submit

.gitattributes

@@ -33,3 +33,5 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+tokenizer.json filter=lfs diff=lfs merge=lfs -text
+g_00500000 filter=lfs diff=lfs merge=lfs -text

config.json

@@ -0,0 +1,46 @@
+{
+  "_name_or_path": "/linzhihang/zhangyuhao/ACLlama_s2s/output/S2S/S2S_finetune_small_lr3e-5_magpie-slice-unit-languge-new-0717-filter_kd_offline_base_merge_0717",
+  "adapter_size": 1280,
+  "architectures": [
+    "ACLlamaForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "audio_token_len": 10,
+  "audio_tower": "/linzhihang/LLMs/whisper-v3",
+  "bos_token_id": 128000,
+  "eos_token_id": [
+    128001,
+    128008,
+    128009
+  ],
+  "head_dim": 128,
+  "hidden_act": "silu",
+  "hidden_size": 3072,
+  "initializer_range": 0.02,
+  "intermediate_size": 8192,
+  "max_position_embeddings": 131072,
+  "mlp_bias": false,
+  "model_type": "ACLlama",
+  "num_attention_heads": 24,
+  "num_hidden_layers": 28,
+  "num_key_value_heads": 8,
+  "pretraining_tp": 1,
+  "rms_norm_eps": 1e-05,
+  "rope_scaling": {
+    "factor": 32.0,
+    "high_freq_factor": 4.0,
+    "low_freq_factor": 1.0,
+    "original_max_position_embeddings": 8192,
+    "rope_type": "llama3"
+  },
+  "rope_theta": 500000.0,
+  "tie_word_embeddings": true,
+  "torch_dtype": "float16",
+  "transformers_version": "4.48.1",
+  "unit_output": "finetune_kd",
+  "unit_vocab": "/linzhihang/zhangyuhao/unit_language/v5.8k.dict",
+  "unit_vocab_size": 7801,
+  "use_cache": false,
+  "vocab_size": 128257
+}

finetune_lora_llama3.1_8B_Instruct_s2s_zyh_3B.sh

@@ -0,0 +1,91 @@
+source /linzhihang/conda_env/init.sh
+conda activate s2s
+cd /linzhihang/zhangyuhao/ACLlama_s2s/scripts
+
+prefix=/linzhihang/zhangyuhao/ACLlama_s2s
+
+# NAME
+TASK=S2S
+stage=finetune  # edit in config
+model_size=small    # edit in config
+lr=3e-5
+# subtag=shareEmbW_CR_0507_base_Echox_s2s_pretrained_0503
+# subtag=FirstTurnS2T+aligner_0505
+# subtag=QA_OneTurn+aligner_Lora_0510
+# subtag=ASR_UnitLanguage_4gram_BPE+aligner_0513
+# subtag=QA_OneTurn_ALL_Lora_0516
+# subtag=QA_OneTurn_ALL_Lora_0517
+#subtag=QA_OneTurn_ALL_Lora_0618_newGen_80k_spm_epoch10-embedcon-10epoch-large-adapter-add-prefix
+subtag=kd_offline_base_merge_0819_bench
+
+
+
+# base_model=/mnt/speech/zhangyuhao/text_to_speech/ACLlama_t2u/Echox_s2s_0516 # unit
+#base_model=/linzhihang/zhangyuhao/ACLlama_s2s/Echox_s2s_unit_language_0529 # unit language
+#base_model=/linzhihang/zhangyuhao/ACLlama_s2s/Echox_s2s_unit_0706 # unit language
+base_model=/linzhihang/zhangyuhao/ACLlama_s2s/output/S2S/S2S_finetune_small_lr3e-5_magpie-slice-unit-languge-new-0717-filter_kd_offline_base_merge_0717
+#/linzhihang/zhangyuhao/ACLlama_s2s/output/S2S/S2S_finetune_small_lr3e-5_S2S-KD-unit-new-0816-filter-3B_kd_offline_base_merge_0816
+
+# DATA
+
+# data_json=/linzhihang/zhangyuhao/ACLlama_s2s/data/magpie_wer-filter-kd-40k-echo-ul-spm.json
+#data_json=/linzhihang/zhangyuhao/ACLlama_s2s/data/magpie-slice-unit-languge-new-0618.json
+#data_json=/linzhihang/zhangyuhao/ACLlama_s2s/data/magpie-slice-unit-languge-new-0629-filter.json
+#data_json=/linzhihang/zhangyuhao/ACLlama_s2s/data/magpie-slice-unit-languge-new-0706-filter.json
+data_json=/linzhihang/zhangyuhao/ACLlama_s2s/data/S2S-KD-unit-language-new-0819-3B-bench-filter.json
+#S2S-KD-unit-new-0819-3B-bench-filter.json
+#S2S-KD-unit-new-0816-filter-3B.json
+#magpie-slice-unit-languge-new-0717-filter.json
+
+
+
+training_set=${TASK}_${stage}_${model_size}_lr${lr}
+model_tag="${training_set}_$(basename "$data_json" .json)_${subtag}"
+
+checkpoint_dir=$prefix/output/$TASK/$model_tag
+echo $checkpoint_dir
+mkdir -p $checkpoint_dir
+cp $0 $checkpoint_dir/
+
+# CMD
+NCCL_P2P_DISABLE=1 \
+NCCL_IB_DISABLE=1 \
+CUDA_VISIBLE_DEVICES=0,1,2,3,4,5,6,7 \
+torchrun \
+--nproc_per_node 8 \
+--nnodes 1 \
+--node_rank 0 \
+--master_addr localhost \
+--master_port 7897 \
+$prefix/finetune_acllama_s2s_zyh.py \
+--audio_model_name_or_path "/linzhihang/LLMs/whisper-v3" \
+--text_model_name_or_path $base_model \
+--data_path "$data_json" \
+--fp16 True \
+--output_dir "$checkpoint_dir" \
+--num_train_epochs 1 \
+--per_device_train_batch_size 1 \
+--per_device_eval_batch_size 1 \
+--gradient_accumulation_steps 1 \
+--evaluation_strategy "no" \
+--save_strategy "steps" \
+--save_steps 200 \
+--save_total_limit 1 \
+--learning_rate $lr \
+--weight_decay 0.1 \
+--adam_beta2 0.95 \
+--warmup_ratio 0.01 \
+--lr_scheduler_type "inverse_sqrt" \
+--logging_steps 1 \
+--report_to "none" \
+--model_max_length 1024 \
+--gradient_checkpointing True \
+--lazy_preprocess True \
+--deepspeed "$prefix/config/ds_config_zero2.json" #\
+#--use_lora #> $checkpoint_dir/train.log # 2>&1
+
+# --use_lora
+#--data_path "$prefix/data/libri_train_update.json" \
+#--text_model_name_or_path "/mnt/user/zhangyuhao/LLM/llama3-instruct/llama3_1-8B/" \
+#--data_path "../data/libri_train_other460.json" \
+#--data_path "../data/train_mt_orgnize.json" \

g_00500000

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0d1f7188b95b06304bc05e524fddf93c7fe682fdd93acff022685663a5e26b97
+size 54051213

generation_config.json

@@ -0,0 +1,12 @@
+{
+  "bos_token_id": 128000,
+  "do_sample": true,
+  "eos_token_id": [
+    128001,
+    128008,
+    128009
+  ],
+  "temperature": 0.6,
+  "top_p": 0.9,
+  "transformers_version": "4.48.1"
+}

model-00001-of-00002.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:28a9f887c592e957a5cd31b9bdfb0386d7042e4dceb5a861ff4ce80507369f2f
+size 4965805056

model-00002-of-00002.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e621a936f89e828103064bc348ac0f4c4224d8c87c74fba3f5bc7ac719fa0b06
+size 3619536920

model.safetensors.index.json

@@ -0,0 +1,348 @@
+{
+  "metadata": {
+    "total_size": 8585302016
+  },
+  "weight_map": {
+    "model.asr_transformer_encoder.layers.0.linear1.bias": "model-00002-of-00002.safetensors",
+    "model.asr_transformer_encoder.layers.0.linear1.weight": "model-00002-of-00002.safetensors",
+    "model.asr_transformer_encoder.layers.0.linear2.bias": "model-00002-of-00002.safetensors",
+    "model.asr_transformer_encoder.layers.0.linear2.weight": "model-00002-of-00002.safetensors",
+    "model.asr_transformer_encoder.layers.0.norm1.bias": "model-00002-of-00002.safetensors",
+    "model.asr_transformer_encoder.layers.0.norm1.weight": "model-00002-of-00002.safetensors",
+    "model.asr_transformer_encoder.layers.0.norm2.bias": "model-00002-of-00002.safetensors",
+    "model.asr_transformer_encoder.layers.0.norm2.weight": "model-00002-of-00002.safetensors",
+    "model.asr_transformer_encoder.layers.0.self_attn.in_proj_bias": "model-00002-of-00002.safetensors",
+    "model.asr_transformer_encoder.layers.0.self_attn.in_proj_weight": "model-00002-of-00002.safetensors",
+    "model.asr_transformer_encoder.layers.0.self_attn.out_proj.bias": "model-00002-of-00002.safetensors",
+    "model.asr_transformer_encoder.layers.0.self_attn.out_proj.weight": "model-00002-of-00002.safetensors",
+    "model.audio_feature_head.weight": "model-00002-of-00002.safetensors",
+    "model.embed_tokens.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.20.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.20.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.20.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.21.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.21.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.21.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.21.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.21.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.21.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.21.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.21.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.21.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.22.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.22.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.22.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.22.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.22.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.22.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.22.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.22.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.22.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.23.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.23.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.23.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.23.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.23.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.23.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.23.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.23.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.23.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.24.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.24.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.24.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.24.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.24.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.24.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.24.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.24.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.24.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.25.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.25.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.25.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.25.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.25.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.25.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.25.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.25.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.25.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.26.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.26.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.26.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.26.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.26.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.26.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.26.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.26.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.26.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.27.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.27.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.27.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.27.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.27.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.27.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.27.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.27.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.27.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.3.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.lbm.atten_layer_norm.bias": "model-00002-of-00002.safetensors",
+    "model.lbm.atten_layer_norm.weight": "model-00002-of-00002.safetensors",
+    "model.lbm.encoder_attn.in_proj_bias": "model-00002-of-00002.safetensors",
+    "model.lbm.encoder_attn.in_proj_weight": "model-00002-of-00002.safetensors",
+    "model.lbm.encoder_attn.out_proj.bias": "model-00002-of-00002.safetensors",
+    "model.lbm.encoder_attn.out_proj.weight": "model-00002-of-00002.safetensors",
+    "model.mm_projector1.bias": "model-00002-of-00002.safetensors",
+    "model.mm_projector1.weight": "model-00002-of-00002.safetensors",
+    "model.norm.weight": "model-00002-of-00002.safetensors",
+    "model.out_norm.bias": "model-00002-of-00002.safetensors",
+    "model.out_norm.weight": "model-00002-of-00002.safetensors",
+    "unit_translator.adapter.bias": "model-00002-of-00002.safetensors",
+    "unit_translator.adapter.weight": "model-00002-of-00002.safetensors",
+    "unit_translator.aligner_MLP.0.bias": "model-00002-of-00002.safetensors",
+    "unit_translator.aligner_MLP.0.weight": "model-00002-of-00002.safetensors",
+    "unit_translator.aligner_MLP.3.bias": "model-00002-of-00002.safetensors",
+    "unit_translator.aligner_MLP.3.weight": "model-00002-of-00002.safetensors",
+    "unit_translator.lm_head.weight": "model-00002-of-00002.safetensors",
+    "unit_translator.model.embed_tokens.weight": "model-00002-of-00002.safetensors",
+    "unit_translator.model.layers.0.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "unit_translator.model.layers.0.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "unit_translator.model.layers.0.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+    "unit_translator.model.layers.0.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+    "unit_translator.model.layers.0.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "unit_translator.model.layers.0.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "unit_translator.model.layers.0.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "unit_translator.model.layers.0.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "unit_translator.model.layers.0.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "unit_translator.model.layers.1.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "unit_translator.model.layers.1.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "unit_translator.model.layers.1.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+    "unit_translator.model.layers.1.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+    "unit_translator.model.layers.1.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "unit_translator.model.layers.1.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "unit_translator.model.layers.1.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "unit_translator.model.layers.1.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "unit_translator.model.layers.1.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "unit_translator.model.layers.2.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "unit_translator.model.layers.2.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "unit_translator.model.layers.2.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+    "unit_translator.model.layers.2.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+    "unit_translator.model.layers.2.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "unit_translator.model.layers.2.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "unit_translator.model.layers.2.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "unit_translator.model.layers.2.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "unit_translator.model.layers.2.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "unit_translator.model.layers.3.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "unit_translator.model.layers.3.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "unit_translator.model.layers.3.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+    "unit_translator.model.layers.3.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+    "unit_translator.model.layers.3.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "unit_translator.model.layers.3.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "unit_translator.model.layers.3.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "unit_translator.model.layers.3.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "unit_translator.model.layers.3.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "unit_translator.model.layers.4.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "unit_translator.model.layers.4.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "unit_translator.model.layers.4.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+    "unit_translator.model.layers.4.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+    "unit_translator.model.layers.4.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "unit_translator.model.layers.4.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "unit_translator.model.layers.4.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "unit_translator.model.layers.4.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "unit_translator.model.layers.4.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "unit_translator.model.layers.5.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "unit_translator.model.layers.5.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "unit_translator.model.layers.5.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+    "unit_translator.model.layers.5.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+    "unit_translator.model.layers.5.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "unit_translator.model.layers.5.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "unit_translator.model.layers.5.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "unit_translator.model.layers.5.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "unit_translator.model.layers.5.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "unit_translator.model.norm.weight": "model-00002-of-00002.safetensors",
+    "unit_translator.unit_embedding.weight": "model-00002-of-00002.safetensors"
+  }
+}

special_tokens_map.json

@@ -0,0 +1,23 @@
+{
+  "bos_token": {
+    "content": "<|begin_of_text|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<|eot_id|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<|eot_id|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:856f538e5cb73db3731eb9f10efb325225743abbfd32d20180383a53e135c349
+size 18733690

trainer_state.json

@@ -0,0 +1,2618 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 1.0,
+  "eval_steps": 500,
+  "global_step": 368,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.002717391304347826,
+      "grad_norm": 0.0,
+      "learning_rate": 0,
+      "loss": 0.5829,
+      "step": 1
+    },
+    {
+      "epoch": 0.005434782608695652,
+      "grad_norm": 0.0,
+      "learning_rate": 0,
+      "loss": 0.685,
+      "step": 2
+    },
+    {
+      "epoch": 0.008152173913043478,
+      "grad_norm": 3.0868945121765137,
+      "learning_rate": 0.0,
+      "loss": 0.7212,
+      "step": 3
+    },
+    {
+      "epoch": 0.010869565217391304,
+      "grad_norm": 3.4650824069976807,
+      "learning_rate": 1.5e-05,
+      "loss": 0.6836,
+      "step": 4
+    },
+    {
+      "epoch": 0.01358695652173913,
+      "grad_norm": 3.4650824069976807,
+      "learning_rate": 1.5e-05,
+      "loss": 0.5367,
+      "step": 5
+    },
+    {
+      "epoch": 0.016304347826086956,
+      "grad_norm": 2.637908935546875,
+      "learning_rate": 2.3774437510817346e-05,
+      "loss": 0.6148,
+      "step": 6
+    },
+    {
+      "epoch": 0.019021739130434784,
+      "grad_norm": 1.4185562133789062,
+      "learning_rate": 3e-05,
+      "loss": 0.562,
+      "step": 7
+    },
+    {
+      "epoch": 0.021739130434782608,
+      "grad_norm": 3.299220323562622,
+      "learning_rate": 3e-05,
+      "loss": 0.6763,
+      "step": 8
+    },
+    {
+      "epoch": 0.024456521739130436,
+      "grad_norm": 3.9320366382598877,
+      "learning_rate": 3e-05,
+      "loss": 0.4243,
+      "step": 9
+    },
+    {
+      "epoch": 0.02717391304347826,
+      "grad_norm": 4.625139236450195,
+      "learning_rate": 3e-05,
+      "loss": 0.6627,
+      "step": 10
+    },
+    {
+      "epoch": 0.029891304347826088,
+      "grad_norm": 4.016164779663086,
+      "learning_rate": 3e-05,
+      "loss": 0.5826,
+      "step": 11
+    },
+    {
+      "epoch": 0.03260869565217391,
+      "grad_norm": 3.5085506439208984,
+      "learning_rate": 3e-05,
+      "loss": 0.7494,
+      "step": 12
+    },
+    {
+      "epoch": 0.035326086956521736,
+      "grad_norm": 3.135777235031128,
+      "learning_rate": 3e-05,
+      "loss": 0.5124,
+      "step": 13
+    },
+    {
+      "epoch": 0.03804347826086957,
+      "grad_norm": 2.25691294670105,
+      "learning_rate": 3e-05,
+      "loss": 0.5156,
+      "step": 14
+    },
+    {
+      "epoch": 0.04076086956521739,
+      "grad_norm": 2.6235625743865967,
+      "learning_rate": 3e-05,
+      "loss": 0.5773,
+      "step": 15
+    },
+    {
+      "epoch": 0.043478260869565216,
+      "grad_norm": 1.977939248085022,
+      "learning_rate": 3e-05,
+      "loss": 0.5534,
+      "step": 16
+    },
+    {
+      "epoch": 0.04619565217391304,
+      "grad_norm": 3.1098241806030273,
+      "learning_rate": 3e-05,
+      "loss": 0.7569,
+      "step": 17
+    },
+    {
+      "epoch": 0.04891304347826087,
+      "grad_norm": 3.5367255210876465,
+      "learning_rate": 3e-05,
+      "loss": 0.6384,
+      "step": 18
+    },
+    {
+      "epoch": 0.051630434782608696,
+      "grad_norm": 2.057596445083618,
+      "learning_rate": 3e-05,
+      "loss": 0.5811,
+      "step": 19
+    },
+    {
+      "epoch": 0.05434782608695652,
+      "grad_norm": 1.8991059064865112,
+      "learning_rate": 3e-05,
+      "loss": 0.4392,
+      "step": 20
+    },
+    {
+      "epoch": 0.057065217391304345,
+      "grad_norm": 2.431248188018799,
+      "learning_rate": 3e-05,
+      "loss": 0.5292,
+      "step": 21
+    },
+    {
+      "epoch": 0.059782608695652176,
+      "grad_norm": 2.1876354217529297,
+      "learning_rate": 3e-05,
+      "loss": 0.4795,
+      "step": 22
+    },
+    {
+      "epoch": 0.0625,
+      "grad_norm": 2.018975257873535,
+      "learning_rate": 3e-05,
+      "loss": 0.5681,
+      "step": 23
+    },
+    {
+      "epoch": 0.06521739130434782,
+      "grad_norm": 1.5382078886032104,
+      "learning_rate": 3e-05,
+      "loss": 0.4643,
+      "step": 24
+    },
+    {
+      "epoch": 0.06793478260869565,
+      "grad_norm": 2.18375563621521,
+      "learning_rate": 3e-05,
+      "loss": 0.4675,
+      "step": 25
+    },
+    {
+      "epoch": 0.07065217391304347,
+      "grad_norm": 1.810476303100586,
+      "learning_rate": 3e-05,
+      "loss": 0.5259,
+      "step": 26
+    },
+    {
+      "epoch": 0.07336956521739131,
+      "grad_norm": 2.9333322048187256,
+      "learning_rate": 3e-05,
+      "loss": 0.6795,
+      "step": 27
+    },
+    {
+      "epoch": 0.07608695652173914,
+      "grad_norm": 1.963409185409546,
+      "learning_rate": 3e-05,
+      "loss": 0.5022,
+      "step": 28
+    },
+    {
+      "epoch": 0.07880434782608696,
+      "grad_norm": 1.5288830995559692,
+      "learning_rate": 3e-05,
+      "loss": 0.5481,
+      "step": 29
+    },
+    {
+      "epoch": 0.08152173913043478,
+      "grad_norm": 2.095547676086426,
+      "learning_rate": 3e-05,
+      "loss": 0.5427,
+      "step": 30
+    },
+    {
+      "epoch": 0.08423913043478261,
+      "grad_norm": 2.3924880027770996,
+      "learning_rate": 3e-05,
+      "loss": 0.5936,
+      "step": 31
+    },
+    {
+      "epoch": 0.08695652173913043,
+      "grad_norm": 2.6183862686157227,
+      "learning_rate": 3e-05,
+      "loss": 0.5452,
+      "step": 32
+    },
+    {
+      "epoch": 0.08967391304347826,
+      "grad_norm": 1.9050800800323486,
+      "learning_rate": 3e-05,
+      "loss": 0.4507,
+      "step": 33
+    },
+    {
+      "epoch": 0.09239130434782608,
+      "grad_norm": 2.0894551277160645,
+      "learning_rate": 3e-05,
+      "loss": 0.6293,
+      "step": 34
+    },
+    {
+      "epoch": 0.09510869565217392,
+      "grad_norm": 2.201720952987671,
+      "learning_rate": 3e-05,
+      "loss": 0.4793,
+      "step": 35
+    },
+    {
+      "epoch": 0.09782608695652174,
+      "grad_norm": 2.516624927520752,
+      "learning_rate": 3e-05,
+      "loss": 0.5658,
+      "step": 36
+    },
+    {
+      "epoch": 0.10054347826086957,
+      "grad_norm": 1.703904628753662,
+      "learning_rate": 3e-05,
+      "loss": 0.4968,
+      "step": 37
+    },
+    {
+      "epoch": 0.10326086956521739,
+      "grad_norm": 2.972416400909424,
+      "learning_rate": 3e-05,
+      "loss": 0.7267,
+      "step": 38
+    },
+    {
+      "epoch": 0.10597826086956522,
+      "grad_norm": 1.9167985916137695,
+      "learning_rate": 3e-05,
+      "loss": 0.4971,
+      "step": 39
+    },
+    {
+      "epoch": 0.10869565217391304,
+      "grad_norm": 1.7569814920425415,
+      "learning_rate": 3e-05,
+      "loss": 0.5191,
+      "step": 40
+    },
+    {
+      "epoch": 0.11141304347826086,
+      "grad_norm": 2.4087071418762207,
+      "learning_rate": 3e-05,
+      "loss": 0.5737,
+      "step": 41
+    },
+    {
+      "epoch": 0.11413043478260869,
+      "grad_norm": 3.517263650894165,
+      "learning_rate": 3e-05,
+      "loss": 0.5688,
+      "step": 42
+    },
+    {
+      "epoch": 0.11684782608695653,
+      "grad_norm": 3.5090434551239014,
+      "learning_rate": 3e-05,
+      "loss": 0.4961,
+      "step": 43
+    },
+    {
+      "epoch": 0.11956521739130435,
+      "grad_norm": 3.181145668029785,
+      "learning_rate": 3e-05,
+      "loss": 0.8124,
+      "step": 44
+    },
+    {
+      "epoch": 0.12228260869565218,
+      "grad_norm": 3.588538885116577,
+      "learning_rate": 3e-05,
+      "loss": 0.5766,
+      "step": 45
+    },
+    {
+      "epoch": 0.125,
+      "grad_norm": 3.123659610748291,
+      "learning_rate": 3e-05,
+      "loss": 0.7286,
+      "step": 46
+    },
+    {
+      "epoch": 0.12771739130434784,
+      "grad_norm": 2.6902639865875244,
+      "learning_rate": 3e-05,
+      "loss": 0.6418,
+      "step": 47
+    },
+    {
+      "epoch": 0.13043478260869565,
+      "grad_norm": 1.6611530780792236,
+      "learning_rate": 3e-05,
+      "loss": 0.5357,
+      "step": 48
+    },
+    {
+      "epoch": 0.1331521739130435,
+      "grad_norm": 2.4058802127838135,
+      "learning_rate": 3e-05,
+      "loss": 0.5909,
+      "step": 49
+    },
+    {
+      "epoch": 0.1358695652173913,
+      "grad_norm": 2.4058802127838135,
+      "learning_rate": 3e-05,
+      "loss": 0.7068,
+      "step": 50
+    },
+    {
+      "epoch": 0.13858695652173914,
+      "grad_norm": 2.513307809829712,
+      "learning_rate": 3e-05,
+      "loss": 0.5384,
+      "step": 51
+    },
+    {
+      "epoch": 0.14130434782608695,
+      "grad_norm": 2.7813773155212402,
+      "learning_rate": 3e-05,
+      "loss": 0.5887,
+      "step": 52
+    },
+    {
+      "epoch": 0.14402173913043478,
+      "grad_norm": 1.950292706489563,
+      "learning_rate": 3e-05,
+      "loss": 0.5399,
+      "step": 53
+    },
+    {
+      "epoch": 0.14673913043478262,
+      "grad_norm": 1.7724437713623047,
+      "learning_rate": 3e-05,
+      "loss": 0.4234,
+      "step": 54
+    },
+    {
+      "epoch": 0.14945652173913043,
+      "grad_norm": 3.3404788970947266,
+      "learning_rate": 3e-05,
+      "loss": 0.7648,
+      "step": 55
+    },
+    {
+      "epoch": 0.15217391304347827,
+      "grad_norm": 2.424994468688965,
+      "learning_rate": 3e-05,
+      "loss": 0.4959,
+      "step": 56
+    },
+    {
+      "epoch": 0.15489130434782608,
+      "grad_norm": 2.12221097946167,
+      "learning_rate": 3e-05,
+      "loss": 0.6107,
+      "step": 57
+    },
+    {
+      "epoch": 0.15760869565217392,
+      "grad_norm": 3.7846357822418213,
+      "learning_rate": 3e-05,
+      "loss": 0.6991,
+      "step": 58
+    },
+    {
+      "epoch": 0.16032608695652173,
+      "grad_norm": 1.7821156978607178,
+      "learning_rate": 3e-05,
+      "loss": 0.4864,
+      "step": 59
+    },
+    {
+      "epoch": 0.16304347826086957,
+      "grad_norm": 1.4435001611709595,
+      "learning_rate": 3e-05,
+      "loss": 0.4782,
+      "step": 60
+    },
+    {
+      "epoch": 0.16576086956521738,
+      "grad_norm": 2.097719669342041,
+      "learning_rate": 3e-05,
+      "loss": 0.4761,
+      "step": 61
+    },
+    {
+      "epoch": 0.16847826086956522,
+      "grad_norm": 2.082066774368286,
+      "learning_rate": 3e-05,
+      "loss": 0.4565,
+      "step": 62
+    },
+    {
+      "epoch": 0.17119565217391305,
+      "grad_norm": 1.8222287893295288,
+      "learning_rate": 3e-05,
+      "loss": 0.5573,
+      "step": 63
+    },
+    {
+      "epoch": 0.17391304347826086,
+      "grad_norm": 2.819382667541504,
+      "learning_rate": 3e-05,
+      "loss": 0.5613,
+      "step": 64
+    },
+    {
+      "epoch": 0.1766304347826087,
+      "grad_norm": 1.8486182689666748,
+      "learning_rate": 3e-05,
+      "loss": 0.5446,
+      "step": 65
+    },
+    {
+      "epoch": 0.1793478260869565,
+      "grad_norm": 1.4942399263381958,
+      "learning_rate": 3e-05,
+      "loss": 0.5744,
+      "step": 66
+    },
+    {
+      "epoch": 0.18206521739130435,
+      "grad_norm": 1.9929646253585815,
+      "learning_rate": 3e-05,
+      "loss": 0.4258,
+      "step": 67
+    },
+    {
+      "epoch": 0.18478260869565216,
+      "grad_norm": 1.7978485822677612,
+      "learning_rate": 3e-05,
+      "loss": 0.4694,
+      "step": 68
+    },
+    {
+      "epoch": 0.1875,
+      "grad_norm": 2.185476064682007,
+      "learning_rate": 3e-05,
+      "loss": 0.5109,
+      "step": 69
+    },
+    {
+      "epoch": 0.19021739130434784,
+      "grad_norm": 2.129399538040161,
+      "learning_rate": 3e-05,
+      "loss": 0.4594,
+      "step": 70
+    },
+    {
+      "epoch": 0.19293478260869565,
+      "grad_norm": 2.488927125930786,
+      "learning_rate": 3e-05,
+      "loss": 0.5293,
+      "step": 71
+    },
+    {
+      "epoch": 0.1956521739130435,
+      "grad_norm": 2.257550001144409,
+      "learning_rate": 3e-05,
+      "loss": 0.5494,
+      "step": 72
+    },
+    {
+      "epoch": 0.1983695652173913,
+      "grad_norm": 2.4466147422790527,
+      "learning_rate": 3e-05,
+      "loss": 0.4834,
+      "step": 73
+    },
+    {
+      "epoch": 0.20108695652173914,
+      "grad_norm": 2.2242984771728516,
+      "learning_rate": 3e-05,
+      "loss": 0.4992,
+      "step": 74
+    },
+    {
+      "epoch": 0.20380434782608695,
+      "grad_norm": 2.867558002471924,
+      "learning_rate": 3e-05,
+      "loss": 0.7131,
+      "step": 75
+    },
+    {
+      "epoch": 0.20652173913043478,
+      "grad_norm": 1.8966355323791504,
+      "learning_rate": 3e-05,
+      "loss": 0.5512,
+      "step": 76
+    },
+    {
+      "epoch": 0.20923913043478262,
+      "grad_norm": 3.1431808471679688,
+      "learning_rate": 3e-05,
+      "loss": 0.4581,
+      "step": 77
+    },
+    {
+      "epoch": 0.21195652173913043,
+      "grad_norm": 3.1641836166381836,
+      "learning_rate": 3e-05,
+      "loss": 0.56,
+      "step": 78
+    },
+    {
+      "epoch": 0.21467391304347827,
+      "grad_norm": 1.869903326034546,
+      "learning_rate": 3e-05,
+      "loss": 0.5766,
+      "step": 79
+    },
+    {
+      "epoch": 0.21739130434782608,
+      "grad_norm": 2.7795424461364746,
+      "learning_rate": 3e-05,
+      "loss": 0.5911,
+      "step": 80
+    },
+    {
+      "epoch": 0.22010869565217392,
+      "grad_norm": 2.3112289905548096,
+      "learning_rate": 3e-05,
+      "loss": 0.5765,
+      "step": 81
+    },
+    {
+      "epoch": 0.22282608695652173,
+      "grad_norm": 1.8768959045410156,
+      "learning_rate": 3e-05,
+      "loss": 0.4489,
+      "step": 82
+    },
+    {
+      "epoch": 0.22554347826086957,
+      "grad_norm": 1.8480079174041748,
+      "learning_rate": 3e-05,
+      "loss": 0.6242,
+      "step": 83
+    },
+    {
+      "epoch": 0.22826086956521738,
+      "grad_norm": 2.04072904586792,
+      "learning_rate": 3e-05,
+      "loss": 0.5061,
+      "step": 84
+    },
+    {
+      "epoch": 0.23097826086956522,
+      "grad_norm": 1.778836965560913,
+      "learning_rate": 3e-05,
+      "loss": 0.5375,
+      "step": 85
+    },
+    {
+      "epoch": 0.23369565217391305,
+      "grad_norm": 2.0883471965789795,
+      "learning_rate": 3e-05,
+      "loss": 0.5375,
+      "step": 86
+    },
+    {
+      "epoch": 0.23641304347826086,
+      "grad_norm": 1.8592745065689087,
+      "learning_rate": 3e-05,
+      "loss": 0.4972,
+      "step": 87
+    },
+    {
+      "epoch": 0.2391304347826087,
+      "grad_norm": 1.817451000213623,
+      "learning_rate": 3e-05,
+      "loss": 0.5005,
+      "step": 88
+    },
+    {
+      "epoch": 0.2418478260869565,
+      "grad_norm": 3.28531813621521,
+      "learning_rate": 3e-05,
+      "loss": 0.6246,
+      "step": 89
+    },
+    {
+      "epoch": 0.24456521739130435,
+      "grad_norm": 1.707846760749817,
+      "learning_rate": 3e-05,
+      "loss": 0.4922,
+      "step": 90
+    },
+    {
+      "epoch": 0.24728260869565216,
+      "grad_norm": 2.3756139278411865,
+      "learning_rate": 3e-05,
+      "loss": 0.5826,
+      "step": 91
+    },
+    {
+      "epoch": 0.25,
+      "grad_norm": 1.2630126476287842,
+      "learning_rate": 3e-05,
+      "loss": 0.5304,
+      "step": 92
+    },
+    {
+      "epoch": 0.25271739130434784,
+      "grad_norm": 2.259389877319336,
+      "learning_rate": 3e-05,
+      "loss": 0.5615,
+      "step": 93
+    },
+    {
+      "epoch": 0.2554347826086957,
+      "grad_norm": 1.3903180360794067,
+      "learning_rate": 3e-05,
+      "loss": 0.4406,
+      "step": 94
+    },
+    {
+      "epoch": 0.25815217391304346,
+      "grad_norm": 1.9753227233886719,
+      "learning_rate": 3e-05,
+      "loss": 0.4809,
+      "step": 95
+    },
+    {
+      "epoch": 0.2608695652173913,
+      "grad_norm": 1.8999621868133545,
+      "learning_rate": 3e-05,
+      "loss": 0.5946,
+      "step": 96
+    },
+    {
+      "epoch": 0.26358695652173914,
+      "grad_norm": 1.9928959608078003,
+      "learning_rate": 3e-05,
+      "loss": 0.5532,
+      "step": 97
+    },
+    {
+      "epoch": 0.266304347826087,
+      "grad_norm": 1.7581721544265747,
+      "learning_rate": 3e-05,
+      "loss": 0.425,
+      "step": 98
+    },
+    {
+      "epoch": 0.26902173913043476,
+      "grad_norm": 3.1840996742248535,
+      "learning_rate": 3e-05,
+      "loss": 0.6684,
+      "step": 99
+    },
+    {
+      "epoch": 0.2717391304347826,
+      "grad_norm": 2.4350733757019043,
+      "learning_rate": 3e-05,
+      "loss": 0.5354,
+      "step": 100
+    },
+    {
+      "epoch": 0.27445652173913043,
+      "grad_norm": 2.1404078006744385,
+      "learning_rate": 3e-05,
+      "loss": 0.4751,
+      "step": 101
+    },
+    {
+      "epoch": 0.27717391304347827,
+      "grad_norm": 3.169942617416382,
+      "learning_rate": 3e-05,
+      "loss": 0.7145,
+      "step": 102
+    },
+    {
+      "epoch": 0.2798913043478261,
+      "grad_norm": 1.2857766151428223,
+      "learning_rate": 3e-05,
+      "loss": 0.4536,
+      "step": 103
+    },
+    {
+      "epoch": 0.2826086956521739,
+      "grad_norm": 2.7525951862335205,
+      "learning_rate": 3e-05,
+      "loss": 0.5892,
+      "step": 104
+    },
+    {
+      "epoch": 0.28532608695652173,
+      "grad_norm": 1.6672691106796265,
+      "learning_rate": 3e-05,
+      "loss": 0.5593,
+      "step": 105
+    },
+    {
+      "epoch": 0.28804347826086957,
+      "grad_norm": 1.2421808242797852,
+      "learning_rate": 3e-05,
+      "loss": 0.4584,
+      "step": 106
+    },
+    {
+      "epoch": 0.2907608695652174,
+      "grad_norm": 3.0662600994110107,
+      "learning_rate": 3e-05,
+      "loss": 0.564,
+      "step": 107
+    },
+    {
+      "epoch": 0.29347826086956524,
+      "grad_norm": 1.9544175863265991,
+      "learning_rate": 3e-05,
+      "loss": 0.5583,
+      "step": 108
+    },
+    {
+      "epoch": 0.296195652173913,
+      "grad_norm": 2.374974250793457,
+      "learning_rate": 3e-05,
+      "loss": 0.5546,
+      "step": 109
+    },
+    {
+      "epoch": 0.29891304347826086,
+      "grad_norm": 1.730804443359375,
+      "learning_rate": 3e-05,
+      "loss": 0.5021,
+      "step": 110
+    },
+    {
+      "epoch": 0.3016304347826087,
+      "grad_norm": 2.308568239212036,
+      "learning_rate": 3e-05,
+      "loss": 0.6143,
+      "step": 111
+    },
+    {
+      "epoch": 0.30434782608695654,
+      "grad_norm": 2.2530875205993652,
+      "learning_rate": 3e-05,
+      "loss": 0.5695,
+      "step": 112
+    },
+    {
+      "epoch": 0.3070652173913043,
+      "grad_norm": 1.4939433336257935,
+      "learning_rate": 3e-05,
+      "loss": 0.4982,
+      "step": 113
+    },
+    {
+      "epoch": 0.30978260869565216,
+      "grad_norm": 1.8290050029754639,
+      "learning_rate": 3e-05,
+      "loss": 0.4945,
+      "step": 114
+    },
+    {
+      "epoch": 0.3125,
+      "grad_norm": 2.5083842277526855,
+      "learning_rate": 3e-05,
+      "loss": 0.4693,
+      "step": 115
+    },
+    {
+      "epoch": 0.31521739130434784,
+      "grad_norm": 3.2338268756866455,
+      "learning_rate": 3e-05,
+      "loss": 0.6578,
+      "step": 116
+    },
+    {
+      "epoch": 0.3179347826086957,
+      "grad_norm": 2.3117527961730957,
+      "learning_rate": 3e-05,
+      "loss": 0.5395,
+      "step": 117
+    },
+    {
+      "epoch": 0.32065217391304346,
+      "grad_norm": 1.5958501100540161,
+      "learning_rate": 3e-05,
+      "loss": 0.4975,
+      "step": 118
+    },
+    {
+      "epoch": 0.3233695652173913,
+      "grad_norm": 2.4708805084228516,
+      "learning_rate": 3e-05,
+      "loss": 0.4652,
+      "step": 119
+    },
+    {
+      "epoch": 0.32608695652173914,
+      "grad_norm": 1.6753262281417847,
+      "learning_rate": 3e-05,
+      "loss": 0.54,
+      "step": 120
+    },
+    {
+      "epoch": 0.328804347826087,
+      "grad_norm": 2.4457895755767822,
+      "learning_rate": 3e-05,
+      "loss": 0.5612,
+      "step": 121
+    },
+    {
+      "epoch": 0.33152173913043476,
+      "grad_norm": 2.059176445007324,
+      "learning_rate": 3e-05,
+      "loss": 0.6599,
+      "step": 122
+    },
+    {
+      "epoch": 0.3342391304347826,
+      "grad_norm": 2.1220273971557617,
+      "learning_rate": 3e-05,
+      "loss": 0.5256,
+      "step": 123
+    },
+    {
+      "epoch": 0.33695652173913043,
+      "grad_norm": 1.756909728050232,
+      "learning_rate": 3e-05,
+      "loss": 0.4839,
+      "step": 124
+    },
+    {
+      "epoch": 0.33967391304347827,
+      "grad_norm": 2.6595497131347656,
+      "learning_rate": 3e-05,
+      "loss": 0.626,
+      "step": 125
+    },
+    {
+      "epoch": 0.3423913043478261,
+      "grad_norm": 2.2510428428649902,
+      "learning_rate": 3e-05,
+      "loss": 0.5029,
+      "step": 126
+    },
+    {
+      "epoch": 0.3451086956521739,
+      "grad_norm": 3.559722423553467,
+      "learning_rate": 3e-05,
+      "loss": 0.647,
+      "step": 127
+    },
+    {
+      "epoch": 0.34782608695652173,
+      "grad_norm": 2.1882569789886475,
+      "learning_rate": 3e-05,
+      "loss": 0.6679,
+      "step": 128
+    },
+    {
+      "epoch": 0.35054347826086957,
+      "grad_norm": 2.5354092121124268,
+      "learning_rate": 3e-05,
+      "loss": 0.6344,
+      "step": 129
+    },
+    {
+      "epoch": 0.3532608695652174,
+      "grad_norm": 2.0643749237060547,
+      "learning_rate": 3e-05,
+      "loss": 0.4314,
+      "step": 130
+    },
+    {
+      "epoch": 0.35597826086956524,
+      "grad_norm": 1.6011096239089966,
+      "learning_rate": 3e-05,
+      "loss": 0.4533,
+      "step": 131
+    },
+    {
+      "epoch": 0.358695652173913,
+      "grad_norm": 2.553201913833618,
+      "learning_rate": 3e-05,
+      "loss": 0.6543,
+      "step": 132
+    },
+    {
+      "epoch": 0.36141304347826086,
+      "grad_norm": 2.410280704498291,
+      "learning_rate": 3e-05,
+      "loss": 0.5308,
+      "step": 133
+    },
+    {
+      "epoch": 0.3641304347826087,
+      "grad_norm": 1.7064554691314697,
+      "learning_rate": 3e-05,
+      "loss": 0.4105,
+      "step": 134
+    },
+    {
+      "epoch": 0.36684782608695654,
+      "grad_norm": 2.6405141353607178,
+      "learning_rate": 3e-05,
+      "loss": 0.5206,
+      "step": 135
+    },
+    {
+      "epoch": 0.3695652173913043,
+      "grad_norm": 1.699704885482788,
+      "learning_rate": 3e-05,
+      "loss": 0.6593,
+      "step": 136
+    },
+    {
+      "epoch": 0.37228260869565216,
+      "grad_norm": 1.8388729095458984,
+      "learning_rate": 3e-05,
+      "loss": 0.4821,
+      "step": 137
+    },
+    {
+      "epoch": 0.375,
+      "grad_norm": 2.7615950107574463,
+      "learning_rate": 3e-05,
+      "loss": 0.5692,
+      "step": 138
+    },
+    {
+      "epoch": 0.37771739130434784,
+      "grad_norm": 1.835686206817627,
+      "learning_rate": 3e-05,
+      "loss": 0.5826,
+      "step": 139
+    },
+    {
+      "epoch": 0.3804347826086957,
+      "grad_norm": 1.6995174884796143,
+      "learning_rate": 3e-05,
+      "loss": 0.4742,
+      "step": 140
+    },
+    {
+      "epoch": 0.38315217391304346,
+      "grad_norm": 1.6208698749542236,
+      "learning_rate": 3e-05,
+      "loss": 0.4437,
+      "step": 141
+    },
+    {
+      "epoch": 0.3858695652173913,
+      "grad_norm": 1.2213523387908936,
+      "learning_rate": 3e-05,
+      "loss": 0.4037,
+      "step": 142
+    },
+    {
+      "epoch": 0.38858695652173914,
+      "grad_norm": 1.5828511714935303,
+      "learning_rate": 3e-05,
+      "loss": 0.4793,
+      "step": 143
+    },
+    {
+      "epoch": 0.391304347826087,
+      "grad_norm": 3.4537761211395264,
+      "learning_rate": 3e-05,
+      "loss": 0.4866,
+      "step": 144
+    },
+    {
+      "epoch": 0.39402173913043476,
+      "grad_norm": 1.346982479095459,
+      "learning_rate": 3e-05,
+      "loss": 0.4468,
+      "step": 145
+    },
+    {
+      "epoch": 0.3967391304347826,
+      "grad_norm": 3.6017796993255615,
+      "learning_rate": 3e-05,
+      "loss": 0.6744,
+      "step": 146
+    },
+    {
+      "epoch": 0.39945652173913043,
+      "grad_norm": 1.896022915840149,
+      "learning_rate": 3e-05,
+      "loss": 0.5816,
+      "step": 147
+    },
+    {
+      "epoch": 0.40217391304347827,
+      "grad_norm": 1.970211148262024,
+      "learning_rate": 3e-05,
+      "loss": 0.5449,
+      "step": 148
+    },
+    {
+      "epoch": 0.4048913043478261,
+      "grad_norm": 2.315774917602539,
+      "learning_rate": 3e-05,
+      "loss": 0.5042,
+      "step": 149
+    },
+    {
+      "epoch": 0.4076086956521739,
+      "grad_norm": 1.7873855829238892,
+      "learning_rate": 3e-05,
+      "loss": 0.5093,
+      "step": 150
+    },
+    {
+      "epoch": 0.41032608695652173,
+      "grad_norm": 2.479135036468506,
+      "learning_rate": 3e-05,
+      "loss": 0.6396,
+      "step": 151
+    },
+    {
+      "epoch": 0.41304347826086957,
+      "grad_norm": 1.4740537405014038,
+      "learning_rate": 3e-05,
+      "loss": 0.4847,
+      "step": 152
+    },
+    {
+      "epoch": 0.4157608695652174,
+      "grad_norm": 1.8202875852584839,
+      "learning_rate": 3e-05,
+      "loss": 0.565,
+      "step": 153
+    },
+    {
+      "epoch": 0.41847826086956524,
+      "grad_norm": 2.06355881690979,
+      "learning_rate": 3e-05,
+      "loss": 0.5481,
+      "step": 154
+    },
+    {
+      "epoch": 0.421195652173913,
+      "grad_norm": 1.522354245185852,
+      "learning_rate": 3e-05,
+      "loss": 0.4835,
+      "step": 155
+    },
+    {
+      "epoch": 0.42391304347826086,
+      "grad_norm": 1.9054198265075684,
+      "learning_rate": 3e-05,
+      "loss": 0.5304,
+      "step": 156
+    },
+    {
+      "epoch": 0.4266304347826087,
+      "grad_norm": 1.7483571767807007,
+      "learning_rate": 3e-05,
+      "loss": 0.4882,
+      "step": 157
+    },
+    {
+      "epoch": 0.42934782608695654,
+      "grad_norm": 1.8139662742614746,
+      "learning_rate": 3e-05,
+      "loss": 0.6033,
+      "step": 158
+    },
+    {
+      "epoch": 0.4320652173913043,
+      "grad_norm": 3.0108718872070312,
+      "learning_rate": 3e-05,
+      "loss": 0.837,
+      "step": 159
+    },
+    {
+      "epoch": 0.43478260869565216,
+      "grad_norm": 1.5665092468261719,
+      "learning_rate": 3e-05,
+      "loss": 0.5234,
+      "step": 160
+    },
+    {
+      "epoch": 0.4375,
+      "grad_norm": 3.6204946041107178,
+      "learning_rate": 3e-05,
+      "loss": 0.5096,
+      "step": 161
+    },
+    {
+      "epoch": 0.44021739130434784,
+      "grad_norm": 2.1590631008148193,
+      "learning_rate": 3e-05,
+      "loss": 0.6107,
+      "step": 162
+    },
+    {
+      "epoch": 0.4429347826086957,
+      "grad_norm": 1.8781529664993286,
+      "learning_rate": 3e-05,
+      "loss": 0.5122,
+      "step": 163
+    },
+    {
+      "epoch": 0.44565217391304346,
+      "grad_norm": 1.7852369546890259,
+      "learning_rate": 3e-05,
+      "loss": 0.4954,
+      "step": 164
+    },
+    {
+      "epoch": 0.4483695652173913,
+      "grad_norm": 1.3662923574447632,
+      "learning_rate": 3e-05,
+      "loss": 0.493,
+      "step": 165
+    },
+    {
+      "epoch": 0.45108695652173914,
+      "grad_norm": 2.7863638401031494,
+      "learning_rate": 3e-05,
+      "loss": 0.6855,
+      "step": 166
+    },
+    {
+      "epoch": 0.453804347826087,
+      "grad_norm": 2.3714311122894287,
+      "learning_rate": 3e-05,
+      "loss": 0.5858,
+      "step": 167
+    },
+    {
+      "epoch": 0.45652173913043476,
+      "grad_norm": 2.097414970397949,
+      "learning_rate": 3e-05,
+      "loss": 0.4895,
+      "step": 168
+    },
+    {
+      "epoch": 0.4592391304347826,
+      "grad_norm": 2.0122313499450684,
+      "learning_rate": 3e-05,
+      "loss": 0.4241,
+      "step": 169
+    },
+    {
+      "epoch": 0.46195652173913043,
+      "grad_norm": 1.5152671337127686,
+      "learning_rate": 3e-05,
+      "loss": 0.4318,
+      "step": 170
+    },
+    {
+      "epoch": 0.46467391304347827,
+      "grad_norm": 2.7088394165039062,
+      "learning_rate": 3e-05,
+      "loss": 0.6514,
+      "step": 171
+    },
+    {
+      "epoch": 0.4673913043478261,
+      "grad_norm": 1.5779036283493042,
+      "learning_rate": 3e-05,
+      "loss": 0.4605,
+      "step": 172
+    },
+    {
+      "epoch": 0.4701086956521739,
+      "grad_norm": 2.193013906478882,
+      "learning_rate": 3e-05,
+      "loss": 0.6419,
+      "step": 173
+    },
+    {
+      "epoch": 0.47282608695652173,
+      "grad_norm": 1.4573429822921753,
+      "learning_rate": 3e-05,
+      "loss": 0.4391,
+      "step": 174
+    },
+    {
+      "epoch": 0.47554347826086957,
+      "grad_norm": 1.719489574432373,
+      "learning_rate": 3e-05,
+      "loss": 0.6271,
+      "step": 175
+    },
+    {
+      "epoch": 0.4782608695652174,
+      "grad_norm": 1.491647481918335,
+      "learning_rate": 3e-05,
+      "loss": 0.5155,
+      "step": 176
+    },
+    {
+      "epoch": 0.48097826086956524,
+      "grad_norm": 2.1597378253936768,
+      "learning_rate": 3e-05,
+      "loss": 0.5132,
+      "step": 177
+    },
+    {
+      "epoch": 0.483695652173913,
+      "grad_norm": 1.62423574924469,
+      "learning_rate": 3e-05,
+      "loss": 0.4454,
+      "step": 178
+    },
+    {
+      "epoch": 0.48641304347826086,
+      "grad_norm": 2.4392452239990234,
+      "learning_rate": 3e-05,
+      "loss": 0.5913,
+      "step": 179
+    },
+    {
+      "epoch": 0.4891304347826087,
+      "grad_norm": 2.407681703567505,
+      "learning_rate": 3e-05,
+      "loss": 0.716,
+      "step": 180
+    },
+    {
+      "epoch": 0.49184782608695654,
+      "grad_norm": 2.402573585510254,
+      "learning_rate": 3e-05,
+      "loss": 0.5874,
+      "step": 181
+    },
+    {
+      "epoch": 0.4945652173913043,
+      "grad_norm": 1.8457105159759521,
+      "learning_rate": 3e-05,
+      "loss": 0.5029,
+      "step": 182
+    },
+    {
+      "epoch": 0.49728260869565216,
+      "grad_norm": 2.4062929153442383,
+      "learning_rate": 3e-05,
+      "loss": 0.6185,
+      "step": 183
+    },
+    {
+      "epoch": 0.5,
+      "grad_norm": 2.0781688690185547,
+      "learning_rate": 3e-05,
+      "loss": 0.6093,
+      "step": 184
+    },
+    {
+      "epoch": 0.5027173913043478,
+      "grad_norm": 1.5497978925704956,
+      "learning_rate": 3e-05,
+      "loss": 0.5322,
+      "step": 185
+    },
+    {
+      "epoch": 0.5054347826086957,
+      "grad_norm": 4.002748489379883,
+      "learning_rate": 3e-05,
+      "loss": 0.7277,
+      "step": 186
+    },
+    {
+      "epoch": 0.5081521739130435,
+      "grad_norm": 1.3743958473205566,
+      "learning_rate": 3e-05,
+      "loss": 0.4846,
+      "step": 187
+    },
+    {
+      "epoch": 0.5108695652173914,
+      "grad_norm": 1.8466805219650269,
+      "learning_rate": 3e-05,
+      "loss": 0.4154,
+      "step": 188
+    },
+    {
+      "epoch": 0.5135869565217391,
+      "grad_norm": 2.3035359382629395,
+      "learning_rate": 3e-05,
+      "loss": 0.5205,
+      "step": 189
+    },
+    {
+      "epoch": 0.5163043478260869,
+      "grad_norm": 2.2272605895996094,
+      "learning_rate": 3e-05,
+      "loss": 0.5532,
+      "step": 190
+    },
+    {
+      "epoch": 0.5190217391304348,
+      "grad_norm": 1.7287522554397583,
+      "learning_rate": 3e-05,
+      "loss": 0.4787,
+      "step": 191
+    },
+    {
+      "epoch": 0.5217391304347826,
+      "grad_norm": 2.0177953243255615,
+      "learning_rate": 3e-05,
+      "loss": 0.4949,
+      "step": 192
+    },
+    {
+      "epoch": 0.5244565217391305,
+      "grad_norm": 1.2431449890136719,
+      "learning_rate": 3e-05,
+      "loss": 0.4026,
+      "step": 193
+    },
+    {
+      "epoch": 0.5271739130434783,
+      "grad_norm": 1.2249736785888672,
+      "learning_rate": 3e-05,
+      "loss": 0.5167,
+      "step": 194
+    },
+    {
+      "epoch": 0.529891304347826,
+      "grad_norm": 1.9964460134506226,
+      "learning_rate": 3e-05,
+      "loss": 0.5396,
+      "step": 195
+    },
+    {
+      "epoch": 0.532608695652174,
+      "grad_norm": 1.4250489473342896,
+      "learning_rate": 3e-05,
+      "loss": 0.4755,
+      "step": 196
+    },
+    {
+      "epoch": 0.5353260869565217,
+      "grad_norm": 3.141204833984375,
+      "learning_rate": 3e-05,
+      "loss": 0.6342,
+      "step": 197
+    },
+    {
+      "epoch": 0.5380434782608695,
+      "grad_norm": 2.3809869289398193,
+      "learning_rate": 3e-05,
+      "loss": 0.5182,
+      "step": 198
+    },
+    {
+      "epoch": 0.5407608695652174,
+      "grad_norm": 3.068497896194458,
+      "learning_rate": 3e-05,
+      "loss": 0.789,
+      "step": 199
+    },
+    {
+      "epoch": 0.5434782608695652,
+      "grad_norm": 1.3273123502731323,
+      "learning_rate": 3e-05,
+      "loss": 0.5361,
+      "step": 200
+    },
+    {
+      "epoch": 0.5461956521739131,
+      "grad_norm": 1.2408298254013062,
+      "learning_rate": 3e-05,
+      "loss": 0.4686,
+      "step": 201
+    },
+    {
+      "epoch": 0.5489130434782609,
+      "grad_norm": 2.427406072616577,
+      "learning_rate": 3e-05,
+      "loss": 0.594,
+      "step": 202
+    },
+    {
+      "epoch": 0.5516304347826086,
+      "grad_norm": 2.725402593612671,
+      "learning_rate": 3e-05,
+      "loss": 0.5054,
+      "step": 203
+    },
+    {
+      "epoch": 0.5543478260869565,
+      "grad_norm": 2.7441318035125732,
+      "learning_rate": 3e-05,
+      "loss": 0.7141,
+      "step": 204
+    },
+    {
+      "epoch": 0.5570652173913043,
+      "grad_norm": 1.1885932683944702,
+      "learning_rate": 3e-05,
+      "loss": 0.3824,
+      "step": 205
+    },
+    {
+      "epoch": 0.5597826086956522,
+      "grad_norm": 1.8625751733779907,
+      "learning_rate": 3e-05,
+      "loss": 0.5073,
+      "step": 206
+    },
+    {
+      "epoch": 0.5625,
+      "grad_norm": 1.8269888162612915,
+      "learning_rate": 3e-05,
+      "loss": 0.5006,
+      "step": 207
+    },
+    {
+      "epoch": 0.5652173913043478,
+      "grad_norm": 1.4195995330810547,
+      "learning_rate": 3e-05,
+      "loss": 0.4833,
+      "step": 208
+    },
+    {
+      "epoch": 0.5679347826086957,
+      "grad_norm": 1.4971282482147217,
+      "learning_rate": 3e-05,
+      "loss": 0.527,
+      "step": 209
+    },
+    {
+      "epoch": 0.5706521739130435,
+      "grad_norm": 2.9385082721710205,
+      "learning_rate": 3e-05,
+      "loss": 0.6193,
+      "step": 210
+    },
+    {
+      "epoch": 0.5733695652173914,
+      "grad_norm": 1.6641467809677124,
+      "learning_rate": 3e-05,
+      "loss": 0.6115,
+      "step": 211
+    },
+    {
+      "epoch": 0.5760869565217391,
+      "grad_norm": 2.1166892051696777,
+      "learning_rate": 3e-05,
+      "loss": 0.6219,
+      "step": 212
+    },
+    {
+      "epoch": 0.5788043478260869,
+      "grad_norm": 2.319002866744995,
+      "learning_rate": 3e-05,
+      "loss": 0.5535,
+      "step": 213
+    },
+    {
+      "epoch": 0.5815217391304348,
+      "grad_norm": 2.2194128036499023,
+      "learning_rate": 3e-05,
+      "loss": 0.5677,
+      "step": 214
+    },
+    {
+      "epoch": 0.5842391304347826,
+      "grad_norm": 2.779507637023926,
+      "learning_rate": 3e-05,
+      "loss": 0.6392,
+      "step": 215
+    },
+    {
+      "epoch": 0.5869565217391305,
+      "grad_norm": 2.3067514896392822,
+      "learning_rate": 3e-05,
+      "loss": 0.6002,
+      "step": 216
+    },
+    {
+      "epoch": 0.5896739130434783,
+      "grad_norm": 1.7696046829223633,
+      "learning_rate": 3e-05,
+      "loss": 0.4951,
+      "step": 217
+    },
+    {
+      "epoch": 0.592391304347826,
+      "grad_norm": 1.649396538734436,
+      "learning_rate": 3e-05,
+      "loss": 0.4799,
+      "step": 218
+    },
+    {
+      "epoch": 0.595108695652174,
+      "grad_norm": 2.204364538192749,
+      "learning_rate": 3e-05,
+      "loss": 0.5045,
+      "step": 219
+    },
+    {
+      "epoch": 0.5978260869565217,
+      "grad_norm": 1.8227981328964233,
+      "learning_rate": 3e-05,
+      "loss": 0.5669,
+      "step": 220
+    },
+    {
+      "epoch": 0.6005434782608695,
+      "grad_norm": 1.8465766906738281,
+      "learning_rate": 3e-05,
+      "loss": 0.6157,
+      "step": 221
+    },
+    {
+      "epoch": 0.6032608695652174,
+      "grad_norm": 2.102964162826538,
+      "learning_rate": 3e-05,
+      "loss": 0.4715,
+      "step": 222
+    },
+    {
+      "epoch": 0.6059782608695652,
+      "grad_norm": 1.6074268817901611,
+      "learning_rate": 3e-05,
+      "loss": 0.48,
+      "step": 223
+    },
+    {
+      "epoch": 0.6086956521739131,
+      "grad_norm": 1.9385191202163696,
+      "learning_rate": 3e-05,
+      "loss": 0.4727,
+      "step": 224
+    },
+    {
+      "epoch": 0.6114130434782609,
+      "grad_norm": 2.639068603515625,
+      "learning_rate": 3e-05,
+      "loss": 0.6696,
+      "step": 225
+    },
+    {
+      "epoch": 0.6141304347826086,
+      "grad_norm": 1.9903889894485474,
+      "learning_rate": 3e-05,
+      "loss": 0.5696,
+      "step": 226
+    },
+    {
+      "epoch": 0.6168478260869565,
+      "grad_norm": 2.7789058685302734,
+      "learning_rate": 3e-05,
+      "loss": 0.6264,
+      "step": 227
+    },
+    {
+      "epoch": 0.6195652173913043,
+      "grad_norm": 2.7294304370880127,
+      "learning_rate": 3e-05,
+      "loss": 0.5185,
+      "step": 228
+    },
+    {
+      "epoch": 0.6222826086956522,
+      "grad_norm": 2.1425840854644775,
+      "learning_rate": 3e-05,
+      "loss": 0.4822,
+      "step": 229
+    },
+    {
+      "epoch": 0.625,
+      "grad_norm": 2.22175669670105,
+      "learning_rate": 3e-05,
+      "loss": 0.6414,
+      "step": 230
+    },
+    {
+      "epoch": 0.6277173913043478,
+      "grad_norm": 2.031067132949829,
+      "learning_rate": 3e-05,
+      "loss": 0.4099,
+      "step": 231
+    },
+    {
+      "epoch": 0.6304347826086957,
+      "grad_norm": 2.0958268642425537,
+      "learning_rate": 3e-05,
+      "loss": 0.7023,
+      "step": 232
+    },
+    {
+      "epoch": 0.6331521739130435,
+      "grad_norm": 2.353266716003418,
+      "learning_rate": 3e-05,
+      "loss": 0.5861,
+      "step": 233
+    },
+    {
+      "epoch": 0.6358695652173914,
+      "grad_norm": 2.579880475997925,
+      "learning_rate": 3e-05,
+      "loss": 0.6487,
+      "step": 234
+    },
+    {
+      "epoch": 0.6385869565217391,
+      "grad_norm": 2.1880500316619873,
+      "learning_rate": 3e-05,
+      "loss": 0.6574,
+      "step": 235
+    },
+    {
+      "epoch": 0.6413043478260869,
+      "grad_norm": 1.692370891571045,
+      "learning_rate": 3e-05,
+      "loss": 0.443,
+      "step": 236
+    },
+    {
+      "epoch": 0.6440217391304348,
+      "grad_norm": 1.253108263015747,
+      "learning_rate": 3e-05,
+      "loss": 0.3879,
+      "step": 237
+    },
+    {
+      "epoch": 0.6467391304347826,
+      "grad_norm": 2.0025603771209717,
+      "learning_rate": 3e-05,
+      "loss": 0.5068,
+      "step": 238
+    },
+    {
+      "epoch": 0.6494565217391305,
+      "grad_norm": 2.018907070159912,
+      "learning_rate": 3e-05,
+      "loss": 0.5783,
+      "step": 239
+    },
+    {
+      "epoch": 0.6521739130434783,
+      "grad_norm": 1.8426876068115234,
+      "learning_rate": 3e-05,
+      "loss": 0.5025,
+      "step": 240
+    },
+    {
+      "epoch": 0.654891304347826,
+      "grad_norm": 1.7005494832992554,
+      "learning_rate": 3e-05,
+      "loss": 0.4745,
+      "step": 241
+    },
+    {
+      "epoch": 0.657608695652174,
+      "grad_norm": 1.503015398979187,
+      "learning_rate": 3e-05,
+      "loss": 0.5462,
+      "step": 242
+    },
+    {
+      "epoch": 0.6603260869565217,
+      "grad_norm": 1.5778555870056152,
+      "learning_rate": 3e-05,
+      "loss": 0.4247,
+      "step": 243
+    },
+    {
+      "epoch": 0.6630434782608695,
+      "grad_norm": 2.054659843444824,
+      "learning_rate": 3e-05,
+      "loss": 0.5232,
+      "step": 244
+    },
+    {
+      "epoch": 0.6657608695652174,
+      "grad_norm": 1.7915781736373901,
+      "learning_rate": 3e-05,
+      "loss": 0.501,
+      "step": 245
+    },
+    {
+      "epoch": 0.6684782608695652,
+      "grad_norm": 2.16675066947937,
+      "learning_rate": 3e-05,
+      "loss": 0.6102,
+      "step": 246
+    },
+    {
+      "epoch": 0.6711956521739131,
+      "grad_norm": 2.503446578979492,
+      "learning_rate": 3e-05,
+      "loss": 0.5873,
+      "step": 247
+    },
+    {
+      "epoch": 0.6739130434782609,
+      "grad_norm": 1.6269731521606445,
+      "learning_rate": 3e-05,
+      "loss": 0.4125,
+      "step": 248
+    },
+    {
+      "epoch": 0.6766304347826086,
+      "grad_norm": 1.795397162437439,
+      "learning_rate": 3e-05,
+      "loss": 0.5117,
+      "step": 249
+    },
+    {
+      "epoch": 0.6793478260869565,
+      "grad_norm": 2.1140689849853516,
+      "learning_rate": 3e-05,
+      "loss": 0.494,
+      "step": 250
+    },
+    {
+      "epoch": 0.6820652173913043,
+      "grad_norm": 1.8272490501403809,
+      "learning_rate": 3e-05,
+      "loss": 0.506,
+      "step": 251
+    },
+    {
+      "epoch": 0.6847826086956522,
+      "grad_norm": 1.8032358884811401,
+      "learning_rate": 3e-05,
+      "loss": 0.5329,
+      "step": 252
+    },
+    {
+      "epoch": 0.6875,
+      "grad_norm": 2.1028482913970947,
+      "learning_rate": 3e-05,
+      "loss": 0.5306,
+      "step": 253
+    },
+    {
+      "epoch": 0.6902173913043478,
+      "grad_norm": 3.13128399848938,
+      "learning_rate": 3e-05,
+      "loss": 0.6008,
+      "step": 254
+    },
+    {
+      "epoch": 0.6929347826086957,
+      "grad_norm": 1.6769640445709229,
+      "learning_rate": 3e-05,
+      "loss": 0.5309,
+      "step": 255
+    },
+    {
+      "epoch": 0.6956521739130435,
+      "grad_norm": 1.5554113388061523,
+      "learning_rate": 3e-05,
+      "loss": 0.5298,
+      "step": 256
+    },
+    {
+      "epoch": 0.6983695652173914,
+      "grad_norm": 1.408714771270752,
+      "learning_rate": 3e-05,
+      "loss": 0.4415,
+      "step": 257
+    },
+    {
+      "epoch": 0.7010869565217391,
+      "grad_norm": 1.4659717082977295,
+      "learning_rate": 3e-05,
+      "loss": 0.3927,
+      "step": 258
+    },
+    {
+      "epoch": 0.7038043478260869,
+      "grad_norm": 1.5185580253601074,
+      "learning_rate": 3e-05,
+      "loss": 0.5113,
+      "step": 259
+    },
+    {
+      "epoch": 0.7065217391304348,
+      "grad_norm": 1.2095264196395874,
+      "learning_rate": 3e-05,
+      "loss": 0.4989,
+      "step": 260
+    },
+    {
+      "epoch": 0.7092391304347826,
+      "grad_norm": 1.7118961811065674,
+      "learning_rate": 3e-05,
+      "loss": 0.5711,
+      "step": 261
+    },
+    {
+      "epoch": 0.7119565217391305,
+      "grad_norm": 1.268684983253479,
+      "learning_rate": 3e-05,
+      "loss": 0.4141,
+      "step": 262
+    },
+    {
+      "epoch": 0.7146739130434783,
+      "grad_norm": 1.5639039278030396,
+      "learning_rate": 3e-05,
+      "loss": 0.3954,
+      "step": 263
+    },
+    {
+      "epoch": 0.717391304347826,
+      "grad_norm": 1.5051435232162476,
+      "learning_rate": 3e-05,
+      "loss": 0.5565,
+      "step": 264
+    },
+    {
+      "epoch": 0.720108695652174,
+      "grad_norm": 2.289475679397583,
+      "learning_rate": 3e-05,
+      "loss": 0.6248,
+      "step": 265
+    },
+    {
+      "epoch": 0.7228260869565217,
+      "grad_norm": 2.8152410984039307,
+      "learning_rate": 3e-05,
+      "loss": 0.5782,
+      "step": 266
+    },
+    {
+      "epoch": 0.7255434782608695,
+      "grad_norm": 1.3830969333648682,
+      "learning_rate": 3e-05,
+      "loss": 0.3414,
+      "step": 267
+    },
+    {
+      "epoch": 0.7282608695652174,
+      "grad_norm": 1.6137558221817017,
+      "learning_rate": 3e-05,
+      "loss": 0.5093,
+      "step": 268
+    },
+    {
+      "epoch": 0.7309782608695652,
+      "grad_norm": 2.456444025039673,
+      "learning_rate": 3e-05,
+      "loss": 0.5382,
+      "step": 269
+    },
+    {
+      "epoch": 0.7336956521739131,
+      "grad_norm": 2.212475061416626,
+      "learning_rate": 3e-05,
+      "loss": 0.4898,
+      "step": 270
+    },
+    {
+      "epoch": 0.7364130434782609,
+      "grad_norm": 2.4707953929901123,
+      "learning_rate": 3e-05,
+      "loss": 0.7499,
+      "step": 271
+    },
+    {
+      "epoch": 0.7391304347826086,
+      "grad_norm": 1.7205860614776611,
+      "learning_rate": 3e-05,
+      "loss": 0.4294,
+      "step": 272
+    },
+    {
+      "epoch": 0.7418478260869565,
+      "grad_norm": 2.055875062942505,
+      "learning_rate": 3e-05,
+      "loss": 0.6336,
+      "step": 273
+    },
+    {
+      "epoch": 0.7445652173913043,
+      "grad_norm": 2.388607978820801,
+      "learning_rate": 3e-05,
+      "loss": 0.5929,
+      "step": 274
+    },
+    {
+      "epoch": 0.7472826086956522,
+      "grad_norm": 2.509526491165161,
+      "learning_rate": 3e-05,
+      "loss": 0.6704,
+      "step": 275
+    },
+    {
+      "epoch": 0.75,
+      "grad_norm": 1.3426049947738647,
+      "learning_rate": 3e-05,
+      "loss": 0.4698,
+      "step": 276
+    },
+    {
+      "epoch": 0.7527173913043478,
+      "grad_norm": 1.8016018867492676,
+      "learning_rate": 3e-05,
+      "loss": 0.5052,
+      "step": 277
+    },
+    {
+      "epoch": 0.7554347826086957,
+      "grad_norm": 1.7007917165756226,
+      "learning_rate": 3e-05,
+      "loss": 0.5279,
+      "step": 278
+    },
+    {
+      "epoch": 0.7581521739130435,
+      "grad_norm": 2.104093074798584,
+      "learning_rate": 3e-05,
+      "loss": 0.5401,
+      "step": 279
+    },
+    {
+      "epoch": 0.7608695652173914,
+      "grad_norm": 2.2383487224578857,
+      "learning_rate": 3e-05,
+      "loss": 0.5503,
+      "step": 280
+    },
+    {
+      "epoch": 0.7635869565217391,
+      "grad_norm": 1.8349882364273071,
+      "learning_rate": 3e-05,
+      "loss": 0.539,
+      "step": 281
+    },
+    {
+      "epoch": 0.7663043478260869,
+      "grad_norm": 1.6838781833648682,
+      "learning_rate": 3e-05,
+      "loss": 0.4281,
+      "step": 282
+    },
+    {
+      "epoch": 0.7690217391304348,
+      "grad_norm": 2.3261828422546387,
+      "learning_rate": 3e-05,
+      "loss": 0.507,
+      "step": 283
+    },
+    {
+      "epoch": 0.7717391304347826,
+      "grad_norm": 1.6605132818222046,
+      "learning_rate": 3e-05,
+      "loss": 0.6602,
+      "step": 284
+    },
+    {
+      "epoch": 0.7744565217391305,
+      "grad_norm": 2.164030075073242,
+      "learning_rate": 3e-05,
+      "loss": 0.4894,
+      "step": 285
+    },
+    {
+      "epoch": 0.7771739130434783,
+      "grad_norm": 1.8546051979064941,
+      "learning_rate": 3e-05,
+      "loss": 0.44,
+      "step": 286
+    },
+    {
+      "epoch": 0.779891304347826,
+      "grad_norm": 2.498770236968994,
+      "learning_rate": 3e-05,
+      "loss": 0.627,
+      "step": 287
+    },
+    {
+      "epoch": 0.782608695652174,
+      "grad_norm": 2.179408550262451,
+      "learning_rate": 3e-05,
+      "loss": 0.6078,
+      "step": 288
+    },
+    {
+      "epoch": 0.7853260869565217,
+      "grad_norm": 1.6444724798202515,
+      "learning_rate": 3e-05,
+      "loss": 0.56,
+      "step": 289
+    },
+    {
+      "epoch": 0.7880434782608695,
+      "grad_norm": 1.3116724491119385,
+      "learning_rate": 3e-05,
+      "loss": 0.4838,
+      "step": 290
+    },
+    {
+      "epoch": 0.7907608695652174,
+      "grad_norm": 2.142730951309204,
+      "learning_rate": 3e-05,
+      "loss": 0.5892,
+      "step": 291
+    },
+    {
+      "epoch": 0.7934782608695652,
+      "grad_norm": 1.947159767150879,
+      "learning_rate": 3e-05,
+      "loss": 0.4796,
+      "step": 292
+    },
+    {
+      "epoch": 0.7961956521739131,
+      "grad_norm": 1.7489418983459473,
+      "learning_rate": 3e-05,
+      "loss": 0.6132,
+      "step": 293
+    },
+    {
+      "epoch": 0.7989130434782609,
+      "grad_norm": 1.7919461727142334,
+      "learning_rate": 3e-05,
+      "loss": 0.4928,
+      "step": 294
+    },
+    {
+      "epoch": 0.8016304347826086,
+      "grad_norm": 2.547961711883545,
+      "learning_rate": 3e-05,
+      "loss": 0.491,
+      "step": 295
+    },
+    {
+      "epoch": 0.8043478260869565,
+      "grad_norm": 2.5002658367156982,
+      "learning_rate": 3e-05,
+      "loss": 0.6737,
+      "step": 296
+    },
+    {
+      "epoch": 0.8070652173913043,
+      "grad_norm": 1.5400561094284058,
+      "learning_rate": 3e-05,
+      "loss": 0.4632,
+      "step": 297
+    },
+    {
+      "epoch": 0.8097826086956522,
+      "grad_norm": 1.2512913942337036,
+      "learning_rate": 3e-05,
+      "loss": 0.563,
+      "step": 298
+    },
+    {
+      "epoch": 0.8125,
+      "grad_norm": 1.7560540437698364,
+      "learning_rate": 3e-05,
+      "loss": 0.5273,
+      "step": 299
+    },
+    {
+      "epoch": 0.8152173913043478,
+      "grad_norm": 2.1025469303131104,
+      "learning_rate": 3e-05,
+      "loss": 0.5813,
+      "step": 300
+    },
+    {
+      "epoch": 0.8179347826086957,
+      "grad_norm": 1.748838186264038,
+      "learning_rate": 3e-05,
+      "loss": 0.5182,
+      "step": 301
+    },
+    {
+      "epoch": 0.8206521739130435,
+      "grad_norm": 1.7875301837921143,
+      "learning_rate": 3e-05,
+      "loss": 0.5248,
+      "step": 302
+    },
+    {
+      "epoch": 0.8233695652173914,
+      "grad_norm": 1.6547335386276245,
+      "learning_rate": 3e-05,
+      "loss": 0.4525,
+      "step": 303
+    },
+    {
+      "epoch": 0.8260869565217391,
+      "grad_norm": 2.5749781131744385,
+      "learning_rate": 3e-05,
+      "loss": 0.5864,
+      "step": 304
+    },
+    {
+      "epoch": 0.8288043478260869,
+      "grad_norm": 2.010406255722046,
+      "learning_rate": 3e-05,
+      "loss": 0.4704,
+      "step": 305
+    },
+    {
+      "epoch": 0.8315217391304348,
+      "grad_norm": 1.61479914188385,
+      "learning_rate": 3e-05,
+      "loss": 0.4981,
+      "step": 306
+    },
+    {
+      "epoch": 0.8342391304347826,
+      "grad_norm": 2.3102943897247314,
+      "learning_rate": 3e-05,
+      "loss": 0.4854,
+      "step": 307
+    },
+    {
+      "epoch": 0.8369565217391305,
+      "grad_norm": 4.297898769378662,
+      "learning_rate": 3e-05,
+      "loss": 0.6969,
+      "step": 308
+    },
+    {
+      "epoch": 0.8396739130434783,
+      "grad_norm": 1.56645667552948,
+      "learning_rate": 3e-05,
+      "loss": 0.4675,
+      "step": 309
+    },
+    {
+      "epoch": 0.842391304347826,
+      "grad_norm": 2.9031646251678467,
+      "learning_rate": 3e-05,
+      "loss": 0.713,
+      "step": 310
+    },
+    {
+      "epoch": 0.845108695652174,
+      "grad_norm": 1.9214662313461304,
+      "learning_rate": 3e-05,
+      "loss": 0.5379,
+      "step": 311
+    },
+    {
+      "epoch": 0.8478260869565217,
+      "grad_norm": 2.371910810470581,
+      "learning_rate": 3e-05,
+      "loss": 0.5174,
+      "step": 312
+    },
+    {
+      "epoch": 0.8505434782608695,
+      "grad_norm": 2.3552019596099854,
+      "learning_rate": 3e-05,
+      "loss": 0.4784,
+      "step": 313
+    },
+    {
+      "epoch": 0.8532608695652174,
+      "grad_norm": 1.3287854194641113,
+      "learning_rate": 3e-05,
+      "loss": 0.5026,
+      "step": 314
+    },
+    {
+      "epoch": 0.8559782608695652,
+      "grad_norm": 3.1460154056549072,
+      "learning_rate": 3e-05,
+      "loss": 0.6437,
+      "step": 315
+    },
+    {
+      "epoch": 0.8586956521739131,
+      "grad_norm": 3.9493658542633057,
+      "learning_rate": 3e-05,
+      "loss": 0.7079,
+      "step": 316
+    },
+    {
+      "epoch": 0.8614130434782609,
+      "grad_norm": 1.8890721797943115,
+      "learning_rate": 3e-05,
+      "loss": 0.4789,
+      "step": 317
+    },
+    {
+      "epoch": 0.8641304347826086,
+      "grad_norm": 2.397585153579712,
+      "learning_rate": 3e-05,
+      "loss": 0.5358,
+      "step": 318
+    },
+    {
+      "epoch": 0.8668478260869565,
+      "grad_norm": 3.6481263637542725,
+      "learning_rate": 3e-05,
+      "loss": 0.6699,
+      "step": 319
+    },
+    {
+      "epoch": 0.8695652173913043,
+      "grad_norm": 3.643709421157837,
+      "learning_rate": 3e-05,
+      "loss": 0.5464,
+      "step": 320
+    },
+    {
+      "epoch": 0.8722826086956522,
+      "grad_norm": 1.8755005598068237,
+      "learning_rate": 3e-05,
+      "loss": 0.5231,
+      "step": 321
+    },
+    {
+      "epoch": 0.875,
+      "grad_norm": 1.2588874101638794,
+      "learning_rate": 3e-05,
+      "loss": 0.3987,
+      "step": 322
+    },
+    {
+      "epoch": 0.8777173913043478,
+      "grad_norm": 1.8710525035858154,
+      "learning_rate": 3e-05,
+      "loss": 0.5137,
+      "step": 323
+    },
+    {
+      "epoch": 0.8804347826086957,
+      "grad_norm": 1.8760285377502441,
+      "learning_rate": 3e-05,
+      "loss": 0.5162,
+      "step": 324
+    },
+    {
+      "epoch": 0.8831521739130435,
+      "grad_norm": 2.302496910095215,
+      "learning_rate": 3e-05,
+      "loss": 0.6338,
+      "step": 325
+    },
+    {
+      "epoch": 0.8858695652173914,
+      "grad_norm": 2.173513650894165,
+      "learning_rate": 3e-05,
+      "loss": 0.5525,
+      "step": 326
+    },
+    {
+      "epoch": 0.8885869565217391,
+      "grad_norm": 3.2215616703033447,
+      "learning_rate": 3e-05,
+      "loss": 0.4746,
+      "step": 327
+    },
+    {
+      "epoch": 0.8913043478260869,
+      "grad_norm": 1.402948021888733,
+      "learning_rate": 3e-05,
+      "loss": 0.4551,
+      "step": 328
+    },
+    {
+      "epoch": 0.8940217391304348,
+      "grad_norm": 2.177567958831787,
+      "learning_rate": 3e-05,
+      "loss": 0.5937,
+      "step": 329
+    },
+    {
+      "epoch": 0.8967391304347826,
+      "grad_norm": 1.9073940515518188,
+      "learning_rate": 3e-05,
+      "loss": 0.6849,
+      "step": 330
+    },
+    {
+      "epoch": 0.8994565217391305,
+      "grad_norm": 2.078620195388794,
+      "learning_rate": 3e-05,
+      "loss": 0.4851,
+      "step": 331
+    },
+    {
+      "epoch": 0.9021739130434783,
+      "grad_norm": 1.666050672531128,
+      "learning_rate": 3e-05,
+      "loss": 0.5361,
+      "step": 332
+    },
+    {
+      "epoch": 0.904891304347826,
+      "grad_norm": 1.8900854587554932,
+      "learning_rate": 3e-05,
+      "loss": 0.5878,
+      "step": 333
+    },
+    {
+      "epoch": 0.907608695652174,
+      "grad_norm": 2.4157278537750244,
+      "learning_rate": 3e-05,
+      "loss": 0.583,
+      "step": 334
+    },
+    {
+      "epoch": 0.9103260869565217,
+      "grad_norm": 1.9325672388076782,
+      "learning_rate": 3e-05,
+      "loss": 0.5036,
+      "step": 335
+    },
+    {
+      "epoch": 0.9130434782608695,
+      "grad_norm": 2.91580867767334,
+      "learning_rate": 3e-05,
+      "loss": 0.6958,
+      "step": 336
+    },
+    {
+      "epoch": 0.9157608695652174,
+      "grad_norm": 2.5430386066436768,
+      "learning_rate": 3e-05,
+      "loss": 0.5496,
+      "step": 337
+    },
+    {
+      "epoch": 0.9184782608695652,
+      "grad_norm": 2.0265727043151855,
+      "learning_rate": 3e-05,
+      "loss": 0.4997,
+      "step": 338
+    },
+    {
+      "epoch": 0.9211956521739131,
+      "grad_norm": 1.583471417427063,
+      "learning_rate": 3e-05,
+      "loss": 0.4641,
+      "step": 339
+    },
+    {
+      "epoch": 0.9239130434782609,
+      "grad_norm": 2.3555848598480225,
+      "learning_rate": 3e-05,
+      "loss": 0.4989,
+      "step": 340
+    },
+    {
+      "epoch": 0.9266304347826086,
+      "grad_norm": 2.229189395904541,
+      "learning_rate": 3e-05,
+      "loss": 0.5665,
+      "step": 341
+    },
+    {
+      "epoch": 0.9293478260869565,
+      "grad_norm": 1.3699840307235718,
+      "learning_rate": 3e-05,
+      "loss": 0.523,
+      "step": 342
+    },
+    {
+      "epoch": 0.9320652173913043,
+      "grad_norm": 1.9070311784744263,
+      "learning_rate": 3e-05,
+      "loss": 0.565,
+      "step": 343
+    },
+    {
+      "epoch": 0.9347826086956522,
+      "grad_norm": 2.588691234588623,
+      "learning_rate": 3e-05,
+      "loss": 0.5814,
+      "step": 344
+    },
+    {
+      "epoch": 0.9375,
+      "grad_norm": 1.7156322002410889,
+      "learning_rate": 3e-05,
+      "loss": 0.5074,
+      "step": 345
+    },
+    {
+      "epoch": 0.9402173913043478,
+      "grad_norm": 2.289538860321045,
+      "learning_rate": 3e-05,
+      "loss": 0.5744,
+      "step": 346
+    },
+    {
+      "epoch": 0.9429347826086957,
+      "grad_norm": 1.0715011358261108,
+      "learning_rate": 3e-05,
+      "loss": 0.4612,
+      "step": 347
+    },
+    {
+      "epoch": 0.9456521739130435,
+      "grad_norm": 1.5660399198532104,
+      "learning_rate": 3e-05,
+      "loss": 0.4354,
+      "step": 348
+    },
+    {
+      "epoch": 0.9483695652173914,
+      "grad_norm": 2.2513458728790283,
+      "learning_rate": 3e-05,
+      "loss": 0.5702,
+      "step": 349
+    },
+    {
+      "epoch": 0.9510869565217391,
+      "grad_norm": 2.5733985900878906,
+      "learning_rate": 3e-05,
+      "loss": 0.6115,
+      "step": 350
+    },
+    {
+      "epoch": 0.9538043478260869,
+      "grad_norm": 1.1676104068756104,
+      "learning_rate": 3e-05,
+      "loss": 0.4158,
+      "step": 351
+    },
+    {
+      "epoch": 0.9565217391304348,
+      "grad_norm": 2.1852831840515137,
+      "learning_rate": 3e-05,
+      "loss": 0.5267,
+      "step": 352
+    },
+    {
+      "epoch": 0.9592391304347826,
+      "grad_norm": 1.339505910873413,
+      "learning_rate": 3e-05,
+      "loss": 0.4204,
+      "step": 353
+    },
+    {
+      "epoch": 0.9619565217391305,
+      "grad_norm": 1.5837697982788086,
+      "learning_rate": 3e-05,
+      "loss": 0.4722,
+      "step": 354
+    },
+    {
+      "epoch": 0.9646739130434783,
+      "grad_norm": 1.5127207040786743,
+      "learning_rate": 3e-05,
+      "loss": 0.4782,
+      "step": 355
+    },
+    {
+      "epoch": 0.967391304347826,
+      "grad_norm": 0.9464153051376343,
+      "learning_rate": 3e-05,
+      "loss": 0.3675,
+      "step": 356
+    },
+    {
+      "epoch": 0.970108695652174,
+      "grad_norm": 3.507087469100952,
+      "learning_rate": 3e-05,
+      "loss": 0.5947,
+      "step": 357
+    },
+    {
+      "epoch": 0.9728260869565217,
+      "grad_norm": 2.1700286865234375,
+      "learning_rate": 3e-05,
+      "loss": 0.5949,
+      "step": 358
+    },
+    {
+      "epoch": 0.9755434782608695,
+      "grad_norm": 1.070860505104065,
+      "learning_rate": 3e-05,
+      "loss": 0.4209,
+      "step": 359
+    },
+    {
+      "epoch": 0.9782608695652174,
+      "grad_norm": 1.8626935482025146,
+      "learning_rate": 3e-05,
+      "loss": 0.4979,
+      "step": 360
+    },
+    {
+      "epoch": 0.9809782608695652,
+      "grad_norm": 1.2900367975234985,
+      "learning_rate": 3e-05,
+      "loss": 0.4778,
+      "step": 361
+    },
+    {
+      "epoch": 0.9836956521739131,
+      "grad_norm": 1.2917206287384033,
+      "learning_rate": 3e-05,
+      "loss": 0.4561,
+      "step": 362
+    },
+    {
+      "epoch": 0.9864130434782609,
+      "grad_norm": 2.3133273124694824,
+      "learning_rate": 3e-05,
+      "loss": 0.5618,
+      "step": 363
+    },
+    {
+      "epoch": 0.9891304347826086,
+      "grad_norm": 1.5823960304260254,
+      "learning_rate": 3e-05,
+      "loss": 0.5549,
+      "step": 364
+    },
+    {
+      "epoch": 0.9918478260869565,
+      "grad_norm": 1.6155842542648315,
+      "learning_rate": 3e-05,
+      "loss": 0.4442,
+      "step": 365
+    },
+    {
+      "epoch": 0.9945652173913043,
+      "grad_norm": 1.7502015829086304,
+      "learning_rate": 3e-05,
+      "loss": 0.5179,
+      "step": 366
+    },
+    {
+      "epoch": 0.9972826086956522,
+      "grad_norm": 2.5052552223205566,
+      "learning_rate": 3e-05,
+      "loss": 0.5613,
+      "step": 367
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 0.9979060888290405,
+      "learning_rate": 3e-05,
+      "loss": 0.3232,
+      "step": 368
+    },
+    {
+      "epoch": 1.0,
+      "step": 368,
+      "total_flos": 6.339170402304e+16,
+      "train_loss": 0.5430356652840324,
+      "train_runtime": 908.8402,
+      "train_samples_per_second": 3.234,
+      "train_steps_per_second": 0.405
+    }
+  ],
+  "logging_steps": 1.0,
+  "max_steps": 368,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 1,
+  "save_steps": 200,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 6.339170402304e+16,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}

vocoder_config.json

@@ -0,0 +1,53 @@
+{
+    "input_wavs_dir": "/private/home/adampolyak/datasets/LJ/LJSpeech-1.1/wavs_16khz_padded",
+    "input_training_file": "/large_experiments/ust/annl/datasets/tts/LJSpeech/filelist/mhubert_vp_en_es_fr_it3_400k/lj_train_layer11_hubert1000_filelist.txt",
+    "input_validation_file": "/large_experiments/ust/annl/datasets/tts/LJSpeech/filelist/mhubert_vp_en_es_fr_it3_400k/lj_dev_layer11_hubert1000_filelist.txt",
+
+    "resblock": "1",
+    "num_gpus": 0,
+    "batch_size": 16,
+    "learning_rate": 0.0002,
+    "adam_b1": 0.8,
+    "adam_b2": 0.99,
+    "lr_decay": 0.999,
+    "seed": 1234,
+
+    "upsample_rates": [5,4,4,2,2],
+    "upsample_kernel_sizes": [11,8,8,4,4],
+    "upsample_initial_channel": 512,
+    "resblock_kernel_sizes": [3,7,11],
+    "resblock_dilation_sizes": [[1,3,5], [1,3,5], [1,3,5]],
+    "num_embeddings": 1000,
+    "embedding_dim": 128,
+    "model_in_dim": 128,
+
+    "segment_size": 8960,
+    "code_hop_size": 320,
+    "f0": false,
+    "num_mels": 80,
+    "num_freq": 1025,
+    "n_fft": 1024,
+    "hop_size": 256,
+    "win_size": 1024,
+
+    "dur_prediction_weight": 1.0,
+    "dur_predictor_params": {
+        "encoder_embed_dim": 128,
+        "var_pred_hidden_dim": 128,
+        "var_pred_kernel_size": 3,
+        "var_pred_dropout": 0.5
+    },
+
+    "sampling_rate": 16000,
+
+    "fmin": 0,
+    "fmax": 8000,
+    "fmax_for_loss": null,
+
+    "num_workers": 4,
+
+    "dist_config": {
+        "dist_backend": "nccl",
+        "dist_url": "env://"
+    }
+}