Upload folder using huggingface_hub

added_tokens.json

@@ -0,0 +1,7 @@
+{
+  "<<LABEL>>": 151646,
+  "<<SEP>>": 151647,
+  "<|endoftext|>": 151643,
+  "<|im_end|>": 151645,
+  "<|im_start|>": 151644
+}

config.json

@@ -0,0 +1,113 @@
+{
+  "_name_or_path": "models/qwen/large",
+  "architecture_type": "uni-encoder",
+  "architectures": [
+    "GLiClassModel"
+  ],
+  "class_token_index": 151646,
+  "contrastive_loss_coef": 0.03,
+  "embed_class_token": true,
+  "encoder_config": {
+    "_name_or_path": "knowledgator/Qwen-encoder-1.5B",
+    "add_cross_attention": false,
+    "architectures": [
+      "Qwen2ForCausalLM"
+    ],
+    "attention_dropout": 0.0,
+    "bad_words_ids": null,
+    "begin_suppress_tokens": null,
+    "bos_token_id": 151643,
+    "chunk_size_feed_forward": 0,
+    "cross_attention_hidden_size": null,
+    "decoder_start_token_id": null,
+    "diversity_penalty": 0.0,
+    "do_sample": false,
+    "early_stopping": false,
+    "encoder_no_repeat_ngram_size": 0,
+    "eos_token_id": 151645,
+    "exponential_decay_length_penalty": null,
+    "finetuning_task": null,
+    "forced_bos_token_id": null,
+    "forced_eos_token_id": null,
+    "hidden_act": "silu",
+    "hidden_size": 1536,
+    "id2label": {
+      "0": "LABEL_0",
+      "1": "LABEL_1"
+    },
+    "initializer_range": 0.02,
+    "intermediate_size": 8960,
+    "is_decoder": false,
+    "is_encoder_decoder": false,
+    "label2id": {
+      "LABEL_0": 0,
+      "LABEL_1": 1
+    },
+    "length_penalty": 1.0,
+    "max_length": 20,
+    "max_position_embeddings": 32768,
+    "max_window_layers": 28,
+    "min_length": 0,
+    "model_type": "qwen2",
+    "no_repeat_ngram_size": 0,
+    "num_attention_heads": 12,
+    "num_beam_groups": 1,
+    "num_beams": 1,
+    "num_hidden_layers": 28,
+    "num_key_value_heads": 2,
+    "num_return_sequences": 1,
+    "output_attentions": false,
+    "output_hidden_states": false,
+    "output_scores": false,
+    "pad_token_id": null,
+    "prefix": null,
+    "problem_type": null,
+    "pruned_heads": {},
+    "remove_invalid_values": false,
+    "repetition_penalty": 1.0,
+    "return_dict": true,
+    "return_dict_in_generate": false,
+    "rms_norm_eps": 1e-06,
+    "rope_theta": 1000000.0,
+    "sep_token_id": null,
+    "sliding_window": null,
+    "suppress_tokens": null,
+    "task_specific_params": null,
+    "temperature": 1.0,
+    "tf_legacy_loss": false,
+    "tie_encoder_decoder": false,
+    "tie_word_embeddings": true,
+    "tokenizer_class": null,
+    "top_k": 50,
+    "top_p": 1.0,
+    "torch_dtype": "float32",
+    "torchscript": false,
+    "typical_p": 1.0,
+    "use_bfloat16": false,
+    "use_cache": true,
+    "use_sliding_window": false,
+    "vocab_size": 151648
+  },
+  "encoder_model_name": "knowledgator/Qwen-encoder-1.5B",
+  "extract_text_features": true,
+  "focal_loss_alpha": -1,
+  "focal_loss_gamma": 2,
+  "hidden_size": 1536,
+  "ignore_index": -100,
+  "initializer_range": 0.03,
+  "logit_scale_init_value": 2.6592,
+  "max_num_classes": 25,
+  "model_type": "GLiClass",
+  "normalize_features": false,
+  "pooling_strategy": "avg",
+  "problem_type": "multi_label_classification",
+  "projector_hidden_act": "gelu",
+  "prompt_first": true,
+  "scorer_type": "simple",
+  "squeeze_layers": false,
+  "text_token_index": 151647,
+  "torch_dtype": "float32",
+  "transformers_version": "4.44.1",
+  "use_lstm": false,
+  "vocab_size": 151648
+}

model-00001-of-00002.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6c23c940fe49a315bb95dd9d8be1076265dd9acc16a76701050d04556748c9e3
+size 4977628572

model-00002-of-00002.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4fd13583ea7c6ead1aceb37ebf076202caa73f39a5f31728a1b3b53a955329b1
+size 1233276416

model.safetensors.index.json

@@ -0,0 +1,354 @@
+{
+  "metadata": {
+    "total_size": 6210861060
+  },
+  "weight_map": {
+    "model.classes_projector.linear_1.bias": "model-00001-of-00002.safetensors",
+    "model.classes_projector.linear_1.weight": "model-00001-of-00002.safetensors",
+    "model.classes_projector.linear_2.bias": "model-00001-of-00002.safetensors",
+    "model.classes_projector.linear_2.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.embed_tokens.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.0.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.0.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.0.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.0.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.0.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.0.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.0.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.0.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.0.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.0.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.0.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.0.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.1.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.1.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.1.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.1.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.1.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.1.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.1.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.1.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.1.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.1.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.1.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.1.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.10.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.10.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.10.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.10.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.10.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.10.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.10.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.10.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.10.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.10.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.10.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.10.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.11.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.11.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.11.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.11.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.11.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.11.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.11.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.11.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.11.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.11.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.11.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.11.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.12.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.12.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.12.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.12.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.12.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.12.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.12.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.12.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.12.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.12.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.12.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.12.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.13.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.13.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.13.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.13.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.13.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.13.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.13.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.13.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.13.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.13.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.13.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.13.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.14.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.14.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.14.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.14.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.14.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.14.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.14.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.14.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.14.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.14.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.14.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.14.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.15.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.15.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.15.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.15.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.15.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.15.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.15.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.15.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.15.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.15.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.15.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.15.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.16.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.16.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.16.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.16.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.16.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.16.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.16.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.16.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.16.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.16.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.16.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.16.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.17.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.17.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.17.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.17.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.17.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.17.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.17.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.17.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.17.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.17.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.17.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.17.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.18.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.18.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.18.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.18.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.18.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.18.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.18.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.18.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.18.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.18.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.18.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.18.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.19.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.19.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.19.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.19.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.19.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.19.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.19.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.19.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.19.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.19.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.19.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.19.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.2.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.2.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.2.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.2.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.2.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.2.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.2.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.2.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.2.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.2.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.2.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.2.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.20.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.20.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.20.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.20.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.20.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.20.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.20.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.20.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.20.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.20.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.20.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.20.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.21.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.encoder_model.layers.21.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.encoder_model.layers.21.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.21.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.encoder_model.layers.21.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.encoder_model.layers.21.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.21.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.21.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.21.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.21.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.21.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.21.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.22.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.encoder_model.layers.22.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.encoder_model.layers.22.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+    "model.encoder_model.layers.22.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.encoder_model.layers.22.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.encoder_model.layers.22.self_attn.k_proj.bias": "model-00002-of-00002.safetensors",
+    "model.encoder_model.layers.22.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.encoder_model.layers.22.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.encoder_model.layers.22.self_attn.q_proj.bias": "model-00002-of-00002.safetensors",
+    "model.encoder_model.layers.22.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.encoder_model.layers.22.self_attn.v_proj.bias": "model-00002-of-00002.safetensors",
+    "model.encoder_model.layers.22.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.encoder_model.layers.23.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.encoder_model.layers.23.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.encoder_model.layers.23.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+    "model.encoder_model.layers.23.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.encoder_model.layers.23.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.encoder_model.layers.23.self_attn.k_proj.bias": "model-00002-of-00002.safetensors",
+    "model.encoder_model.layers.23.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.encoder_model.layers.23.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.encoder_model.layers.23.self_attn.q_proj.bias": "model-00002-of-00002.safetensors",
+    "model.encoder_model.layers.23.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.encoder_model.layers.23.self_attn.v_proj.bias": "model-00002-of-00002.safetensors",
+    "model.encoder_model.layers.23.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.encoder_model.layers.24.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.encoder_model.layers.24.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.encoder_model.layers.24.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+    "model.encoder_model.layers.24.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.encoder_model.layers.24.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.encoder_model.layers.24.self_attn.k_proj.bias": "model-00002-of-00002.safetensors",
+    "model.encoder_model.layers.24.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.encoder_model.layers.24.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.encoder_model.layers.24.self_attn.q_proj.bias": "model-00002-of-00002.safetensors",
+    "model.encoder_model.layers.24.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.encoder_model.layers.24.self_attn.v_proj.bias": "model-00002-of-00002.safetensors",
+    "model.encoder_model.layers.24.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.encoder_model.layers.25.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.encoder_model.layers.25.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.encoder_model.layers.25.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+    "model.encoder_model.layers.25.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.encoder_model.layers.25.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.encoder_model.layers.25.self_attn.k_proj.bias": "model-00002-of-00002.safetensors",
+    "model.encoder_model.layers.25.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.encoder_model.layers.25.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.encoder_model.layers.25.self_attn.q_proj.bias": "model-00002-of-00002.safetensors",
+    "model.encoder_model.layers.25.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.encoder_model.layers.25.self_attn.v_proj.bias": "model-00002-of-00002.safetensors",
+    "model.encoder_model.layers.25.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.encoder_model.layers.26.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.encoder_model.layers.26.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.encoder_model.layers.26.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+    "model.encoder_model.layers.26.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.encoder_model.layers.26.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.encoder_model.layers.26.self_attn.k_proj.bias": "model-00002-of-00002.safetensors",
+    "model.encoder_model.layers.26.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.encoder_model.layers.26.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.encoder_model.layers.26.self_attn.q_proj.bias": "model-00002-of-00002.safetensors",
+    "model.encoder_model.layers.26.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.encoder_model.layers.26.self_attn.v_proj.bias": "model-00002-of-00002.safetensors",
+    "model.encoder_model.layers.26.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.encoder_model.layers.27.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.encoder_model.layers.27.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.encoder_model.layers.27.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+    "model.encoder_model.layers.27.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.encoder_model.layers.27.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.encoder_model.layers.27.self_attn.k_proj.bias": "model-00002-of-00002.safetensors",
+    "model.encoder_model.layers.27.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.encoder_model.layers.27.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.encoder_model.layers.27.self_attn.q_proj.bias": "model-00002-of-00002.safetensors",
+    "model.encoder_model.layers.27.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.encoder_model.layers.27.self_attn.v_proj.bias": "model-00002-of-00002.safetensors",
+    "model.encoder_model.layers.27.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.encoder_model.layers.3.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.3.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.3.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.3.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.3.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.3.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.3.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.3.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.3.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.3.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.3.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.3.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.4.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.4.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.4.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.4.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.4.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.4.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.4.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.4.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.4.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.4.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.4.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.4.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.5.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.5.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.5.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.5.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.5.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.5.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.5.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.5.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.5.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.5.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.5.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.5.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.6.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.6.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.6.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.6.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.6.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.6.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.6.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.6.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.6.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.6.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.6.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.6.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.7.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.7.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.7.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.7.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.7.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.7.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.7.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.7.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.7.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.7.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.7.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.7.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.8.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.8.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.8.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.8.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.8.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.8.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.8.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.8.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.8.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.8.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.8.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.8.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.9.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.9.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.9.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.9.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.9.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.9.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.9.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.9.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.9.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.9.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.9.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.encoder_model.layers.9.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.encoder_model.norm.weight": "model-00002-of-00002.safetensors",
+    "model.logit_scale": "model-00001-of-00002.safetensors",
+    "model.text_projector.linear_1.bias": "model-00001-of-00002.safetensors",
+    "model.text_projector.linear_1.weight": "model-00001-of-00002.safetensors",
+    "model.text_projector.linear_2.bias": "model-00001-of-00002.safetensors",
+    "model.text_projector.linear_2.weight": "model-00001-of-00002.safetensors"
+  }
+}

special_tokens_map.json

@@ -0,0 +1,27 @@
+{
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>"
+  ],
+  "eos_token": {
+    "content": "<|im_end|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "mask_token": {
+    "content": "_",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer_config.json

@@ -0,0 +1,68 @@
+{
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "62": {
+      "content": "_",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151643": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151644": {
+      "content": "<|im_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151645": {
+      "content": "<|im_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151646": {
+      "content": "<<LABEL>>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151647": {
+      "content": "<<SEP>>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>"
+  ],
+  "bos_token": null,
+  "chat_template": "{% for message in messages %}{% if loop.first and messages[0]['role'] != 'system' %}{{ '<|im_start|>system\nYou are a helpful assistant.<|im_end|>\n' }}{% endif %}{{'<|im_start|>' + message['role'] + '\n' + message['content'] + '<|im_end|>' + '\n'}}{% endfor %}{% if add_generation_prompt %}{{ '<|im_start|>assistant\n' }}{% endif %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|im_end|>",
+  "errors": "replace",
+  "mask_token": "_",
+  "model_max_length": 32768,
+  "pad_token": "<|endoftext|>",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

trainer_state.json

@@ -0,0 +1,2145 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 1.7980221756068326,
+  "eval_steps": 500,
+  "global_step": 30000,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.005993407252022775,
+      "grad_norm": 1.0038542747497559,
+      "learning_rate": 2.996703626011388e-07,
+      "loss": 0.1541,
+      "step": 100
+    },
+    {
+      "epoch": 0.01198681450404555,
+      "grad_norm": 0.5848270058631897,
+      "learning_rate": 5.993407252022776e-07,
+      "loss": 0.1208,
+      "step": 200
+    },
+    {
+      "epoch": 0.017980221756068324,
+      "grad_norm": 0.9016316533088684,
+      "learning_rate": 8.990110878034162e-07,
+      "loss": 0.1106,
+      "step": 300
+    },
+    {
+      "epoch": 0.0239736290080911,
+      "grad_norm": 0.21770495176315308,
+      "learning_rate": 1.1986814504045551e-06,
+      "loss": 0.1085,
+      "step": 400
+    },
+    {
+      "epoch": 0.029967036260113874,
+      "grad_norm": 0.1711203008890152,
+      "learning_rate": 1.4983518130056939e-06,
+      "loss": 0.1064,
+      "step": 500
+    },
+    {
+      "epoch": 0.03596044351213665,
+      "grad_norm": 0.22904595732688904,
+      "learning_rate": 1.7980221756068325e-06,
+      "loss": 0.1033,
+      "step": 600
+    },
+    {
+      "epoch": 0.041953850764159424,
+      "grad_norm": 0.060541968792676926,
+      "learning_rate": 2.0976925382079712e-06,
+      "loss": 0.1036,
+      "step": 700
+    },
+    {
+      "epoch": 0.0479472580161822,
+      "grad_norm": 1.8464969396591187,
+      "learning_rate": 2.3973629008091102e-06,
+      "loss": 0.1024,
+      "step": 800
+    },
+    {
+      "epoch": 0.05394066526820498,
+      "grad_norm": 0.9996387362480164,
+      "learning_rate": 2.6970332634102492e-06,
+      "loss": 0.103,
+      "step": 900
+    },
+    {
+      "epoch": 0.05993407252022775,
+      "grad_norm": 0.11365871131420135,
+      "learning_rate": 2.9967036260113878e-06,
+      "loss": 0.1025,
+      "step": 1000
+    },
+    {
+      "epoch": 0.06592747977225052,
+      "grad_norm": 0.34113025665283203,
+      "learning_rate": 3.2963739886125264e-06,
+      "loss": 0.1028,
+      "step": 1100
+    },
+    {
+      "epoch": 0.0719208870242733,
+      "grad_norm": 0.21321642398834229,
+      "learning_rate": 3.596044351213665e-06,
+      "loss": 0.1014,
+      "step": 1200
+    },
+    {
+      "epoch": 0.07791429427629608,
+      "grad_norm": 0.6457433104515076,
+      "learning_rate": 3.8957147138148035e-06,
+      "loss": 0.1011,
+      "step": 1300
+    },
+    {
+      "epoch": 0.08390770152831885,
+      "grad_norm": 0.7026138305664062,
+      "learning_rate": 4.1953850764159425e-06,
+      "loss": 0.1011,
+      "step": 1400
+    },
+    {
+      "epoch": 0.08990110878034162,
+      "grad_norm": 0.22752709686756134,
+      "learning_rate": 4.4950554390170815e-06,
+      "loss": 0.101,
+      "step": 1500
+    },
+    {
+      "epoch": 0.0958945160323644,
+      "grad_norm": 0.16539303958415985,
+      "learning_rate": 4.7947258016182205e-06,
+      "loss": 0.1008,
+      "step": 1600
+    },
+    {
+      "epoch": 0.10188792328438717,
+      "grad_norm": 0.10093910992145538,
+      "learning_rate": 5.0943961642193594e-06,
+      "loss": 0.1005,
+      "step": 1700
+    },
+    {
+      "epoch": 0.10788133053640996,
+      "grad_norm": 0.06153643876314163,
+      "learning_rate": 5.3940665268204984e-06,
+      "loss": 0.1011,
+      "step": 1800
+    },
+    {
+      "epoch": 0.11387473778843273,
+      "grad_norm": 0.5297218561172485,
+      "learning_rate": 5.6937368894216374e-06,
+      "loss": 0.1001,
+      "step": 1900
+    },
+    {
+      "epoch": 0.1198681450404555,
+      "grad_norm": 0.710209310054779,
+      "learning_rate": 5.9934072520227756e-06,
+      "loss": 0.1004,
+      "step": 2000
+    },
+    {
+      "epoch": 0.12586155229247828,
+      "grad_norm": 0.043333955109119415,
+      "learning_rate": 6.2930776146239146e-06,
+      "loss": 0.1011,
+      "step": 2100
+    },
+    {
+      "epoch": 0.13185495954450105,
+      "grad_norm": 0.2191428393125534,
+      "learning_rate": 6.592747977225053e-06,
+      "loss": 0.1002,
+      "step": 2200
+    },
+    {
+      "epoch": 0.13784836679652382,
+      "grad_norm": 0.24634818732738495,
+      "learning_rate": 6.892418339826191e-06,
+      "loss": 0.1003,
+      "step": 2300
+    },
+    {
+      "epoch": 0.1438417740485466,
+      "grad_norm": 0.19500990211963654,
+      "learning_rate": 7.19208870242733e-06,
+      "loss": 0.1017,
+      "step": 2400
+    },
+    {
+      "epoch": 0.1498351813005694,
+      "grad_norm": 0.20727553963661194,
+      "learning_rate": 7.491759065028469e-06,
+      "loss": 0.1009,
+      "step": 2500
+    },
+    {
+      "epoch": 0.15582858855259216,
+      "grad_norm": 0.19461679458618164,
+      "learning_rate": 7.791429427629607e-06,
+      "loss": 0.1006,
+      "step": 2600
+    },
+    {
+      "epoch": 0.16182199580461493,
+      "grad_norm": 0.12335213273763657,
+      "learning_rate": 8.091099790230747e-06,
+      "loss": 0.1008,
+      "step": 2700
+    },
+    {
+      "epoch": 0.1678154030566377,
+      "grad_norm": 0.1840120255947113,
+      "learning_rate": 8.390770152831885e-06,
+      "loss": 0.0996,
+      "step": 2800
+    },
+    {
+      "epoch": 0.17380881030866047,
+      "grad_norm": 0.047822967171669006,
+      "learning_rate": 8.690440515433025e-06,
+      "loss": 0.1006,
+      "step": 2900
+    },
+    {
+      "epoch": 0.17980221756068324,
+      "grad_norm": 0.05022828280925751,
+      "learning_rate": 8.990110878034163e-06,
+      "loss": 0.1004,
+      "step": 3000
+    },
+    {
+      "epoch": 0.18579562481270603,
+      "grad_norm": 0.08173201233148575,
+      "learning_rate": 9.289781240635301e-06,
+      "loss": 0.1016,
+      "step": 3100
+    },
+    {
+      "epoch": 0.1917890320647288,
+      "grad_norm": 0.25700005888938904,
+      "learning_rate": 9.589451603236441e-06,
+      "loss": 0.1012,
+      "step": 3200
+    },
+    {
+      "epoch": 0.19778243931675157,
+      "grad_norm": 0.929848849773407,
+      "learning_rate": 9.889121965837579e-06,
+      "loss": 0.1003,
+      "step": 3300
+    },
+    {
+      "epoch": 0.20377584656877434,
+      "grad_norm": 0.16289019584655762,
+      "learning_rate": 9.97902307461792e-06,
+      "loss": 0.1005,
+      "step": 3400
+    },
+    {
+      "epoch": 0.2097692538207971,
+      "grad_norm": 0.8773489594459534,
+      "learning_rate": 9.94572636766224e-06,
+      "loss": 0.1007,
+      "step": 3500
+    },
+    {
+      "epoch": 0.2157626610728199,
+      "grad_norm": 0.0891551822423935,
+      "learning_rate": 9.912429660706557e-06,
+      "loss": 0.1001,
+      "step": 3600
+    },
+    {
+      "epoch": 0.22175606832484268,
+      "grad_norm": 0.06410754472017288,
+      "learning_rate": 9.879132953750875e-06,
+      "loss": 0.0993,
+      "step": 3700
+    },
+    {
+      "epoch": 0.22774947557686545,
+      "grad_norm": 0.21573640406131744,
+      "learning_rate": 9.845836246795192e-06,
+      "loss": 0.1004,
+      "step": 3800
+    },
+    {
+      "epoch": 0.23374288282888822,
+      "grad_norm": 0.07671983540058136,
+      "learning_rate": 9.812539539839511e-06,
+      "loss": 0.1006,
+      "step": 3900
+    },
+    {
+      "epoch": 0.239736290080911,
+      "grad_norm": 0.6497980952262878,
+      "learning_rate": 9.779242832883829e-06,
+      "loss": 0.1,
+      "step": 4000
+    },
+    {
+      "epoch": 0.24572969733293376,
+      "grad_norm": 0.7966341972351074,
+      "learning_rate": 9.745946125928146e-06,
+      "loss": 0.0982,
+      "step": 4100
+    },
+    {
+      "epoch": 0.25172310458495656,
+      "grad_norm": 0.13860899209976196,
+      "learning_rate": 9.712649418972464e-06,
+      "loss": 0.1007,
+      "step": 4200
+    },
+    {
+      "epoch": 0.2577165118369793,
+      "grad_norm": 0.21768826246261597,
+      "learning_rate": 9.679352712016781e-06,
+      "loss": 0.0996,
+      "step": 4300
+    },
+    {
+      "epoch": 0.2637099190890021,
+      "grad_norm": 0.5174150466918945,
+      "learning_rate": 9.6460560050611e-06,
+      "loss": 0.1006,
+      "step": 4400
+    },
+    {
+      "epoch": 0.2697033263410249,
+      "grad_norm": 0.5629359483718872,
+      "learning_rate": 9.61275929810542e-06,
+      "loss": 0.1014,
+      "step": 4500
+    },
+    {
+      "epoch": 0.27569673359304764,
+      "grad_norm": 0.21343261003494263,
+      "learning_rate": 9.579462591149737e-06,
+      "loss": 0.1011,
+      "step": 4600
+    },
+    {
+      "epoch": 0.28169014084507044,
+      "grad_norm": 0.36314478516578674,
+      "learning_rate": 9.546165884194054e-06,
+      "loss": 0.1006,
+      "step": 4700
+    },
+    {
+      "epoch": 0.2876835480970932,
+      "grad_norm": 0.07583048939704895,
+      "learning_rate": 9.512869177238371e-06,
+      "loss": 0.098,
+      "step": 4800
+    },
+    {
+      "epoch": 0.293676955349116,
+      "grad_norm": 0.014798377640545368,
+      "learning_rate": 9.479572470282689e-06,
+      "loss": 0.099,
+      "step": 4900
+    },
+    {
+      "epoch": 0.2996703626011388,
+      "grad_norm": 0.21406540274620056,
+      "learning_rate": 9.446275763327008e-06,
+      "loss": 0.0993,
+      "step": 5000
+    },
+    {
+      "epoch": 0.3056637698531615,
+      "grad_norm": 0.36721229553222656,
+      "learning_rate": 9.412979056371325e-06,
+      "loss": 0.1006,
+      "step": 5100
+    },
+    {
+      "epoch": 0.3116571771051843,
+      "grad_norm": 0.18309548497200012,
+      "learning_rate": 9.379682349415643e-06,
+      "loss": 0.0995,
+      "step": 5200
+    },
+    {
+      "epoch": 0.31765058435720706,
+      "grad_norm": 0.1970544010400772,
+      "learning_rate": 9.346385642459962e-06,
+      "loss": 0.0999,
+      "step": 5300
+    },
+    {
+      "epoch": 0.32364399160922985,
+      "grad_norm": 0.052760008722543716,
+      "learning_rate": 9.31308893550428e-06,
+      "loss": 0.0992,
+      "step": 5400
+    },
+    {
+      "epoch": 0.3296373988612526,
+      "grad_norm": 0.1352636069059372,
+      "learning_rate": 9.279792228548597e-06,
+      "loss": 0.099,
+      "step": 5500
+    },
+    {
+      "epoch": 0.3356308061132754,
+      "grad_norm": 0.024126173928380013,
+      "learning_rate": 9.246495521592916e-06,
+      "loss": 0.1001,
+      "step": 5600
+    },
+    {
+      "epoch": 0.3416242133652982,
+      "grad_norm": 0.12636955082416534,
+      "learning_rate": 9.213198814637233e-06,
+      "loss": 0.1003,
+      "step": 5700
+    },
+    {
+      "epoch": 0.34761762061732093,
+      "grad_norm": 0.021808089688420296,
+      "learning_rate": 9.17990210768155e-06,
+      "loss": 0.1,
+      "step": 5800
+    },
+    {
+      "epoch": 0.35361102786934373,
+      "grad_norm": 0.11340193450450897,
+      "learning_rate": 9.146605400725868e-06,
+      "loss": 0.1005,
+      "step": 5900
+    },
+    {
+      "epoch": 0.3596044351213665,
+      "grad_norm": 0.20579053461551666,
+      "learning_rate": 9.113308693770187e-06,
+      "loss": 0.1007,
+      "step": 6000
+    },
+    {
+      "epoch": 0.36559784237338927,
+      "grad_norm": 0.14165253937244415,
+      "learning_rate": 9.080011986814505e-06,
+      "loss": 0.0997,
+      "step": 6100
+    },
+    {
+      "epoch": 0.37159124962541207,
+      "grad_norm": 0.20944170653820038,
+      "learning_rate": 9.046715279858824e-06,
+      "loss": 0.0997,
+      "step": 6200
+    },
+    {
+      "epoch": 0.3775846568774348,
+      "grad_norm": 0.15122222900390625,
+      "learning_rate": 9.013418572903141e-06,
+      "loss": 0.0993,
+      "step": 6300
+    },
+    {
+      "epoch": 0.3835780641294576,
+      "grad_norm": 0.1885826140642166,
+      "learning_rate": 8.980121865947459e-06,
+      "loss": 0.0989,
+      "step": 6400
+    },
+    {
+      "epoch": 0.38957147138148035,
+      "grad_norm": 0.04748326912522316,
+      "learning_rate": 8.946825158991776e-06,
+      "loss": 0.0998,
+      "step": 6500
+    },
+    {
+      "epoch": 0.39556487863350315,
+      "grad_norm": 0.018379371613264084,
+      "learning_rate": 8.913528452036094e-06,
+      "loss": 0.0997,
+      "step": 6600
+    },
+    {
+      "epoch": 0.40155828588552595,
+      "grad_norm": 0.22810649871826172,
+      "learning_rate": 8.880231745080413e-06,
+      "loss": 0.0999,
+      "step": 6700
+    },
+    {
+      "epoch": 0.4075516931375487,
+      "grad_norm": 0.028912872076034546,
+      "learning_rate": 8.84693503812473e-06,
+      "loss": 0.0977,
+      "step": 6800
+    },
+    {
+      "epoch": 0.4135451003895715,
+      "grad_norm": 0.09700625389814377,
+      "learning_rate": 8.813638331169047e-06,
+      "loss": 0.0998,
+      "step": 6900
+    },
+    {
+      "epoch": 0.4195385076415942,
+      "grad_norm": 0.04979974031448364,
+      "learning_rate": 8.780341624213367e-06,
+      "loss": 0.0997,
+      "step": 7000
+    },
+    {
+      "epoch": 0.425531914893617,
+      "grad_norm": 0.10137055069208145,
+      "learning_rate": 8.747044917257684e-06,
+      "loss": 0.0991,
+      "step": 7100
+    },
+    {
+      "epoch": 0.4315253221456398,
+      "grad_norm": 0.0385773666203022,
+      "learning_rate": 8.713748210302001e-06,
+      "loss": 0.0996,
+      "step": 7200
+    },
+    {
+      "epoch": 0.43751872939766256,
+      "grad_norm": 0.05161421373486519,
+      "learning_rate": 8.68045150334632e-06,
+      "loss": 0.1003,
+      "step": 7300
+    },
+    {
+      "epoch": 0.44351213664968536,
+      "grad_norm": 0.14135126769542694,
+      "learning_rate": 8.647154796390638e-06,
+      "loss": 0.0996,
+      "step": 7400
+    },
+    {
+      "epoch": 0.4495055439017081,
+      "grad_norm": 0.40400320291519165,
+      "learning_rate": 8.613858089434955e-06,
+      "loss": 0.0995,
+      "step": 7500
+    },
+    {
+      "epoch": 0.4554989511537309,
+      "grad_norm": 0.5337746143341064,
+      "learning_rate": 8.580561382479273e-06,
+      "loss": 0.1002,
+      "step": 7600
+    },
+    {
+      "epoch": 0.46149235840575364,
+      "grad_norm": 0.25852474570274353,
+      "learning_rate": 8.547264675523592e-06,
+      "loss": 0.0988,
+      "step": 7700
+    },
+    {
+      "epoch": 0.46748576565777644,
+      "grad_norm": 0.02634989097714424,
+      "learning_rate": 8.51396796856791e-06,
+      "loss": 0.0998,
+      "step": 7800
+    },
+    {
+      "epoch": 0.47347917290979924,
+      "grad_norm": 0.045602887868881226,
+      "learning_rate": 8.480671261612228e-06,
+      "loss": 0.0993,
+      "step": 7900
+    },
+    {
+      "epoch": 0.479472580161822,
+      "grad_norm": 0.4190271198749542,
+      "learning_rate": 8.447374554656546e-06,
+      "loss": 0.0992,
+      "step": 8000
+    },
+    {
+      "epoch": 0.4854659874138448,
+      "grad_norm": 0.03802090883255005,
+      "learning_rate": 8.414077847700863e-06,
+      "loss": 0.1003,
+      "step": 8100
+    },
+    {
+      "epoch": 0.4914593946658675,
+      "grad_norm": 0.187332421541214,
+      "learning_rate": 8.38078114074518e-06,
+      "loss": 0.099,
+      "step": 8200
+    },
+    {
+      "epoch": 0.4974528019178903,
+      "grad_norm": 0.08949125558137894,
+      "learning_rate": 8.347484433789498e-06,
+      "loss": 0.0991,
+      "step": 8300
+    },
+    {
+      "epoch": 0.5034462091699131,
+      "grad_norm": 0.32420963048934937,
+      "learning_rate": 8.314187726833817e-06,
+      "loss": 0.1001,
+      "step": 8400
+    },
+    {
+      "epoch": 0.5094396164219359,
+      "grad_norm": 0.06165235489606857,
+      "learning_rate": 8.280891019878135e-06,
+      "loss": 0.0994,
+      "step": 8500
+    },
+    {
+      "epoch": 0.5154330236739586,
+      "grad_norm": 0.14306558668613434,
+      "learning_rate": 8.247594312922452e-06,
+      "loss": 0.1,
+      "step": 8600
+    },
+    {
+      "epoch": 0.5214264309259814,
+      "grad_norm": 0.1607675403356552,
+      "learning_rate": 8.21429760596677e-06,
+      "loss": 0.0999,
+      "step": 8700
+    },
+    {
+      "epoch": 0.5274198381780042,
+      "grad_norm": 0.16511593759059906,
+      "learning_rate": 8.181000899011089e-06,
+      "loss": 0.1012,
+      "step": 8800
+    },
+    {
+      "epoch": 0.533413245430027,
+      "grad_norm": 0.08098283410072327,
+      "learning_rate": 8.147704192055406e-06,
+      "loss": 0.0998,
+      "step": 8900
+    },
+    {
+      "epoch": 0.5394066526820498,
+      "grad_norm": 0.18850570917129517,
+      "learning_rate": 8.114407485099725e-06,
+      "loss": 0.1004,
+      "step": 9000
+    },
+    {
+      "epoch": 0.5454000599340725,
+      "grad_norm": 0.03290531411767006,
+      "learning_rate": 8.081110778144043e-06,
+      "loss": 0.0998,
+      "step": 9100
+    },
+    {
+      "epoch": 0.5513934671860953,
+      "grad_norm": 0.799527645111084,
+      "learning_rate": 8.04781407118836e-06,
+      "loss": 0.0995,
+      "step": 9200
+    },
+    {
+      "epoch": 0.5573868744381181,
+      "grad_norm": 0.45988729596138,
+      "learning_rate": 8.014517364232677e-06,
+      "loss": 0.1,
+      "step": 9300
+    },
+    {
+      "epoch": 0.5633802816901409,
+      "grad_norm": 0.035269495099782944,
+      "learning_rate": 7.981220657276996e-06,
+      "loss": 0.0993,
+      "step": 9400
+    },
+    {
+      "epoch": 0.5693736889421637,
+      "grad_norm": 0.29002872109413147,
+      "learning_rate": 7.947923950321314e-06,
+      "loss": 0.0994,
+      "step": 9500
+    },
+    {
+      "epoch": 0.5753670961941864,
+      "grad_norm": 0.06160065904259682,
+      "learning_rate": 7.914627243365631e-06,
+      "loss": 0.0995,
+      "step": 9600
+    },
+    {
+      "epoch": 0.5813605034462092,
+      "grad_norm": 0.14128552377223969,
+      "learning_rate": 7.88133053640995e-06,
+      "loss": 0.0998,
+      "step": 9700
+    },
+    {
+      "epoch": 0.587353910698232,
+      "grad_norm": 0.14790260791778564,
+      "learning_rate": 7.848033829454268e-06,
+      "loss": 0.0996,
+      "step": 9800
+    },
+    {
+      "epoch": 0.5933473179502547,
+      "grad_norm": 0.4272247552871704,
+      "learning_rate": 7.814737122498585e-06,
+      "loss": 0.0996,
+      "step": 9900
+    },
+    {
+      "epoch": 0.5993407252022775,
+      "grad_norm": 0.39960777759552,
+      "learning_rate": 7.781440415542903e-06,
+      "loss": 0.1001,
+      "step": 10000
+    },
+    {
+      "epoch": 0.6053341324543002,
+      "grad_norm": 0.27847522497177124,
+      "learning_rate": 7.748143708587222e-06,
+      "loss": 0.0986,
+      "step": 10100
+    },
+    {
+      "epoch": 0.611327539706323,
+      "grad_norm": 0.06485463678836823,
+      "learning_rate": 7.71484700163154e-06,
+      "loss": 0.0985,
+      "step": 10200
+    },
+    {
+      "epoch": 0.6173209469583458,
+      "grad_norm": 0.05133885145187378,
+      "learning_rate": 7.681550294675857e-06,
+      "loss": 0.0996,
+      "step": 10300
+    },
+    {
+      "epoch": 0.6233143542103686,
+      "grad_norm": 0.0617876760661602,
+      "learning_rate": 7.648253587720174e-06,
+      "loss": 0.0989,
+      "step": 10400
+    },
+    {
+      "epoch": 0.6293077614623914,
+      "grad_norm": 0.06740323454141617,
+      "learning_rate": 7.614956880764492e-06,
+      "loss": 0.1,
+      "step": 10500
+    },
+    {
+      "epoch": 0.6353011687144141,
+      "grad_norm": 0.20278900861740112,
+      "learning_rate": 7.5816601738088115e-06,
+      "loss": 0.0999,
+      "step": 10600
+    },
+    {
+      "epoch": 0.6412945759664369,
+      "grad_norm": 0.06642678380012512,
+      "learning_rate": 7.548363466853129e-06,
+      "loss": 0.0991,
+      "step": 10700
+    },
+    {
+      "epoch": 0.6472879832184597,
+      "grad_norm": 0.058061473071575165,
+      "learning_rate": 7.515066759897447e-06,
+      "loss": 0.0997,
+      "step": 10800
+    },
+    {
+      "epoch": 0.6532813904704825,
+      "grad_norm": 0.20366477966308594,
+      "learning_rate": 7.4817700529417646e-06,
+      "loss": 0.0983,
+      "step": 10900
+    },
+    {
+      "epoch": 0.6592747977225052,
+      "grad_norm": 0.016720598563551903,
+      "learning_rate": 7.448473345986083e-06,
+      "loss": 0.0991,
+      "step": 11000
+    },
+    {
+      "epoch": 0.665268204974528,
+      "grad_norm": 0.0325213186442852,
+      "learning_rate": 7.4151766390304e-06,
+      "loss": 0.0988,
+      "step": 11100
+    },
+    {
+      "epoch": 0.6712616122265508,
+      "grad_norm": 0.06385581195354462,
+      "learning_rate": 7.381879932074718e-06,
+      "loss": 0.0996,
+      "step": 11200
+    },
+    {
+      "epoch": 0.6772550194785736,
+      "grad_norm": 0.2656920850276947,
+      "learning_rate": 7.348583225119036e-06,
+      "loss": 0.1001,
+      "step": 11300
+    },
+    {
+      "epoch": 0.6832484267305964,
+      "grad_norm": 0.04199494421482086,
+      "learning_rate": 7.315286518163355e-06,
+      "loss": 0.0994,
+      "step": 11400
+    },
+    {
+      "epoch": 0.6892418339826191,
+      "grad_norm": 0.23018859326839447,
+      "learning_rate": 7.2819898112076725e-06,
+      "loss": 0.1008,
+      "step": 11500
+    },
+    {
+      "epoch": 0.6952352412346419,
+      "grad_norm": 0.16758233308792114,
+      "learning_rate": 7.248693104251991e-06,
+      "loss": 0.0996,
+      "step": 11600
+    },
+    {
+      "epoch": 0.7012286484866647,
+      "grad_norm": 0.07982943207025528,
+      "learning_rate": 7.215396397296308e-06,
+      "loss": 0.0987,
+      "step": 11700
+    },
+    {
+      "epoch": 0.7072220557386875,
+      "grad_norm": 0.17444317042827606,
+      "learning_rate": 7.1820996903406256e-06,
+      "loss": 0.1003,
+      "step": 11800
+    },
+    {
+      "epoch": 0.7132154629907103,
+      "grad_norm": 0.019602667540311813,
+      "learning_rate": 7.148802983384944e-06,
+      "loss": 0.0998,
+      "step": 11900
+    },
+    {
+      "epoch": 0.719208870242733,
+      "grad_norm": 0.06751447916030884,
+      "learning_rate": 7.115506276429261e-06,
+      "loss": 0.0996,
+      "step": 12000
+    },
+    {
+      "epoch": 0.7252022774947557,
+      "grad_norm": 0.6664422154426575,
+      "learning_rate": 7.0822095694735795e-06,
+      "loss": 0.0999,
+      "step": 12100
+    },
+    {
+      "epoch": 0.7311956847467785,
+      "grad_norm": 0.11597349494695663,
+      "learning_rate": 7.048912862517897e-06,
+      "loss": 0.098,
+      "step": 12200
+    },
+    {
+      "epoch": 0.7371890919988013,
+      "grad_norm": 0.021992051973938942,
+      "learning_rate": 7.015616155562216e-06,
+      "loss": 0.0997,
+      "step": 12300
+    },
+    {
+      "epoch": 0.7431824992508241,
+      "grad_norm": 0.04850628226995468,
+      "learning_rate": 6.9823194486065335e-06,
+      "loss": 0.1002,
+      "step": 12400
+    },
+    {
+      "epoch": 0.7491759065028468,
+      "grad_norm": 0.05067530274391174,
+      "learning_rate": 6.949022741650852e-06,
+      "loss": 0.0993,
+      "step": 12500
+    },
+    {
+      "epoch": 0.7551693137548696,
+      "grad_norm": 0.0720600038766861,
+      "learning_rate": 6.915726034695169e-06,
+      "loss": 0.0993,
+      "step": 12600
+    },
+    {
+      "epoch": 0.7611627210068924,
+      "grad_norm": 0.05677128955721855,
+      "learning_rate": 6.8824293277394874e-06,
+      "loss": 0.0999,
+      "step": 12700
+    },
+    {
+      "epoch": 0.7671561282589152,
+      "grad_norm": 0.06587408483028412,
+      "learning_rate": 6.849132620783805e-06,
+      "loss": 0.0989,
+      "step": 12800
+    },
+    {
+      "epoch": 0.773149535510938,
+      "grad_norm": 0.1323852837085724,
+      "learning_rate": 6.815835913828123e-06,
+      "loss": 0.0992,
+      "step": 12900
+    },
+    {
+      "epoch": 0.7791429427629607,
+      "grad_norm": 0.016273049637675285,
+      "learning_rate": 6.7825392068724405e-06,
+      "loss": 0.0999,
+      "step": 13000
+    },
+    {
+      "epoch": 0.7851363500149835,
+      "grad_norm": 0.11021006107330322,
+      "learning_rate": 6.749242499916758e-06,
+      "loss": 0.0993,
+      "step": 13100
+    },
+    {
+      "epoch": 0.7911297572670063,
+      "grad_norm": 0.03431228548288345,
+      "learning_rate": 6.715945792961077e-06,
+      "loss": 0.0995,
+      "step": 13200
+    },
+    {
+      "epoch": 0.7971231645190291,
+      "grad_norm": 0.19659392535686493,
+      "learning_rate": 6.682649086005395e-06,
+      "loss": 0.0991,
+      "step": 13300
+    },
+    {
+      "epoch": 0.8031165717710519,
+      "grad_norm": 0.17087887227535248,
+      "learning_rate": 6.649352379049713e-06,
+      "loss": 0.0999,
+      "step": 13400
+    },
+    {
+      "epoch": 0.8091099790230746,
+      "grad_norm": 0.04571348801255226,
+      "learning_rate": 6.61605567209403e-06,
+      "loss": 0.0987,
+      "step": 13500
+    },
+    {
+      "epoch": 0.8151033862750974,
+      "grad_norm": 0.06407848745584488,
+      "learning_rate": 6.5827589651383484e-06,
+      "loss": 0.0995,
+      "step": 13600
+    },
+    {
+      "epoch": 0.8210967935271202,
+      "grad_norm": 0.041546691209077835,
+      "learning_rate": 6.549462258182666e-06,
+      "loss": 0.1001,
+      "step": 13700
+    },
+    {
+      "epoch": 0.827090200779143,
+      "grad_norm": 0.027348553761839867,
+      "learning_rate": 6.516165551226984e-06,
+      "loss": 0.099,
+      "step": 13800
+    },
+    {
+      "epoch": 0.8330836080311658,
+      "grad_norm": 0.08637899160385132,
+      "learning_rate": 6.4828688442713015e-06,
+      "loss": 0.0987,
+      "step": 13900
+    },
+    {
+      "epoch": 0.8390770152831885,
+      "grad_norm": 0.02239215560257435,
+      "learning_rate": 6.44957213731562e-06,
+      "loss": 0.0992,
+      "step": 14000
+    },
+    {
+      "epoch": 0.8450704225352113,
+      "grad_norm": 0.03839031606912613,
+      "learning_rate": 6.416275430359938e-06,
+      "loss": 0.0987,
+      "step": 14100
+    },
+    {
+      "epoch": 0.851063829787234,
+      "grad_norm": 0.08829605579376221,
+      "learning_rate": 6.382978723404256e-06,
+      "loss": 0.0988,
+      "step": 14200
+    },
+    {
+      "epoch": 0.8570572370392568,
+      "grad_norm": 0.2102310210466385,
+      "learning_rate": 6.349682016448574e-06,
+      "loss": 0.0991,
+      "step": 14300
+    },
+    {
+      "epoch": 0.8630506442912796,
+      "grad_norm": 0.033268921077251434,
+      "learning_rate": 6.316385309492892e-06,
+      "loss": 0.0983,
+      "step": 14400
+    },
+    {
+      "epoch": 0.8690440515433023,
+      "grad_norm": 0.040406860411167145,
+      "learning_rate": 6.2830886025372095e-06,
+      "loss": 0.0973,
+      "step": 14500
+    },
+    {
+      "epoch": 0.8750374587953251,
+      "grad_norm": 0.04502435401082039,
+      "learning_rate": 6.249791895581528e-06,
+      "loss": 0.0999,
+      "step": 14600
+    },
+    {
+      "epoch": 0.8810308660473479,
+      "grad_norm": 0.026787765324115753,
+      "learning_rate": 6.216495188625845e-06,
+      "loss": 0.1002,
+      "step": 14700
+    },
+    {
+      "epoch": 0.8870242732993707,
+      "grad_norm": 0.13938550651073456,
+      "learning_rate": 6.1831984816701626e-06,
+      "loss": 0.099,
+      "step": 14800
+    },
+    {
+      "epoch": 0.8930176805513935,
+      "grad_norm": 0.135736882686615,
+      "learning_rate": 6.149901774714481e-06,
+      "loss": 0.0999,
+      "step": 14900
+    },
+    {
+      "epoch": 0.8990110878034162,
+      "grad_norm": 0.15793441236019135,
+      "learning_rate": 6.1166050677588e-06,
+      "loss": 0.0989,
+      "step": 15000
+    },
+    {
+      "epoch": 0.905004495055439,
+      "grad_norm": 0.08003054559230804,
+      "learning_rate": 6.083308360803117e-06,
+      "loss": 0.0991,
+      "step": 15100
+    },
+    {
+      "epoch": 0.9109979023074618,
+      "grad_norm": 0.0406046137213707,
+      "learning_rate": 6.050011653847435e-06,
+      "loss": 0.0985,
+      "step": 15200
+    },
+    {
+      "epoch": 0.9169913095594846,
+      "grad_norm": 0.04845142737030983,
+      "learning_rate": 6.016714946891753e-06,
+      "loss": 0.099,
+      "step": 15300
+    },
+    {
+      "epoch": 0.9229847168115073,
+      "grad_norm": 0.023725613951683044,
+      "learning_rate": 5.9834182399360705e-06,
+      "loss": 0.0995,
+      "step": 15400
+    },
+    {
+      "epoch": 0.9289781240635301,
+      "grad_norm": 0.04039942845702171,
+      "learning_rate": 5.950121532980389e-06,
+      "loss": 0.0996,
+      "step": 15500
+    },
+    {
+      "epoch": 0.9349715313155529,
+      "grad_norm": 0.026223324239253998,
+      "learning_rate": 5.916824826024706e-06,
+      "loss": 0.0997,
+      "step": 15600
+    },
+    {
+      "epoch": 0.9409649385675757,
+      "grad_norm": 0.27790361642837524,
+      "learning_rate": 5.883528119069024e-06,
+      "loss": 0.099,
+      "step": 15700
+    },
+    {
+      "epoch": 0.9469583458195985,
+      "grad_norm": 0.058665353804826736,
+      "learning_rate": 5.850231412113343e-06,
+      "loss": 0.0998,
+      "step": 15800
+    },
+    {
+      "epoch": 0.9529517530716212,
+      "grad_norm": 0.02454979345202446,
+      "learning_rate": 5.816934705157661e-06,
+      "loss": 0.1003,
+      "step": 15900
+    },
+    {
+      "epoch": 0.958945160323644,
+      "grad_norm": 0.06335332244634628,
+      "learning_rate": 5.783637998201978e-06,
+      "loss": 0.0995,
+      "step": 16000
+    },
+    {
+      "epoch": 0.9649385675756668,
+      "grad_norm": 0.10977023839950562,
+      "learning_rate": 5.750341291246297e-06,
+      "loss": 0.0974,
+      "step": 16100
+    },
+    {
+      "epoch": 0.9709319748276896,
+      "grad_norm": 0.025012081488966942,
+      "learning_rate": 5.717044584290614e-06,
+      "loss": 0.0992,
+      "step": 16200
+    },
+    {
+      "epoch": 0.9769253820797124,
+      "grad_norm": 0.022056572139263153,
+      "learning_rate": 5.683747877334932e-06,
+      "loss": 0.099,
+      "step": 16300
+    },
+    {
+      "epoch": 0.982918789331735,
+      "grad_norm": 0.10653574019670486,
+      "learning_rate": 5.65045117037925e-06,
+      "loss": 0.1009,
+      "step": 16400
+    },
+    {
+      "epoch": 0.9889121965837578,
+      "grad_norm": 0.04108584672212601,
+      "learning_rate": 5.617154463423567e-06,
+      "loss": 0.0995,
+      "step": 16500
+    },
+    {
+      "epoch": 0.9949056038357806,
+      "grad_norm": 0.05849546194076538,
+      "learning_rate": 5.583857756467885e-06,
+      "loss": 0.0991,
+      "step": 16600
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.7163220105947296,
+      "eval_f1": 0.6846578561420671,
+      "eval_loss": 0.09925069659948349,
+      "eval_precision": 0.6620272988688642,
+      "eval_recall": 0.7163220105947296,
+      "eval_runtime": 253.2812,
+      "eval_samples_per_second": 58.555,
+      "eval_steps_per_second": 7.32,
+      "step": 16685
+    },
+    {
+      "epoch": 1.0008990110878033,
+      "grad_norm": 0.022984975948929787,
+      "learning_rate": 5.5505610495122045e-06,
+      "loss": 0.0986,
+      "step": 16700
+    },
+    {
+      "epoch": 1.0068924183398262,
+      "grad_norm": 0.022939996793866158,
+      "learning_rate": 5.517264342556522e-06,
+      "loss": 0.0987,
+      "step": 16800
+    },
+    {
+      "epoch": 1.012885825591849,
+      "grad_norm": 0.014104674570262432,
+      "learning_rate": 5.483967635600839e-06,
+      "loss": 0.098,
+      "step": 16900
+    },
+    {
+      "epoch": 1.0188792328438718,
+      "grad_norm": 0.02679424174129963,
+      "learning_rate": 5.450670928645158e-06,
+      "loss": 0.0978,
+      "step": 17000
+    },
+    {
+      "epoch": 1.0248726400958945,
+      "grad_norm": 0.007641422096639872,
+      "learning_rate": 5.417374221689475e-06,
+      "loss": 0.0982,
+      "step": 17100
+    },
+    {
+      "epoch": 1.0308660473479172,
+      "grad_norm": 0.03306014835834503,
+      "learning_rate": 5.384077514733793e-06,
+      "loss": 0.0972,
+      "step": 17200
+    },
+    {
+      "epoch": 1.0368594545999401,
+      "grad_norm": 0.053686805069446564,
+      "learning_rate": 5.350780807778111e-06,
+      "loss": 0.0976,
+      "step": 17300
+    },
+    {
+      "epoch": 1.0428528618519628,
+      "grad_norm": 0.05771619454026222,
+      "learning_rate": 5.317484100822429e-06,
+      "loss": 0.0986,
+      "step": 17400
+    },
+    {
+      "epoch": 1.0488462691039857,
+      "grad_norm": 0.08814644813537598,
+      "learning_rate": 5.2841873938667464e-06,
+      "loss": 0.0978,
+      "step": 17500
+    },
+    {
+      "epoch": 1.0548396763560084,
+      "grad_norm": 0.18690770864486694,
+      "learning_rate": 5.2508906869110655e-06,
+      "loss": 0.0986,
+      "step": 17600
+    },
+    {
+      "epoch": 1.060833083608031,
+      "grad_norm": 0.01733516715466976,
+      "learning_rate": 5.217593979955383e-06,
+      "loss": 0.0987,
+      "step": 17700
+    },
+    {
+      "epoch": 1.066826490860054,
+      "grad_norm": 0.12736350297927856,
+      "learning_rate": 5.184297272999701e-06,
+      "loss": 0.0987,
+      "step": 17800
+    },
+    {
+      "epoch": 1.0728198981120767,
+      "grad_norm": 0.1056164801120758,
+      "learning_rate": 5.151000566044019e-06,
+      "loss": 0.0984,
+      "step": 17900
+    },
+    {
+      "epoch": 1.0788133053640996,
+      "grad_norm": 0.014793598093092442,
+      "learning_rate": 5.117703859088337e-06,
+      "loss": 0.0973,
+      "step": 18000
+    },
+    {
+      "epoch": 1.0848067126161223,
+      "grad_norm": 0.015460701659321785,
+      "learning_rate": 5.084407152132654e-06,
+      "loss": 0.0975,
+      "step": 18100
+    },
+    {
+      "epoch": 1.090800119868145,
+      "grad_norm": 0.01873720809817314,
+      "learning_rate": 5.051110445176972e-06,
+      "loss": 0.0987,
+      "step": 18200
+    },
+    {
+      "epoch": 1.0967935271201679,
+      "grad_norm": 0.01362083200365305,
+      "learning_rate": 5.01781373822129e-06,
+      "loss": 0.0984,
+      "step": 18300
+    },
+    {
+      "epoch": 1.1027869343721906,
+      "grad_norm": 0.019239643588662148,
+      "learning_rate": 4.984517031265608e-06,
+      "loss": 0.0986,
+      "step": 18400
+    },
+    {
+      "epoch": 1.1087803416242135,
+      "grad_norm": 0.03043738380074501,
+      "learning_rate": 4.951220324309926e-06,
+      "loss": 0.0986,
+      "step": 18500
+    },
+    {
+      "epoch": 1.1147737488762361,
+      "grad_norm": 0.12742625176906586,
+      "learning_rate": 4.917923617354244e-06,
+      "loss": 0.0983,
+      "step": 18600
+    },
+    {
+      "epoch": 1.1207671561282588,
+      "grad_norm": 0.18706390261650085,
+      "learning_rate": 4.884626910398562e-06,
+      "loss": 0.0991,
+      "step": 18700
+    },
+    {
+      "epoch": 1.1267605633802817,
+      "grad_norm": 0.21639247238636017,
+      "learning_rate": 4.85133020344288e-06,
+      "loss": 0.0984,
+      "step": 18800
+    },
+    {
+      "epoch": 1.1327539706323044,
+      "grad_norm": 0.06566977500915527,
+      "learning_rate": 4.818033496487198e-06,
+      "loss": 0.0983,
+      "step": 18900
+    },
+    {
+      "epoch": 1.1387473778843273,
+      "grad_norm": 0.06070713698863983,
+      "learning_rate": 4.784736789531515e-06,
+      "loss": 0.0984,
+      "step": 19000
+    },
+    {
+      "epoch": 1.14474078513635,
+      "grad_norm": 0.01181207038462162,
+      "learning_rate": 4.751440082575834e-06,
+      "loss": 0.0967,
+      "step": 19100
+    },
+    {
+      "epoch": 1.1507341923883727,
+      "grad_norm": 0.014737865887582302,
+      "learning_rate": 4.718143375620152e-06,
+      "loss": 0.0979,
+      "step": 19200
+    },
+    {
+      "epoch": 1.1567275996403956,
+      "grad_norm": 0.09678817540407181,
+      "learning_rate": 4.684846668664469e-06,
+      "loss": 0.0987,
+      "step": 19300
+    },
+    {
+      "epoch": 1.1627210068924183,
+      "grad_norm": 0.042408786714076996,
+      "learning_rate": 4.6515499617087876e-06,
+      "loss": 0.0989,
+      "step": 19400
+    },
+    {
+      "epoch": 1.168714414144441,
+      "grad_norm": 0.026022640988230705,
+      "learning_rate": 4.618253254753105e-06,
+      "loss": 0.0976,
+      "step": 19500
+    },
+    {
+      "epoch": 1.174707821396464,
+      "grad_norm": 0.029273219406604767,
+      "learning_rate": 4.584956547797423e-06,
+      "loss": 0.0971,
+      "step": 19600
+    },
+    {
+      "epoch": 1.1807012286484866,
+      "grad_norm": 0.019847506657242775,
+      "learning_rate": 4.5516598408417415e-06,
+      "loss": 0.0979,
+      "step": 19700
+    },
+    {
+      "epoch": 1.1866946359005095,
+      "grad_norm": 0.09878643602132797,
+      "learning_rate": 4.518363133886059e-06,
+      "loss": 0.0983,
+      "step": 19800
+    },
+    {
+      "epoch": 1.1926880431525322,
+      "grad_norm": 0.031175991520285606,
+      "learning_rate": 4.485066426930376e-06,
+      "loss": 0.0986,
+      "step": 19900
+    },
+    {
+      "epoch": 1.198681450404555,
+      "grad_norm": 0.057501692324876785,
+      "learning_rate": 4.4517697199746955e-06,
+      "loss": 0.098,
+      "step": 20000
+    },
+    {
+      "epoch": 1.2046748576565778,
+      "grad_norm": 0.06590039283037186,
+      "learning_rate": 4.418473013019013e-06,
+      "loss": 0.0978,
+      "step": 20100
+    },
+    {
+      "epoch": 1.2106682649086005,
+      "grad_norm": 0.021284347400069237,
+      "learning_rate": 4.38517630606333e-06,
+      "loss": 0.0982,
+      "step": 20200
+    },
+    {
+      "epoch": 1.2166616721606234,
+      "grad_norm": 0.26528528332710266,
+      "learning_rate": 4.3518795991076486e-06,
+      "loss": 0.0978,
+      "step": 20300
+    },
+    {
+      "epoch": 1.222655079412646,
+      "grad_norm": 0.06511413305997849,
+      "learning_rate": 4.318582892151967e-06,
+      "loss": 0.099,
+      "step": 20400
+    },
+    {
+      "epoch": 1.2286484866646687,
+      "grad_norm": 0.048637229949235916,
+      "learning_rate": 4.285286185196284e-06,
+      "loss": 0.0978,
+      "step": 20500
+    },
+    {
+      "epoch": 1.2346418939166917,
+      "grad_norm": 0.021369967609643936,
+      "learning_rate": 4.2519894782406025e-06,
+      "loss": 0.0984,
+      "step": 20600
+    },
+    {
+      "epoch": 1.2406353011687143,
+      "grad_norm": 0.5301014184951782,
+      "learning_rate": 4.21869277128492e-06,
+      "loss": 0.0975,
+      "step": 20700
+    },
+    {
+      "epoch": 1.2466287084207373,
+      "grad_norm": 0.01625882275402546,
+      "learning_rate": 4.185396064329238e-06,
+      "loss": 0.0983,
+      "step": 20800
+    },
+    {
+      "epoch": 1.25262211567276,
+      "grad_norm": 0.16671617329120636,
+      "learning_rate": 4.1520993573735565e-06,
+      "loss": 0.0986,
+      "step": 20900
+    },
+    {
+      "epoch": 1.2586155229247828,
+      "grad_norm": 0.15410806238651276,
+      "learning_rate": 4.118802650417874e-06,
+      "loss": 0.0992,
+      "step": 21000
+    },
+    {
+      "epoch": 1.2646089301768055,
+      "grad_norm": 0.030888069421052933,
+      "learning_rate": 4.085505943462192e-06,
+      "loss": 0.0981,
+      "step": 21100
+    },
+    {
+      "epoch": 1.2706023374288282,
+      "grad_norm": 0.014099245890974998,
+      "learning_rate": 4.05220923650651e-06,
+      "loss": 0.0985,
+      "step": 21200
+    },
+    {
+      "epoch": 1.2765957446808511,
+      "grad_norm": 0.019736776128411293,
+      "learning_rate": 4.018912529550828e-06,
+      "loss": 0.0976,
+      "step": 21300
+    },
+    {
+      "epoch": 1.2825891519328738,
+      "grad_norm": 0.011912483721971512,
+      "learning_rate": 3.985615822595146e-06,
+      "loss": 0.0987,
+      "step": 21400
+    },
+    {
+      "epoch": 1.2885825591848965,
+      "grad_norm": 0.030794724822044373,
+      "learning_rate": 3.9523191156394635e-06,
+      "loss": 0.0986,
+      "step": 21500
+    },
+    {
+      "epoch": 1.2945759664369194,
+      "grad_norm": 0.018548903986811638,
+      "learning_rate": 3.919022408683781e-06,
+      "loss": 0.0993,
+      "step": 21600
+    },
+    {
+      "epoch": 1.300569373688942,
+      "grad_norm": 0.0847175121307373,
+      "learning_rate": 3.885725701728099e-06,
+      "loss": 0.0985,
+      "step": 21700
+    },
+    {
+      "epoch": 1.306562780940965,
+      "grad_norm": 0.02340412326157093,
+      "learning_rate": 3.8524289947724175e-06,
+      "loss": 0.0988,
+      "step": 21800
+    },
+    {
+      "epoch": 1.3125561881929877,
+      "grad_norm": 0.06334514170885086,
+      "learning_rate": 3.819132287816735e-06,
+      "loss": 0.0988,
+      "step": 21900
+    },
+    {
+      "epoch": 1.3185495954450106,
+      "grad_norm": 0.03040698543190956,
+      "learning_rate": 3.785835580861053e-06,
+      "loss": 0.0984,
+      "step": 22000
+    },
+    {
+      "epoch": 1.3245430026970333,
+      "grad_norm": 0.4113400876522064,
+      "learning_rate": 3.752538873905371e-06,
+      "loss": 0.0989,
+      "step": 22100
+    },
+    {
+      "epoch": 1.330536409949056,
+      "grad_norm": 0.03668079525232315,
+      "learning_rate": 3.7192421669496893e-06,
+      "loss": 0.0983,
+      "step": 22200
+    },
+    {
+      "epoch": 1.3365298172010789,
+      "grad_norm": 0.040583133697509766,
+      "learning_rate": 3.685945459994007e-06,
+      "loss": 0.0982,
+      "step": 22300
+    },
+    {
+      "epoch": 1.3425232244531016,
+      "grad_norm": 0.2958316504955292,
+      "learning_rate": 3.652648753038325e-06,
+      "loss": 0.0976,
+      "step": 22400
+    },
+    {
+      "epoch": 1.3485166317051243,
+      "grad_norm": 0.07291315495967865,
+      "learning_rate": 3.6193520460826424e-06,
+      "loss": 0.0993,
+      "step": 22500
+    },
+    {
+      "epoch": 1.3545100389571472,
+      "grad_norm": 0.16284628212451935,
+      "learning_rate": 3.586055339126961e-06,
+      "loss": 0.0979,
+      "step": 22600
+    },
+    {
+      "epoch": 1.3605034462091699,
+      "grad_norm": 0.018767952919006348,
+      "learning_rate": 3.5527586321712785e-06,
+      "loss": 0.0983,
+      "step": 22700
+    },
+    {
+      "epoch": 1.3664968534611928,
+      "grad_norm": 0.32872429490089417,
+      "learning_rate": 3.5194619252155963e-06,
+      "loss": 0.0971,
+      "step": 22800
+    },
+    {
+      "epoch": 1.3724902607132154,
+      "grad_norm": 0.10787306725978851,
+      "learning_rate": 3.486165218259914e-06,
+      "loss": 0.0988,
+      "step": 22900
+    },
+    {
+      "epoch": 1.3784836679652384,
+      "grad_norm": 0.11872507631778717,
+      "learning_rate": 3.452868511304232e-06,
+      "loss": 0.0985,
+      "step": 23000
+    },
+    {
+      "epoch": 1.384477075217261,
+      "grad_norm": 0.01392502523958683,
+      "learning_rate": 3.4195718043485503e-06,
+      "loss": 0.0987,
+      "step": 23100
+    },
+    {
+      "epoch": 1.3904704824692837,
+      "grad_norm": 0.2750115394592285,
+      "learning_rate": 3.386275097392868e-06,
+      "loss": 0.0997,
+      "step": 23200
+    },
+    {
+      "epoch": 1.3964638897213066,
+      "grad_norm": 0.01633540168404579,
+      "learning_rate": 3.352978390437186e-06,
+      "loss": 0.0994,
+      "step": 23300
+    },
+    {
+      "epoch": 1.4024572969733293,
+      "grad_norm": 0.018824318423867226,
+      "learning_rate": 3.319681683481504e-06,
+      "loss": 0.0985,
+      "step": 23400
+    },
+    {
+      "epoch": 1.408450704225352,
+      "grad_norm": 0.09030365198850632,
+      "learning_rate": 3.286384976525822e-06,
+      "loss": 0.0984,
+      "step": 23500
+    },
+    {
+      "epoch": 1.414444111477375,
+      "grad_norm": 0.014728711917996407,
+      "learning_rate": 3.25308826957014e-06,
+      "loss": 0.0986,
+      "step": 23600
+    },
+    {
+      "epoch": 1.4204375187293976,
+      "grad_norm": 0.1276281625032425,
+      "learning_rate": 3.2197915626144578e-06,
+      "loss": 0.0988,
+      "step": 23700
+    },
+    {
+      "epoch": 1.4264309259814205,
+      "grad_norm": 0.0615570992231369,
+      "learning_rate": 3.1864948556587756e-06,
+      "loss": 0.0984,
+      "step": 23800
+    },
+    {
+      "epoch": 1.4324243332334432,
+      "grad_norm": 0.04221341758966446,
+      "learning_rate": 3.1531981487030935e-06,
+      "loss": 0.0977,
+      "step": 23900
+    },
+    {
+      "epoch": 1.4384177404854661,
+      "grad_norm": 0.022126736119389534,
+      "learning_rate": 3.1199014417474117e-06,
+      "loss": 0.0982,
+      "step": 24000
+    },
+    {
+      "epoch": 1.4444111477374888,
+      "grad_norm": 0.02455032989382744,
+      "learning_rate": 3.0866047347917296e-06,
+      "loss": 0.0976,
+      "step": 24100
+    },
+    {
+      "epoch": 1.4504045549895115,
+      "grad_norm": 0.5411427617073059,
+      "learning_rate": 3.053308027836047e-06,
+      "loss": 0.0971,
+      "step": 24200
+    },
+    {
+      "epoch": 1.4563979622415344,
+      "grad_norm": 0.22378544509410858,
+      "learning_rate": 3.020011320880365e-06,
+      "loss": 0.0991,
+      "step": 24300
+    },
+    {
+      "epoch": 1.462391369493557,
+      "grad_norm": 0.03697386384010315,
+      "learning_rate": 2.9867146139246835e-06,
+      "loss": 0.0987,
+      "step": 24400
+    },
+    {
+      "epoch": 1.4683847767455798,
+      "grad_norm": 0.05567880719900131,
+      "learning_rate": 2.953417906969001e-06,
+      "loss": 0.0973,
+      "step": 24500
+    },
+    {
+      "epoch": 1.4743781839976027,
+      "grad_norm": 0.008923672139644623,
+      "learning_rate": 2.9201212000133188e-06,
+      "loss": 0.0974,
+      "step": 24600
+    },
+    {
+      "epoch": 1.4803715912496254,
+      "grad_norm": 0.7570419907569885,
+      "learning_rate": 2.8868244930576366e-06,
+      "loss": 0.099,
+      "step": 24700
+    },
+    {
+      "epoch": 1.486364998501648,
+      "grad_norm": 0.03995317593216896,
+      "learning_rate": 2.853527786101955e-06,
+      "loss": 0.0976,
+      "step": 24800
+    },
+    {
+      "epoch": 1.492358405753671,
+      "grad_norm": 0.015565958805382252,
+      "learning_rate": 2.8202310791462727e-06,
+      "loss": 0.0971,
+      "step": 24900
+    },
+    {
+      "epoch": 1.4983518130056939,
+      "grad_norm": 0.06947562098503113,
+      "learning_rate": 2.7869343721905906e-06,
+      "loss": 0.0986,
+      "step": 25000
+    },
+    {
+      "epoch": 1.5043452202577166,
+      "grad_norm": 0.09315433353185654,
+      "learning_rate": 2.7536376652349084e-06,
+      "loss": 0.099,
+      "step": 25100
+    },
+    {
+      "epoch": 1.5103386275097392,
+      "grad_norm": 0.03829352185130119,
+      "learning_rate": 2.7203409582792263e-06,
+      "loss": 0.0986,
+      "step": 25200
+    },
+    {
+      "epoch": 1.5163320347617621,
+      "grad_norm": 0.0827546939253807,
+      "learning_rate": 2.6870442513235445e-06,
+      "loss": 0.0981,
+      "step": 25300
+    },
+    {
+      "epoch": 1.5223254420137848,
+      "grad_norm": 0.009773751720786095,
+      "learning_rate": 2.6537475443678624e-06,
+      "loss": 0.0981,
+      "step": 25400
+    },
+    {
+      "epoch": 1.5283188492658075,
+      "grad_norm": 0.03792467713356018,
+      "learning_rate": 2.6204508374121802e-06,
+      "loss": 0.0971,
+      "step": 25500
+    },
+    {
+      "epoch": 1.5343122565178304,
+      "grad_norm": 0.2956952452659607,
+      "learning_rate": 2.587154130456498e-06,
+      "loss": 0.0973,
+      "step": 25600
+    },
+    {
+      "epoch": 1.5403056637698531,
+      "grad_norm": 0.007948148064315319,
+      "learning_rate": 2.5538574235008163e-06,
+      "loss": 0.0992,
+      "step": 25700
+    },
+    {
+      "epoch": 1.5462990710218758,
+      "grad_norm": 0.030265573412179947,
+      "learning_rate": 2.520560716545134e-06,
+      "loss": 0.0978,
+      "step": 25800
+    },
+    {
+      "epoch": 1.5522924782738987,
+      "grad_norm": 0.06282780319452286,
+      "learning_rate": 2.487264009589452e-06,
+      "loss": 0.0988,
+      "step": 25900
+    },
+    {
+      "epoch": 1.5582858855259216,
+      "grad_norm": 0.19928309321403503,
+      "learning_rate": 2.4539673026337694e-06,
+      "loss": 0.0973,
+      "step": 26000
+    },
+    {
+      "epoch": 1.5642792927779443,
+      "grad_norm": 0.3808630108833313,
+      "learning_rate": 2.4206705956780877e-06,
+      "loss": 0.0983,
+      "step": 26100
+    },
+    {
+      "epoch": 1.570272700029967,
+      "grad_norm": 0.07692461460828781,
+      "learning_rate": 2.3873738887224055e-06,
+      "loss": 0.0971,
+      "step": 26200
+    },
+    {
+      "epoch": 1.57626610728199,
+      "grad_norm": 0.0316767655313015,
+      "learning_rate": 2.3540771817667234e-06,
+      "loss": 0.0969,
+      "step": 26300
+    },
+    {
+      "epoch": 1.5822595145340126,
+      "grad_norm": 0.054147470742464066,
+      "learning_rate": 2.3207804748110412e-06,
+      "loss": 0.0987,
+      "step": 26400
+    },
+    {
+      "epoch": 1.5882529217860353,
+      "grad_norm": 0.021603476256132126,
+      "learning_rate": 2.2874837678553595e-06,
+      "loss": 0.0993,
+      "step": 26500
+    },
+    {
+      "epoch": 1.5942463290380582,
+      "grad_norm": 0.008439415134489536,
+      "learning_rate": 2.2541870608996773e-06,
+      "loss": 0.0961,
+      "step": 26600
+    },
+    {
+      "epoch": 1.6002397362900809,
+      "grad_norm": 0.019180960953235626,
+      "learning_rate": 2.220890353943995e-06,
+      "loss": 0.0973,
+      "step": 26700
+    },
+    {
+      "epoch": 1.6062331435421036,
+      "grad_norm": 0.0572553388774395,
+      "learning_rate": 2.187593646988313e-06,
+      "loss": 0.0985,
+      "step": 26800
+    },
+    {
+      "epoch": 1.6122265507941265,
+      "grad_norm": 0.06225167587399483,
+      "learning_rate": 2.154296940032631e-06,
+      "loss": 0.0995,
+      "step": 26900
+    },
+    {
+      "epoch": 1.6182199580461494,
+      "grad_norm": 0.024926872923970222,
+      "learning_rate": 2.1210002330769487e-06,
+      "loss": 0.0981,
+      "step": 27000
+    },
+    {
+      "epoch": 1.624213365298172,
+      "grad_norm": 0.04497619345784187,
+      "learning_rate": 2.0877035261212665e-06,
+      "loss": 0.0983,
+      "step": 27100
+    },
+    {
+      "epoch": 1.6302067725501947,
+      "grad_norm": 0.008968952111899853,
+      "learning_rate": 2.054406819165585e-06,
+      "loss": 0.0989,
+      "step": 27200
+    },
+    {
+      "epoch": 1.6362001798022177,
+      "grad_norm": 0.5678277611732483,
+      "learning_rate": 2.0211101122099027e-06,
+      "loss": 0.0977,
+      "step": 27300
+    },
+    {
+      "epoch": 1.6421935870542403,
+      "grad_norm": 0.006168660242110491,
+      "learning_rate": 1.9878134052542205e-06,
+      "loss": 0.0987,
+      "step": 27400
+    },
+    {
+      "epoch": 1.648186994306263,
+      "grad_norm": 0.012832088395953178,
+      "learning_rate": 1.9545166982985383e-06,
+      "loss": 0.0977,
+      "step": 27500
+    },
+    {
+      "epoch": 1.654180401558286,
+      "grad_norm": 0.019368445500731468,
+      "learning_rate": 1.9212199913428566e-06,
+      "loss": 0.0983,
+      "step": 27600
+    },
+    {
+      "epoch": 1.6601738088103086,
+      "grad_norm": 0.014382258988916874,
+      "learning_rate": 1.8879232843871742e-06,
+      "loss": 0.0983,
+      "step": 27700
+    },
+    {
+      "epoch": 1.6661672160623313,
+      "grad_norm": 0.020989634096622467,
+      "learning_rate": 1.8546265774314923e-06,
+      "loss": 0.098,
+      "step": 27800
+    },
+    {
+      "epoch": 1.6721606233143542,
+      "grad_norm": 0.06429729610681534,
+      "learning_rate": 1.8213298704758101e-06,
+      "loss": 0.0983,
+      "step": 27900
+    },
+    {
+      "epoch": 1.6781540305663771,
+      "grad_norm": 0.06610631942749023,
+      "learning_rate": 1.7880331635201282e-06,
+      "loss": 0.0982,
+      "step": 28000
+    },
+    {
+      "epoch": 1.6841474378183996,
+      "grad_norm": 0.018632225692272186,
+      "learning_rate": 1.7547364565644458e-06,
+      "loss": 0.0981,
+      "step": 28100
+    },
+    {
+      "epoch": 1.6901408450704225,
+      "grad_norm": 0.0523594506084919,
+      "learning_rate": 1.7214397496087637e-06,
+      "loss": 0.0984,
+      "step": 28200
+    },
+    {
+      "epoch": 1.6961342523224454,
+      "grad_norm": 0.2683553993701935,
+      "learning_rate": 1.6881430426530817e-06,
+      "loss": 0.0982,
+      "step": 28300
+    },
+    {
+      "epoch": 1.702127659574468,
+      "grad_norm": 0.0381014347076416,
+      "learning_rate": 1.6548463356973996e-06,
+      "loss": 0.0982,
+      "step": 28400
+    },
+    {
+      "epoch": 1.7081210668264908,
+      "grad_norm": 0.0280121061950922,
+      "learning_rate": 1.6215496287417176e-06,
+      "loss": 0.0993,
+      "step": 28500
+    },
+    {
+      "epoch": 1.7141144740785137,
+      "grad_norm": 0.028142858296632767,
+      "learning_rate": 1.5882529217860355e-06,
+      "loss": 0.0979,
+      "step": 28600
+    },
+    {
+      "epoch": 1.7201078813305364,
+      "grad_norm": 0.17541886866092682,
+      "learning_rate": 1.5549562148303535e-06,
+      "loss": 0.0974,
+      "step": 28700
+    },
+    {
+      "epoch": 1.726101288582559,
+      "grad_norm": 0.02304765395820141,
+      "learning_rate": 1.5216595078746714e-06,
+      "loss": 0.0972,
+      "step": 28800
+    },
+    {
+      "epoch": 1.732094695834582,
+      "grad_norm": 0.039642009884119034,
+      "learning_rate": 1.4883628009189894e-06,
+      "loss": 0.0982,
+      "step": 28900
+    },
+    {
+      "epoch": 1.7380881030866049,
+      "grad_norm": 0.013521491549909115,
+      "learning_rate": 1.455066093963307e-06,
+      "loss": 0.0969,
+      "step": 29000
+    },
+    {
+      "epoch": 1.7440815103386273,
+      "grad_norm": 0.01796823740005493,
+      "learning_rate": 1.421769387007625e-06,
+      "loss": 0.0983,
+      "step": 29100
+    },
+    {
+      "epoch": 1.7500749175906503,
+      "grad_norm": 0.013001805171370506,
+      "learning_rate": 1.388472680051943e-06,
+      "loss": 0.0973,
+      "step": 29200
+    },
+    {
+      "epoch": 1.7560683248426732,
+      "grad_norm": 0.16073040664196014,
+      "learning_rate": 1.3551759730962608e-06,
+      "loss": 0.0977,
+      "step": 29300
+    },
+    {
+      "epoch": 1.7620617320946959,
+      "grad_norm": 0.0181583184748888,
+      "learning_rate": 1.3218792661405788e-06,
+      "loss": 0.098,
+      "step": 29400
+    },
+    {
+      "epoch": 1.7680551393467185,
+      "grad_norm": 0.0457487516105175,
+      "learning_rate": 1.2885825591848967e-06,
+      "loss": 0.0978,
+      "step": 29500
+    },
+    {
+      "epoch": 1.7740485465987414,
+      "grad_norm": 0.04452770575881004,
+      "learning_rate": 1.2552858522292147e-06,
+      "loss": 0.0982,
+      "step": 29600
+    },
+    {
+      "epoch": 1.7800419538507641,
+      "grad_norm": 0.5107665657997131,
+      "learning_rate": 1.2219891452735326e-06,
+      "loss": 0.0972,
+      "step": 29700
+    },
+    {
+      "epoch": 1.7860353611027868,
+      "grad_norm": 0.01403922587633133,
+      "learning_rate": 1.1886924383178504e-06,
+      "loss": 0.0986,
+      "step": 29800
+    },
+    {
+      "epoch": 1.7920287683548097,
+      "grad_norm": 0.020552983507514,
+      "learning_rate": 1.1553957313621683e-06,
+      "loss": 0.0989,
+      "step": 29900
+    },
+    {
+      "epoch": 1.7980221756068326,
+      "grad_norm": 0.011963661760091782,
+      "learning_rate": 1.1220990244064863e-06,
+      "loss": 0.098,
+      "step": 30000
+    }
+  ],
+  "logging_steps": 100,
+  "max_steps": 33370,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 2,
+  "save_steps": 10000,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 1.3389545272056584e+18,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:56160131670a4e6dc1e0982bdb220ed3ca17b733f3153925ac7e0c5c0019409c
+size 5240