
+      "learning_rate": 1.036787048180683e-06,
+      "loss": 0.0,
+      "step": 632
+    },
+    {
+      "epoch": 16.23,
+      "learning_rate": 1.007625672029977e-06,
+      "loss": 0.0,
+      "step": 633
+    },
+    {
+      "epoch": 16.26,
+      "learning_rate": 9.788696740969295e-07,
+      "loss": 0.0,
+      "step": 634
+    },
+    {
+      "epoch": 16.28,
+      "learning_rate": 9.50519668157095e-07,
+      "loss": 0.0001,
+      "step": 635
+    },
+    {
+      "epoch": 16.31,
+      "learning_rate": 9.225762593204379e-07,
+      "loss": 0.0,
+      "step": 636
+    },
+    {
+      "epoch": 16.33,
+      "learning_rate": 8.950400440184004e-07,
+      "loss": 0.0,
+      "step": 637
+    },
+    {
+      "epoch": 16.36,
+      "learning_rate": 8.679116099911855e-07,
+      "loss": 0.0,
+      "step": 638
+    },
+    {
+      "epoch": 16.38,
+      "learning_rate": 8.411915362751988e-07,
+      "loss": 0.0,
+      "step": 639
+    },
+    {
+      "epoch": 16.41,
+      "learning_rate": 8.148803931907023e-07,
+      "loss": 0.0,
+      "step": 640
+    },
+    {
+      "epoch": 16.44,
+      "learning_rate": 7.889787423296335e-07,
+      "loss": 0.0,
+      "step": 641
+    },
+    {
+      "epoch": 16.46,
+      "learning_rate": 7.634871365436192e-07,
+      "loss": 0.0001,
+      "step": 642
+    },
+    {
+      "epoch": 16.49,
+      "learning_rate": 7.384061199321757e-07,
+      "loss": 0.0,
+      "step": 643
+    },
+    {
+      "epoch": 16.51,
+      "learning_rate": 7.137362278311033e-07,
+      "loss": 0.0,
+      "step": 644
+    },
+    {
+      "epoch": 16.54,
+      "learning_rate": 6.894779868010415e-07,
+      "loss": 0.0001,
+      "step": 645
+    },
+    {
+      "epoch": 16.56,
+      "learning_rate": 6.656319146162516e-07,
+      "loss": 0.0,
+      "step": 646
+    },
+    {
+      "epoch": 16.59,
+      "learning_rate": 6.421985202535497e-07,
+      "loss": 0.0001,
+      "step": 647
+    },
+    {
+      "epoch": 16.62,
+      "learning_rate": 6.191783038814492e-07,
+      "loss": 0.0,
+      "step": 648
+    },
+    {
+      "epoch": 16.64,
+      "learning_rate": 5.965717568494844e-07,
+      "loss": 0.0,
+      "step": 649
+    },
+    {
+      "epoch": 16.67,
+      "learning_rate": 5.743793616777216e-07,
+      "loss": 0.0,
+      "step": 650
+    },
+    {
+      "epoch": 16.69,
+      "learning_rate": 5.526015920464689e-07,
+      "loss": 0.0,
+      "step": 651
+    },
+    {
+      "epoch": 16.72,
+      "learning_rate": 5.312389127861428e-07,
+      "loss": 0.0,
+      "step": 652
+    },
+    {
+      "epoch": 16.74,
+      "learning_rate": 5.102917798673779e-07,
+      "loss": 0.0001,
+      "step": 653
+    },
+    {
+      "epoch": 16.77,
+      "learning_rate": 4.89760640391268e-07,
+      "loss": 0.0,
+      "step": 654
+    },
+    {
+      "epoch": 16.79,
+      "learning_rate": 4.696459325798386e-07,
+      "loss": 0.0,
+      "step": 655
+    },
+    {
+      "epoch": 16.82,
+      "learning_rate": 4.499480857666849e-07,
+      "loss": 0.0001,
+      "step": 656
+    },
+    {
+      "epoch": 16.85,
+      "learning_rate": 4.306675203878219e-07,
+      "loss": 0.0,
+      "step": 657
+    },
+    {
+      "epoch": 16.87,
+      "learning_rate": 4.118046479726823e-07,
+      "loss": 0.0,
+      "step": 658
+    },
+    {
+      "epoch": 16.9,
+      "learning_rate": 3.9335987113537035e-07,
+      "loss": 0.0,
+      "step": 659
+    },
+    {
+      "epoch": 16.92,
+      "learning_rate": 3.75333583566031e-07,
+      "loss": 0.0,
+      "step": 660
+    },
+    {
+      "epoch": 16.95,
+      "learning_rate": 3.57726170022481e-07,
+      "loss": 0.0,
+      "step": 661
+    },
+    {
+      "epoch": 16.97,
+      "learning_rate": 3.4053800632196434e-07,
+      "loss": 0.0001,
+      "step": 662
+    },
+    {
+      "epoch": 17.0,
+      "learning_rate": 3.2376945933316086e-07,
+      "loss": 0.0,
+      "step": 663
+    },
+    {
+      "epoch": 17.03,
+      "learning_rate": 3.074208869683282e-07,
+      "loss": 0.0,
+      "step": 664
+    },
+    {
+      "epoch": 17.05,
+      "learning_rate": 2.914926381756855e-07,
+      "loss": 0.0,
+      "step": 665
+    },
+    {
+      "epoch": 17.08,
+      "learning_rate": 2.7598505293194855e-07,
+      "loss": 0.0,
+      "step": 666
+    },
+    {
+      "epoch": 17.1,
+      "learning_rate": 2.6089846223508853e-07,
+      "loss": 0.0,
+      "step": 667
+    },
+    {
+      "epoch": 17.13,
+      "learning_rate": 2.462331880972468e-07,
+      "loss": 0.0,
+      "step": 668
+    },
+    {
+      "epoch": 17.15,
+      "learning_rate": 2.3198954353788271e-07,
+      "loss": 0.0,
+      "step": 669
+    },
+    {
+      "epoch": 17.18,
+      "learning_rate": 2.1816783257708084e-07,
+      "loss": 0.0,
+      "step": 670
+    },
+    {
+      "epoch": 17.21,
+      "learning_rate": 2.0476835022906317e-07,
+      "loss": 0.0,
+      "step": 671
+    },
+    {
+      "epoch": 17.23,
+      "learning_rate": 1.9179138249589836e-07,
+      "loss": 0.0,
+      "step": 672
+    },
+    {
+      "epoch": 17.26,
+      "learning_rate": 1.792372063613823e-07,
+      "loss": 0.0,
+      "step": 673
+    },
+    {
+      "epoch": 17.28,
+      "learning_rate": 1.6710608978514509e-07,
+      "loss": 0.0,
+      "step": 674
+    },
+    {
+      "epoch": 17.31,
+      "learning_rate": 1.5539829169691989e-07,
+      "loss": 0.0001,
+      "step": 675
+    },
+    {
+      "epoch": 17.33,
+      "learning_rate": 1.4411406199102084e-07,
+      "loss": 0.0001,
+      "step": 676
+    },
+    {
+      "epoch": 17.36,
+      "learning_rate": 1.3325364152100063e-07,
+      "loss": 0.0,
+      "step": 677
+    },
+    {
+      "epoch": 17.38,
+      "learning_rate": 1.2281726209452782e-07,
+      "loss": 0.0001,
+      "step": 678
+    },
+    {
+      "epoch": 17.41,
+      "learning_rate": 1.1280514646841545e-07,
+      "loss": 0.0,
+      "step": 679
+    },
+    {
+      "epoch": 17.44,
+      "learning_rate": 1.0321750834388911e-07,
+      "loss": 0.0,
+      "step": 680
+    },
+    {
+      "epoch": 17.46,
+      "learning_rate": 9.405455236200844e-08,
+      "loss": 0.0,
+      "step": 681
+    },
+    {
+      "epoch": 17.49,
+      "learning_rate": 8.531647409931065e-08,
+      "loss": 0.0,
+      "step": 682
+    },
+    {
+      "epoch": 17.51,
+      "learning_rate": 7.700346006362714e-08,
+      "loss": 0.0,
+      "step": 683
+    },
+    {
+      "epoch": 17.54,
+      "learning_rate": 6.91156876901089e-08,
+      "loss": 0.0,
+      "step": 684
+    },
+    {
+      "epoch": 17.56,
+      "learning_rate": 6.165332533744072e-08,
+      "loss": 0.0,
+      "step": 685
+    },
+    {
+      "epoch": 17.59,
+      "learning_rate": 5.4616532284239576e-08,
+      "loss": 0.0001,
+      "step": 686
+    },
+    {
+      "epoch": 17.62,
+      "learning_rate": 4.800545872566176e-08,
+      "loss": 0.0001,
+      "step": 687
+    },
+    {
+      "epoch": 17.64,
+      "learning_rate": 4.182024577019439e-08,
+      "loss": 0.0,
+      "step": 688
+    },
+    {
+      "epoch": 17.67,
+      "learning_rate": 3.606102543664447e-08,
+      "loss": 0.0,
+      "step": 689
+    },
+    {
+      "epoch": 17.69,
+      "learning_rate": 3.072792065132113e-08,
+      "loss": 0.0,
+      "step": 690
+    },
+    {
+      "epoch": 17.72,
+      "learning_rate": 2.582104524541107e-08,
+      "loss": 0.0,
+      "step": 691
+    },
+    {
+      "epoch": 17.74,
+      "learning_rate": 2.1340503952551606e-08,
+      "loss": 0.0,
+      "step": 692
+    },
+    {
+      "epoch": 17.77,
+      "learning_rate": 1.7286392406588026e-08,
+      "loss": 0.0001,
+      "step": 693
+    },
+    {
+      "epoch": 17.79,
+      "learning_rate": 1.365879713954188e-08,
+      "loss": 0.0,
+      "step": 694
+    },
+    {
+      "epoch": 17.82,
+      "learning_rate": 1.0457795579756901e-08,
+      "loss": 0.0,
+      "step": 695
+    },
+    {
+      "epoch": 17.85,
+      "learning_rate": 7.683456050251447e-09,
+      "loss": 0.0,
+      "step": 696
+    },
+    {
+      "epoch": 17.85,
+      "eval_test_accuracy": 85.57692307692307,
+      "eval_test_average": 85.16548761316201,
+      "eval_test_loss": 1.0252264738082886,
+      "eval_test_recall": 84.75405214940098,
+      "eval_test_runtime": 782.1434,
+      "eval_test_samples_per_second": 1.357,
+      "eval_test_steps_per_second": 0.006,
+      "step": 696
+    },
+    {
+      "epoch": 17.85,
+      "eval_val_accuracy": 85.07109004739335,
+      "eval_val_average": 84.84065407207038,
+      "eval_val_loss": 1.024724006652832,
+      "eval_val_recall": 84.6102180967474,
+      "eval_val_runtime": 782.6042,
+      "eval_val_samples_per_second": 1.346,
+      "eval_val_steps_per_second": 0.006,
+      "step": 696
+    },
+    {
+      "epoch": 17.87,
+      "learning_rate": 5.335837767255214e-09,
+      "loss": 0.0,
+      "step": 697
+    },
+    {
+      "epoch": 17.9,
+      "learning_rate": 3.414990838945809e-09,
+      "loss": 0.0,
+      "step": 698
+    },
+    {
+      "epoch": 17.92,
+      "learning_rate": 1.920956264387375e-09,
+      "loss": 0.0,
+      "step": 699
+    },
+    {
+      "epoch": 17.95,
+      "learning_rate": 8.537659326424141e-10,
+      "loss": 0.0,
+      "step": 700
+    },
+    {
+      "epoch": 17.97,
+      "learning_rate": 2.1344262210565293e-10,
+      "loss": 0.0,
+      "step": 701
+    },
+    {
+      "epoch": 18.0,
+      "learning_rate": 0.0,
+      "loss": 0.0,
+      "step": 702
+    },
+    {
+      "epoch": 18.0,
+      "step": 702,
+      "total_flos": 6.135775686914212e+17,
+      "train_loss": 0.034780138895060625,
+      "train_runtime": 14932.2475,
+      "train_samples_per_second": 5.93,
+      "train_steps_per_second": 0.047
+    }
+  ],
+  "max_steps": 702,
+  "num_train_epochs": 18,
+  "total_flos": 6.135775686914212e+17,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:eb9cdc8f80afab5311b3222468462bc9782415f64fdf14cf1c41ffd61c871b7f
+size 4600