Initial commit

README.md

@@ -0,0 +1,60 @@
+---
+license: other
+base_model: meta-llama/Llama-2-7b-chat-hf
+tags:
+- llama-factory
+- full
+- generated_from_trainer
+model-index:
+- name: train_2024-07-16-09-05-28_llama2
+  results: []
+---
+
+<!-- This model card has been generated automatically according to the information the Trainer had access to. You
+should probably proofread and complete it, then remove this comment. -->
+
+# train_2024-07-16-09-05-28_llama2
+
+This model is a fine-tuned version of [meta-llama/Llama-2-7b-chat-hf](https://huggingface.co/meta-llama/Llama-2-7b-chat-hf) on the truth_train_0716 dataset.
+
+## Model description
+
+More information needed
+
+## Intended uses & limitations
+
+More information needed
+
+## Training and evaluation data
+
+More information needed
+
+## Training procedure
+
+### Training hyperparameters
+
+The following hyperparameters were used during training:
+- learning_rate: 5e-06
+- train_batch_size: 2
+- eval_batch_size: 8
+- seed: 42
+- distributed_type: multi-GPU
+- num_devices: 8
+- gradient_accumulation_steps: 8
+- total_train_batch_size: 128
+- total_eval_batch_size: 64
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: cosine
+- lr_scheduler_warmup_steps: 10
+- num_epochs: 5.0
+
+### Training results
+
+
+
+### Framework versions
+
+- Transformers 4.42.3
+- Pytorch 2.3.0a0+ebedce2
+- Datasets 2.20.0
+- Tokenizers 0.19.1

all_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 4.887459807073955,
+    "num_input_tokens_seen": 1299392,
+    "total_flos": 5.151317702790349e+16,
+    "train_loss": 0.3433768034317166,
+    "train_runtime": 2162.0959,
+    "train_samples_per_second": 11.489,
+    "train_steps_per_second": 0.088
+}

checkpoint-190/config.json

@@ -0,0 +1,29 @@
+{
+  "_name_or_path": "meta-llama/Llama-2-7b-chat-hf",
+  "architectures": [
+    "LlamaForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "bos_token_id": 1,
+  "eos_token_id": 2,
+  "hidden_act": "silu",
+  "hidden_size": 4096,
+  "initializer_range": 0.02,
+  "intermediate_size": 11008,
+  "max_position_embeddings": 4096,
+  "mlp_bias": false,
+  "model_type": "llama",
+  "num_attention_heads": 32,
+  "num_hidden_layers": 32,
+  "num_key_value_heads": 32,
+  "pretraining_tp": 1,
+  "rms_norm_eps": 1e-05,
+  "rope_scaling": null,
+  "rope_theta": 10000.0,
+  "tie_word_embeddings": false,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.42.3",
+  "use_cache": false,
+  "vocab_size": 32000
+}

checkpoint-190/generation_config.json

@@ -0,0 +1,10 @@
+{
+  "bos_token_id": 1,
+  "do_sample": true,
+  "eos_token_id": 2,
+  "max_length": 4096,
+  "pad_token_id": 0,
+  "temperature": 0.6,
+  "top_p": 0.9,
+  "transformers_version": "4.42.3"
+}

checkpoint-190/global_step190/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e1e96a89cd156105a6cbcb0cc0958f7970eb3b52a02c3525fd5f206487d962f0
+size 10107631180

checkpoint-190/global_step190/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:584e847239f9463d01f59de838f70c500d6df1e42742105aacc922197ba35ce6
+size 10107631372

checkpoint-190/global_step190/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:afce19f072fa6aa461d9893678f22d62847a38c40f6b0762d94540902d5fa9bd
+size 10107631436

checkpoint-190/global_step190/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3fa61774cb1ebc3769aa4b8587d988a1b97b0370a5acfb04da33974d7ada710f
+size 10107631372

checkpoint-190/global_step190/bf16_zero_pp_rank_4_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b9280c6e4504ce36a553485fca0bc8c4f3882c9e3420b37b3cc5af22ea33e06b
+size 10107631436

checkpoint-190/global_step190/bf16_zero_pp_rank_5_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2cc7bb260f54a427c5aa166b894a23b4053aeef0b1a9fa03c81bcbbc0e025ae1
+size 10107631500

checkpoint-190/global_step190/bf16_zero_pp_rank_6_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:35edca3844aabdf871d78908f25f5efa028288ae8bbdba4aa5379f5263cf2c0c
+size 10107631436

checkpoint-190/global_step190/bf16_zero_pp_rank_7_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3647ef2e82720cf924aff42ec47ce772d6ffc8ebe9b2e400db6e63fce2a68dd2
+size 10107630988

checkpoint-190/global_step190/mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5bdf19aed6cf7a9bdc707ca65b5d2a0ce2817a5c720593fcf97ea4197475aef5
+size 13476919288

checkpoint-190/latest

@@ -0,0 +1 @@
+global_step190

checkpoint-190/model-00001-of-00003.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4ad2a4f8c844b27f40af4f7ebd576dc04d590cf97967cd141f9ff0e37e9b7f06
+size 4938985352

checkpoint-190/model-00002-of-00003.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0d6911d18e1230b44c2dcec0b8b93af000e046d3a3672425c51c3905490a01f9
+size 4947390880

checkpoint-190/model-00003-of-00003.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:38aea6ed7c69811e9b91746ce9ba805addb67a766dad6944d6b53d929f4af38e
+size 3590488816

checkpoint-190/model.safetensors.index.json

@@ -0,0 +1,298 @@
+{
+  "metadata": {
+    "total_size": 13476831232
+  },
+  "weight_map": {
+    "lm_head.weight": "model-00003-of-00003.safetensors",
+    "model.embed_tokens.weight": "model-00001-of-00003.safetensors",
+    "model.layers.0.input_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.0.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.0.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.0.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.1.input_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.1.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.1.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.1.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.1.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.1.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.1.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.10.input_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.10.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.10.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.10.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.10.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.10.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.10.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.10.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.10.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.11.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.11.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.11.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.11.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.11.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.11.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.11.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.11.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.11.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.12.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.12.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.12.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.12.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.12.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.12.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.12.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.12.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.12.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.13.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.13.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.13.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.13.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.13.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.13.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.13.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.13.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.13.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.14.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.14.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.14.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.14.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.14.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.14.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.14.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.14.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.14.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.15.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.15.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.15.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.15.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.15.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.15.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.15.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.15.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.15.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.16.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.16.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.16.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.16.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.16.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.16.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.16.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.16.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.16.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.17.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.17.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.17.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.17.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.17.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.17.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.17.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.17.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.17.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.18.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.18.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.18.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.18.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.18.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.18.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.18.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.18.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.18.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.19.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.19.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.19.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.19.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.19.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.19.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.19.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.19.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.19.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.2.input_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.2.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.2.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.2.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.2.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.2.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.2.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.2.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.2.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.20.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.20.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.20.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.20.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.20.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.20.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.20.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.20.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.20.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.21.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.21.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.21.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.21.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.21.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.21.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.21.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.21.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.21.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.22.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.22.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.22.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.22.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.22.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.22.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.22.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.22.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.22.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.23.input_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.23.mlp.down_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.23.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.23.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.23.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.23.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.23.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.23.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.23.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.24.input_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.24.mlp.down_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.24.mlp.gate_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.24.mlp.up_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.24.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.24.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.24.self_attn.o_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.24.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.24.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.25.input_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.25.mlp.down_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.25.mlp.gate_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.25.mlp.up_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.25.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.25.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.25.self_attn.o_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.25.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.25.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.26.input_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.26.mlp.down_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.26.mlp.gate_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.26.mlp.up_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.26.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.26.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.26.self_attn.o_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.26.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.26.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.27.input_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.27.mlp.down_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.27.mlp.gate_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.27.mlp.up_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.27.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.27.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.27.self_attn.o_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.27.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.27.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.28.input_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.28.mlp.down_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.28.mlp.gate_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.28.mlp.up_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.28.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.28.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.28.self_attn.o_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.28.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.28.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.29.input_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.29.mlp.down_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.29.mlp.gate_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.29.mlp.up_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.29.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.29.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.29.self_attn.o_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.29.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.29.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.3.input_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.3.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.3.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.3.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.3.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.3.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.3.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.3.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.3.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.30.input_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.30.mlp.down_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.30.mlp.gate_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.30.mlp.up_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.30.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.30.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.30.self_attn.o_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.30.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.30.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.31.input_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.31.mlp.down_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.31.mlp.gate_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.31.mlp.up_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.31.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.31.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.31.self_attn.o_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.31.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.31.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.4.input_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.4.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.4.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.4.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.4.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.4.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.4.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.4.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.4.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.5.input_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.5.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.5.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.5.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.5.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.5.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.5.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.5.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.5.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.6.input_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.6.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.6.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.6.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.6.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.6.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.6.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.6.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.6.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.7.input_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.7.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.7.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.7.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.7.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.7.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.7.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.7.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.7.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.8.input_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.8.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.8.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.8.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.8.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.8.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.8.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.8.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.8.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.9.input_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.9.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.9.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.9.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.9.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.9.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.9.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.9.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.9.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "model.norm.weight": "model-00003-of-00003.safetensors"
+  }
+}

checkpoint-190/rng_state_0.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5a0ef6f96a48e59aa52c4b471312c2a62378c19acc7ebbae839612b03a7d775a
+size 15984

checkpoint-190/rng_state_1.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ab11d533c0fdad46ea8b8e295ba5fdb705e078eeb88cc28f37d82913508766e9
+size 15984

checkpoint-190/rng_state_2.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:615c168147e3465ce5bfab6da2ff4afc68566ce00ec0f0c6c9fc988038a58d0a
+size 15984

checkpoint-190/rng_state_3.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:79f71e8f8674ecaef9f8cdcbf7ac457a8b8ff15b12694ba2a2fffcb4b43f0f08
+size 15984

checkpoint-190/rng_state_4.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:88cf6d674dab5545c300a55135f08ca935730a3d35e2c419fb0b333f19482c19
+size 15984

checkpoint-190/rng_state_5.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2754f2cd8824702f027870d93748b3c0491b0ecd30f1e3d8e937116b2be6151f
+size 15984

checkpoint-190/rng_state_6.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1385124ac55604598f45ea6e2d141f29456647d3e7c10d12ca64ec93d312be8d
+size 15984

checkpoint-190/rng_state_7.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:416538efaec7391fa8fe782fb15146b83e5612d9e1961292c34c53e964806873
+size 15984

checkpoint-190/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e610dd5e9f27fcc75487bd147a6e92e36d0c3d068f260ff6e94230dced07d21d
+size 1064

checkpoint-190/special_tokens_map.json

@@ -0,0 +1,24 @@
+{
+  "bos_token": {
+    "content": "<s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "</s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "</s>",
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

checkpoint-190/tokenizer.model

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

checkpoint-190/tokenizer_config.json

@@ -0,0 +1,44 @@
+{
+  "add_bos_token": true,
+  "add_eos_token": false,
+  "add_prefix_space": null,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<s>",
+  "chat_template": "{% if messages[0]['role'] == 'system' %}{% set system_message = messages[0]['content'] %}{% endif %}{% for message in messages %}{% set content = message['content'] %}{% if loop.index0 == 0 and system_message is defined %}{% set content = '<<SYS>>\n' + system_message + '\n<</SYS>>\n\n' + message['content'] %}{% endif %}{% if message['role'] == 'user' %}{{ '<s>' + '[INST] ' + content + ' [/INST]' }}{% elif message['role'] == 'assistant' %}{{ content + '</s>' }}{% endif %}{% endfor %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "</s>",
+  "legacy": false,
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "</s>",
+  "padding_side": "right",
+  "sp_model_kwargs": {},
+  "split_special_tokens": false,
+  "tokenizer_class": "LlamaTokenizer",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false
+}

checkpoint-190/trainer_state.json

@@ -0,0 +1,1553 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 4.887459807073955,
+  "eval_steps": 500,
+  "global_step": 190,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.02572347266881029,
+      "grad_norm": 268.0186767578125,
+      "learning_rate": 5.000000000000001e-07,
+      "loss": 8.3599,
+      "num_input_tokens_seen": 7120,
+      "step": 1
+    },
+    {
+      "epoch": 0.05144694533762058,
+      "grad_norm": 277.9050598144531,
+      "learning_rate": 1.0000000000000002e-06,
+      "loss": 8.1891,
+      "num_input_tokens_seen": 13888,
+      "step": 2
+    },
+    {
+      "epoch": 0.07717041800643087,
+      "grad_norm": 277.69873046875,
+      "learning_rate": 1.5e-06,
+      "loss": 8.0792,
+      "num_input_tokens_seen": 20656,
+      "step": 3
+    },
+    {
+      "epoch": 0.10289389067524116,
+      "grad_norm": 267.486328125,
+      "learning_rate": 2.0000000000000003e-06,
+      "loss": 7.9682,
+      "num_input_tokens_seen": 27184,
+      "step": 4
+    },
+    {
+      "epoch": 0.12861736334405144,
+      "grad_norm": 301.225830078125,
+      "learning_rate": 2.5e-06,
+      "loss": 6.9482,
+      "num_input_tokens_seen": 34416,
+      "step": 5
+    },
+    {
+      "epoch": 0.15434083601286175,
+      "grad_norm": 137.74415588378906,
+      "learning_rate": 3e-06,
+      "loss": 5.1505,
+      "num_input_tokens_seen": 41056,
+      "step": 6
+    },
+    {
+      "epoch": 0.18006430868167203,
+      "grad_norm": 113.7622299194336,
+      "learning_rate": 3.5e-06,
+      "loss": 4.7491,
+      "num_input_tokens_seen": 47536,
+      "step": 7
+    },
+    {
+      "epoch": 0.2057877813504823,
+      "grad_norm": 109.39883422851562,
+      "learning_rate": 4.000000000000001e-06,
+      "loss": 3.2164,
+      "num_input_tokens_seen": 54464,
+      "step": 8
+    },
+    {
+      "epoch": 0.2315112540192926,
+      "grad_norm": 119.0561294555664,
+      "learning_rate": 4.5e-06,
+      "loss": 2.7761,
+      "num_input_tokens_seen": 61520,
+      "step": 9
+    },
+    {
+      "epoch": 0.2572347266881029,
+      "grad_norm": 111.31031036376953,
+      "learning_rate": 5e-06,
+      "loss": 0.6703,
+      "num_input_tokens_seen": 68464,
+      "step": 10
+    },
+    {
+      "epoch": 0.2829581993569132,
+      "grad_norm": 39.81684494018555,
+      "learning_rate": 4.9996192378909785e-06,
+      "loss": 0.3255,
+      "num_input_tokens_seen": 75152,
+      "step": 11
+    },
+    {
+      "epoch": 0.3086816720257235,
+      "grad_norm": 41.08443069458008,
+      "learning_rate": 4.99847706754774e-06,
+      "loss": 0.3301,
+      "num_input_tokens_seen": 81888,
+      "step": 12
+    },
+    {
+      "epoch": 0.33440514469453375,
+      "grad_norm": 17.01146125793457,
+      "learning_rate": 4.9965738368864345e-06,
+      "loss": 0.2121,
+      "num_input_tokens_seen": 88688,
+      "step": 13
+    },
+    {
+      "epoch": 0.36012861736334406,
+      "grad_norm": 73.99099731445312,
+      "learning_rate": 4.993910125649561e-06,
+      "loss": 1.1565,
+      "num_input_tokens_seen": 95616,
+      "step": 14
+    },
+    {
+      "epoch": 0.3858520900321543,
+      "grad_norm": 114.83367919921875,
+      "learning_rate": 4.990486745229364e-06,
+      "loss": 0.8054,
+      "num_input_tokens_seen": 102144,
+      "step": 15
+    },
+    {
+      "epoch": 0.4115755627009646,
+      "grad_norm": 18.247591018676758,
+      "learning_rate": 4.986304738420684e-06,
+      "loss": 0.2386,
+      "num_input_tokens_seen": 109120,
+      "step": 16
+    },
+    {
+      "epoch": 0.43729903536977494,
+      "grad_norm": 27.782032012939453,
+      "learning_rate": 4.981365379103306e-06,
+      "loss": 0.3161,
+      "num_input_tokens_seen": 115744,
+      "step": 17
+    },
+    {
+      "epoch": 0.4630225080385852,
+      "grad_norm": 21.16512107849121,
+      "learning_rate": 4.975670171853926e-06,
+      "loss": 0.2773,
+      "num_input_tokens_seen": 122704,
+      "step": 18
+    },
+    {
+      "epoch": 0.4887459807073955,
+      "grad_norm": 6.738564491271973,
+      "learning_rate": 4.9692208514878445e-06,
+      "loss": 0.2062,
+      "num_input_tokens_seen": 129552,
+      "step": 19
+    },
+    {
+      "epoch": 0.5144694533762058,
+      "grad_norm": 3.6603872776031494,
+      "learning_rate": 4.962019382530521e-06,
+      "loss": 0.1837,
+      "num_input_tokens_seen": 136544,
+      "step": 20
+    },
+    {
+      "epoch": 0.5401929260450161,
+      "grad_norm": 10.339789390563965,
+      "learning_rate": 4.9540679586191605e-06,
+      "loss": 0.1735,
+      "num_input_tokens_seen": 143488,
+      "step": 21
+    },
+    {
+      "epoch": 0.5659163987138264,
+      "grad_norm": 4.833702087402344,
+      "learning_rate": 4.9453690018345144e-06,
+      "loss": 0.1588,
+      "num_input_tokens_seen": 150224,
+      "step": 22
+    },
+    {
+      "epoch": 0.5916398713826366,
+      "grad_norm": 4.9161152839660645,
+      "learning_rate": 4.935925161963089e-06,
+      "loss": 0.1443,
+      "num_input_tokens_seen": 157232,
+      "step": 23
+    },
+    {
+      "epoch": 0.617363344051447,
+      "grad_norm": 9.033807754516602,
+      "learning_rate": 4.925739315689991e-06,
+      "loss": 0.157,
+      "num_input_tokens_seen": 163776,
+      "step": 24
+    },
+    {
+      "epoch": 0.6430868167202572,
+      "grad_norm": 4.518654823303223,
+      "learning_rate": 4.914814565722671e-06,
+      "loss": 0.1199,
+      "num_input_tokens_seen": 170352,
+      "step": 25
+    },
+    {
+      "epoch": 0.6688102893890675,
+      "grad_norm": 10.43909740447998,
+      "learning_rate": 4.903154239845798e-06,
+      "loss": 0.1539,
+      "num_input_tokens_seen": 177120,
+      "step": 26
+    },
+    {
+      "epoch": 0.6945337620578779,
+      "grad_norm": 9.000858306884766,
+      "learning_rate": 4.890761889907589e-06,
+      "loss": 0.1208,
+      "num_input_tokens_seen": 184096,
+      "step": 27
+    },
+    {
+      "epoch": 0.7202572347266881,
+      "grad_norm": 3.3685858249664307,
+      "learning_rate": 4.8776412907378845e-06,
+      "loss": 0.0954,
+      "num_input_tokens_seen": 191040,
+      "step": 28
+    },
+    {
+      "epoch": 0.7459807073954984,
+      "grad_norm": 6.876232147216797,
+      "learning_rate": 4.863796438998293e-06,
+      "loss": 0.1387,
+      "num_input_tokens_seen": 198064,
+      "step": 29
+    },
+    {
+      "epoch": 0.7717041800643086,
+      "grad_norm": 11.019433975219727,
+      "learning_rate": 4.849231551964771e-06,
+      "loss": 0.1484,
+      "num_input_tokens_seen": 205136,
+      "step": 30
+    },
+    {
+      "epoch": 0.797427652733119,
+      "grad_norm": 4.780524730682373,
+      "learning_rate": 4.833951066243004e-06,
+      "loss": 0.0998,
+      "num_input_tokens_seen": 212000,
+      "step": 31
+    },
+    {
+      "epoch": 0.8231511254019293,
+      "grad_norm": 3.613060235977173,
+      "learning_rate": 4.817959636416969e-06,
+      "loss": 0.1068,
+      "num_input_tokens_seen": 218720,
+      "step": 32
+    },
+    {
+      "epoch": 0.8488745980707395,
+      "grad_norm": 5.576949596405029,
+      "learning_rate": 4.801262133631101e-06,
+      "loss": 0.0801,
+      "num_input_tokens_seen": 225856,
+      "step": 33
+    },
+    {
+      "epoch": 0.8745980707395499,
+      "grad_norm": 3.3204915523529053,
+      "learning_rate": 4.783863644106502e-06,
+      "loss": 0.1066,
+      "num_input_tokens_seen": 232640,
+      "step": 34
+    },
+    {
+      "epoch": 0.9003215434083601,
+      "grad_norm": 3.090853452682495,
+      "learning_rate": 4.765769467591626e-06,
+      "loss": 0.1038,
+      "num_input_tokens_seen": 239504,
+      "step": 35
+    },
+    {
+      "epoch": 0.9260450160771704,
+      "grad_norm": 6.979610443115234,
+      "learning_rate": 4.746985115747918e-06,
+      "loss": 0.106,
+      "num_input_tokens_seen": 246288,
+      "step": 36
+    },
+    {
+      "epoch": 0.9517684887459807,
+      "grad_norm": 3.491868495941162,
+      "learning_rate": 4.72751631047092e-06,
+      "loss": 0.1107,
+      "num_input_tokens_seen": 253136,
+      "step": 37
+    },
+    {
+      "epoch": 0.977491961414791,
+      "grad_norm": 6.108020782470703,
+      "learning_rate": 4.707368982147318e-06,
+      "loss": 0.1372,
+      "num_input_tokens_seen": 260160,
+      "step": 38
+    },
+    {
+      "epoch": 1.0032154340836013,
+      "grad_norm": 1.9764081239700317,
+      "learning_rate": 4.68654926784849e-06,
+      "loss": 0.0816,
+      "num_input_tokens_seen": 267120,
+      "step": 39
+    },
+    {
+      "epoch": 1.0289389067524115,
+      "grad_norm": 2.5603554248809814,
+      "learning_rate": 4.665063509461098e-06,
+      "loss": 0.0743,
+      "num_input_tokens_seen": 274112,
+      "step": 40
+    },
+    {
+      "epoch": 1.0546623794212218,
+      "grad_norm": 5.457398414611816,
+      "learning_rate": 4.642918251755281e-06,
+      "loss": 0.072,
+      "num_input_tokens_seen": 281136,
+      "step": 41
+    },
+    {
+      "epoch": 1.0803858520900322,
+      "grad_norm": 4.488720417022705,
+      "learning_rate": 4.620120240391065e-06,
+      "loss": 0.0596,
+      "num_input_tokens_seen": 288048,
+      "step": 42
+    },
+    {
+      "epoch": 1.1061093247588425,
+      "grad_norm": 2.8057987689971924,
+      "learning_rate": 4.596676419863561e-06,
+      "loss": 0.0544,
+      "num_input_tokens_seen": 295120,
+      "step": 43
+    },
+    {
+      "epoch": 1.1318327974276527,
+      "grad_norm": 2.921278476715088,
+      "learning_rate": 4.572593931387604e-06,
+      "loss": 0.0342,
+      "num_input_tokens_seen": 302000,
+      "step": 44
+    },
+    {
+      "epoch": 1.157556270096463,
+      "grad_norm": 2.9281697273254395,
+      "learning_rate": 4.54788011072248e-06,
+      "loss": 0.0394,
+      "num_input_tokens_seen": 308672,
+      "step": 45
+    },
+    {
+      "epoch": 1.1832797427652733,
+      "grad_norm": 4.087917327880859,
+      "learning_rate": 4.522542485937369e-06,
+      "loss": 0.0196,
+      "num_input_tokens_seen": 315600,
+      "step": 46
+    },
+    {
+      "epoch": 1.2090032154340835,
+      "grad_norm": 3.390829086303711,
+      "learning_rate": 4.496588775118232e-06,
+      "loss": 0.0411,
+      "num_input_tokens_seen": 322464,
+      "step": 47
+    },
+    {
+      "epoch": 1.234726688102894,
+      "grad_norm": 4.550433158874512,
+      "learning_rate": 4.470026884016805e-06,
+      "loss": 0.0257,
+      "num_input_tokens_seen": 329024,
+      "step": 48
+    },
+    {
+      "epoch": 1.2604501607717042,
+      "grad_norm": 6.6677446365356445,
+      "learning_rate": 4.442864903642428e-06,
+      "loss": 0.0289,
+      "num_input_tokens_seen": 336032,
+      "step": 49
+    },
+    {
+      "epoch": 1.2861736334405145,
+      "grad_norm": 7.874734878540039,
+      "learning_rate": 4.415111107797445e-06,
+      "loss": 0.1193,
+      "num_input_tokens_seen": 342704,
+      "step": 50
+    },
+    {
+      "epoch": 1.3118971061093248,
+      "grad_norm": 5.971132278442383,
+      "learning_rate": 4.386773950556931e-06,
+      "loss": 0.0883,
+      "num_input_tokens_seen": 349472,
+      "step": 51
+    },
+    {
+      "epoch": 1.337620578778135,
+      "grad_norm": 4.0566725730896,
+      "learning_rate": 4.357862063693486e-06,
+      "loss": 0.0377,
+      "num_input_tokens_seen": 356272,
+      "step": 52
+    },
+    {
+      "epoch": 1.3633440514469453,
+      "grad_norm": 5.443573474884033,
+      "learning_rate": 4.328384254047927e-06,
+      "loss": 0.0602,
+      "num_input_tokens_seen": 363040,
+      "step": 53
+    },
+    {
+      "epoch": 1.3890675241157555,
+      "grad_norm": 6.038252353668213,
+      "learning_rate": 4.2983495008466285e-06,
+      "loss": 0.083,
+      "num_input_tokens_seen": 369664,
+      "step": 54
+    },
+    {
+      "epoch": 1.414790996784566,
+      "grad_norm": 2.8046696186065674,
+      "learning_rate": 4.267766952966369e-06,
+      "loss": 0.0358,
+      "num_input_tokens_seen": 376704,
+      "step": 55
+    },
+    {
+      "epoch": 1.4405144694533762,
+      "grad_norm": 2.451719284057617,
+      "learning_rate": 4.236645926147493e-06,
+      "loss": 0.0321,
+      "num_input_tokens_seen": 383600,
+      "step": 56
+    },
+    {
+      "epoch": 1.4662379421221865,
+      "grad_norm": 3.4111475944519043,
+      "learning_rate": 4.204995900156247e-06,
+      "loss": 0.0452,
+      "num_input_tokens_seen": 390592,
+      "step": 57
+    },
+    {
+      "epoch": 1.4919614147909968,
+      "grad_norm": 3.065139055252075,
+      "learning_rate": 4.172826515897146e-06,
+      "loss": 0.0915,
+      "num_input_tokens_seen": 397360,
+      "step": 58
+    },
+    {
+      "epoch": 1.517684887459807,
+      "grad_norm": 3.0692198276519775,
+      "learning_rate": 4.140147572476269e-06,
+      "loss": 0.0651,
+      "num_input_tokens_seen": 404048,
+      "step": 59
+    },
+    {
+      "epoch": 1.5434083601286175,
+      "grad_norm": 3.6201350688934326,
+      "learning_rate": 4.106969024216348e-06,
+      "loss": 0.0868,
+      "num_input_tokens_seen": 410960,
+      "step": 60
+    },
+    {
+      "epoch": 1.5691318327974275,
+      "grad_norm": 2.289900779724121,
+      "learning_rate": 4.073300977624594e-06,
+      "loss": 0.0554,
+      "num_input_tokens_seen": 417888,
+      "step": 61
+    },
+    {
+      "epoch": 1.594855305466238,
+      "grad_norm": 1.4743808507919312,
+      "learning_rate": 4.039153688314146e-06,
+      "loss": 0.0336,
+      "num_input_tokens_seen": 424880,
+      "step": 62
+    },
+    {
+      "epoch": 1.6205787781350482,
+      "grad_norm": 2.1891098022460938,
+      "learning_rate": 4.0045375578801216e-06,
+      "loss": 0.0455,
+      "num_input_tokens_seen": 432000,
+      "step": 63
+    },
+    {
+      "epoch": 1.6463022508038585,
+      "grad_norm": 1.8203191757202148,
+      "learning_rate": 3.969463130731183e-06,
+      "loss": 0.0406,
+      "num_input_tokens_seen": 438672,
+      "step": 64
+    },
+    {
+      "epoch": 1.6720257234726688,
+      "grad_norm": 2.453317165374756,
+      "learning_rate": 3.933941090877615e-06,
+      "loss": 0.0461,
+      "num_input_tokens_seen": 445440,
+      "step": 65
+    },
+    {
+      "epoch": 1.697749196141479,
+      "grad_norm": 2.035806894302368,
+      "learning_rate": 3.897982258676867e-06,
+      "loss": 0.0466,
+      "num_input_tokens_seen": 452064,
+      "step": 66
+    },
+    {
+      "epoch": 1.7234726688102895,
+      "grad_norm": 3.169175386428833,
+      "learning_rate": 3.861597587537568e-06,
+      "loss": 0.0382,
+      "num_input_tokens_seen": 458992,
+      "step": 67
+    },
+    {
+      "epoch": 1.7491961414790995,
+      "grad_norm": 2.0320234298706055,
+      "learning_rate": 3.824798160583012e-06,
+      "loss": 0.0426,
+      "num_input_tokens_seen": 465568,
+      "step": 68
+    },
+    {
+      "epoch": 1.77491961414791,
+      "grad_norm": 4.094840049743652,
+      "learning_rate": 3.787595187275136e-06,
+      "loss": 0.0264,
+      "num_input_tokens_seen": 472496,
+      "step": 69
+    },
+    {
+      "epoch": 1.8006430868167203,
+      "grad_norm": 4.245150089263916,
+      "learning_rate": 3.7500000000000005e-06,
+      "loss": 0.0567,
+      "num_input_tokens_seen": 479392,
+      "step": 70
+    },
+    {
+      "epoch": 1.8263665594855305,
+      "grad_norm": 3.226271152496338,
+      "learning_rate": 3.7120240506158433e-06,
+      "loss": 0.0688,
+      "num_input_tokens_seen": 486416,
+      "step": 71
+    },
+    {
+      "epoch": 1.852090032154341,
+      "grad_norm": 2.070322275161743,
+      "learning_rate": 3.6736789069647273e-06,
+      "loss": 0.0351,
+      "num_input_tokens_seen": 492896,
+      "step": 72
+    },
+    {
+      "epoch": 1.877813504823151,
+      "grad_norm": 2.4622035026550293,
+      "learning_rate": 3.634976249348867e-06,
+      "loss": 0.0246,
+      "num_input_tokens_seen": 499760,
+      "step": 73
+    },
+    {
+      "epoch": 1.9035369774919615,
+      "grad_norm": 3.902181625366211,
+      "learning_rate": 3.595927866972694e-06,
+      "loss": 0.0364,
+      "num_input_tokens_seen": 506816,
+      "step": 74
+    },
+    {
+      "epoch": 1.9292604501607717,
+      "grad_norm": 4.0147480964660645,
+      "learning_rate": 3.556545654351749e-06,
+      "loss": 0.0352,
+      "num_input_tokens_seen": 513648,
+      "step": 75
+    },
+    {
+      "epoch": 1.954983922829582,
+      "grad_norm": 3.6791763305664062,
+      "learning_rate": 3.516841607689501e-06,
+      "loss": 0.0915,
+      "num_input_tokens_seen": 520480,
+      "step": 76
+    },
+    {
+      "epoch": 1.9807073954983923,
+      "grad_norm": 1.641350269317627,
+      "learning_rate": 3.476827821223184e-06,
+      "loss": 0.0327,
+      "num_input_tokens_seen": 527056,
+      "step": 77
+    },
+    {
+      "epoch": 2.0064308681672025,
+      "grad_norm": 3.6227402687072754,
+      "learning_rate": 3.436516483539781e-06,
+      "loss": 0.0448,
+      "num_input_tokens_seen": 534112,
+      "step": 78
+    },
+    {
+      "epoch": 2.032154340836013,
+      "grad_norm": 1.654789924621582,
+      "learning_rate": 3.39591987386325e-06,
+      "loss": 0.0186,
+      "num_input_tokens_seen": 541024,
+      "step": 79
+    },
+    {
+      "epoch": 2.057877813504823,
+      "grad_norm": 2.0418519973754883,
+      "learning_rate": 3.3550503583141726e-06,
+      "loss": 0.0342,
+      "num_input_tokens_seen": 547888,
+      "step": 80
+    },
+    {
+      "epoch": 2.0836012861736335,
+      "grad_norm": 1.0242717266082764,
+      "learning_rate": 3.313920386142892e-06,
+      "loss": 0.0079,
+      "num_input_tokens_seen": 554592,
+      "step": 81
+    },
+    {
+      "epoch": 2.1093247588424435,
+      "grad_norm": 1.59933602809906,
+      "learning_rate": 3.272542485937369e-06,
+      "loss": 0.0177,
+      "num_input_tokens_seen": 561296,
+      "step": 82
+    },
+    {
+      "epoch": 2.135048231511254,
+      "grad_norm": 1.83909273147583,
+      "learning_rate": 3.230929261806842e-06,
+      "loss": 0.0139,
+      "num_input_tokens_seen": 567872,
+      "step": 83
+    },
+    {
+      "epoch": 2.1607717041800645,
+      "grad_norm": 1.933048963546753,
+      "learning_rate": 3.189093389542498e-06,
+      "loss": 0.0103,
+      "num_input_tokens_seen": 575072,
+      "step": 84
+    },
+    {
+      "epoch": 2.1864951768488745,
+      "grad_norm": 2.1397433280944824,
+      "learning_rate": 3.147047612756302e-06,
+      "loss": 0.0221,
+      "num_input_tokens_seen": 582080,
+      "step": 85
+    },
+    {
+      "epoch": 2.212218649517685,
+      "grad_norm": 0.7958673238754272,
+      "learning_rate": 3.1048047389991693e-06,
+      "loss": 0.0021,
+      "num_input_tokens_seen": 588816,
+      "step": 86
+    },
+    {
+      "epoch": 2.237942122186495,
+      "grad_norm": 1.8060617446899414,
+      "learning_rate": 3.062377635859663e-06,
+      "loss": 0.011,
+      "num_input_tokens_seen": 596032,
+      "step": 87
+    },
+    {
+      "epoch": 2.2636655948553055,
+      "grad_norm": 1.7997971773147583,
+      "learning_rate": 3.019779227044398e-06,
+      "loss": 0.0081,
+      "num_input_tokens_seen": 602672,
+      "step": 88
+    },
+    {
+      "epoch": 2.289389067524116,
+      "grad_norm": 2.0229434967041016,
+      "learning_rate": 2.9770224884413625e-06,
+      "loss": 0.0149,
+      "num_input_tokens_seen": 609424,
+      "step": 89
+    },
+    {
+      "epoch": 2.315112540192926,
+      "grad_norm": 0.5248342156410217,
+      "learning_rate": 2.9341204441673267e-06,
+      "loss": 0.001,
+      "num_input_tokens_seen": 616448,
+      "step": 90
+    },
+    {
+      "epoch": 2.3408360128617365,
+      "grad_norm": 1.1899443864822388,
+      "learning_rate": 2.8910861626005774e-06,
+      "loss": 0.007,
+      "num_input_tokens_seen": 623296,
+      "step": 91
+    },
+    {
+      "epoch": 2.3665594855305465,
+      "grad_norm": 3.1920082569122314,
+      "learning_rate": 2.847932752400164e-06,
+      "loss": 0.0089,
+      "num_input_tokens_seen": 630064,
+      "step": 92
+    },
+    {
+      "epoch": 2.392282958199357,
+      "grad_norm": 0.37337368726730347,
+      "learning_rate": 2.804673358512869e-06,
+      "loss": 0.0013,
+      "num_input_tokens_seen": 637088,
+      "step": 93
+    },
+    {
+      "epoch": 2.418006430868167,
+      "grad_norm": 4.124266147613525,
+      "learning_rate": 2.761321158169134e-06,
+      "loss": 0.0267,
+      "num_input_tokens_seen": 644080,
+      "step": 94
+    },
+    {
+      "epoch": 2.4437299035369775,
+      "grad_norm": 1.9108864068984985,
+      "learning_rate": 2.717889356869146e-06,
+      "loss": 0.0171,
+      "num_input_tokens_seen": 650848,
+      "step": 95
+    },
+    {
+      "epoch": 2.469453376205788,
+      "grad_norm": 3.225116729736328,
+      "learning_rate": 2.6743911843603134e-06,
+      "loss": 0.0375,
+      "num_input_tokens_seen": 657424,
+      "step": 96
+    },
+    {
+      "epoch": 2.495176848874598,
+      "grad_norm": 1.8133978843688965,
+      "learning_rate": 2.6308398906073603e-06,
+      "loss": 0.0101,
+      "num_input_tokens_seen": 664128,
+      "step": 97
+    },
+    {
+      "epoch": 2.5209003215434085,
+      "grad_norm": 3.179337739944458,
+      "learning_rate": 2.587248741756253e-06,
+      "loss": 0.0282,
+      "num_input_tokens_seen": 671120,
+      "step": 98
+    },
+    {
+      "epoch": 2.5466237942122185,
+      "grad_norm": 1.5852500200271606,
+      "learning_rate": 2.543631016093209e-06,
+      "loss": 0.0069,
+      "num_input_tokens_seen": 677920,
+      "step": 99
+    },
+    {
+      "epoch": 2.572347266881029,
+      "grad_norm": 2.1428685188293457,
+      "learning_rate": 2.5e-06,
+      "loss": 0.0135,
+      "num_input_tokens_seen": 684960,
+      "step": 100
+    },
+    {
+      "epoch": 2.598070739549839,
+      "grad_norm": 1.0775537490844727,
+      "learning_rate": 2.4563689839067913e-06,
+      "loss": 0.0062,
+      "num_input_tokens_seen": 691856,
+      "step": 101
+    },
+    {
+      "epoch": 2.6237942122186495,
+      "grad_norm": 1.5073256492614746,
+      "learning_rate": 2.4127512582437486e-06,
+      "loss": 0.005,
+      "num_input_tokens_seen": 698512,
+      "step": 102
+    },
+    {
+      "epoch": 2.64951768488746,
+      "grad_norm": 1.7251828908920288,
+      "learning_rate": 2.3691601093926406e-06,
+      "loss": 0.0285,
+      "num_input_tokens_seen": 705440,
+      "step": 103
+    },
+    {
+      "epoch": 2.67524115755627,
+      "grad_norm": 2.271348237991333,
+      "learning_rate": 2.325608815639687e-06,
+      "loss": 0.0225,
+      "num_input_tokens_seen": 712528,
+      "step": 104
+    },
+    {
+      "epoch": 2.7009646302250805,
+      "grad_norm": 0.9558768272399902,
+      "learning_rate": 2.2821106431308546e-06,
+      "loss": 0.028,
+      "num_input_tokens_seen": 719168,
+      "step": 105
+    },
+    {
+      "epoch": 2.7266881028938905,
+      "grad_norm": 2.1378886699676514,
+      "learning_rate": 2.238678841830867e-06,
+      "loss": 0.0176,
+      "num_input_tokens_seen": 725904,
+      "step": 106
+    },
+    {
+      "epoch": 2.752411575562701,
+      "grad_norm": 0.9674012660980225,
+      "learning_rate": 2.195326641487132e-06,
+      "loss": 0.0047,
+      "num_input_tokens_seen": 732480,
+      "step": 107
+    },
+    {
+      "epoch": 2.778135048231511,
+      "grad_norm": 0.6068861484527588,
+      "learning_rate": 2.1520672475998374e-06,
+      "loss": 0.0135,
+      "num_input_tokens_seen": 739184,
+      "step": 108
+    },
+    {
+      "epoch": 2.8038585209003215,
+      "grad_norm": 0.833519458770752,
+      "learning_rate": 2.1089138373994226e-06,
+      "loss": 0.0044,
+      "num_input_tokens_seen": 746320,
+      "step": 109
+    },
+    {
+      "epoch": 2.829581993569132,
+      "grad_norm": 1.4018948078155518,
+      "learning_rate": 2.0658795558326745e-06,
+      "loss": 0.0252,
+      "num_input_tokens_seen": 753136,
+      "step": 110
+    },
+    {
+      "epoch": 2.855305466237942,
+      "grad_norm": 1.5698492527008057,
+      "learning_rate": 2.022977511558638e-06,
+      "loss": 0.0249,
+      "num_input_tokens_seen": 760096,
+      "step": 111
+    },
+    {
+      "epoch": 2.8810289389067525,
+      "grad_norm": 2.051377534866333,
+      "learning_rate": 1.9802207729556023e-06,
+      "loss": 0.0146,
+      "num_input_tokens_seen": 767104,
+      "step": 112
+    },
+    {
+      "epoch": 2.906752411575563,
+      "grad_norm": 0.48683854937553406,
+      "learning_rate": 1.937622364140338e-06,
+      "loss": 0.0044,
+      "num_input_tokens_seen": 773872,
+      "step": 113
+    },
+    {
+      "epoch": 2.932475884244373,
+      "grad_norm": 0.9895896911621094,
+      "learning_rate": 1.895195261000831e-06,
+      "loss": 0.0054,
+      "num_input_tokens_seen": 780736,
+      "step": 114
+    },
+    {
+      "epoch": 2.958199356913183,
+      "grad_norm": 0.8025338053703308,
+      "learning_rate": 1.852952387243698e-06,
+      "loss": 0.0106,
+      "num_input_tokens_seen": 787424,
+      "step": 115
+    },
+    {
+      "epoch": 2.9839228295819935,
+      "grad_norm": 1.7318856716156006,
+      "learning_rate": 1.8109066104575023e-06,
+      "loss": 0.0167,
+      "num_input_tokens_seen": 794224,
+      "step": 116
+    },
+    {
+      "epoch": 3.009646302250804,
+      "grad_norm": 1.5407195091247559,
+      "learning_rate": 1.7690707381931585e-06,
+      "loss": 0.009,
+      "num_input_tokens_seen": 801088,
+      "step": 117
+    },
+    {
+      "epoch": 3.035369774919614,
+      "grad_norm": 0.25947141647338867,
+      "learning_rate": 1.7274575140626318e-06,
+      "loss": 0.0024,
+      "num_input_tokens_seen": 808048,
+      "step": 118
+    },
+    {
+      "epoch": 3.0610932475884245,
+      "grad_norm": 2.6516857147216797,
+      "learning_rate": 1.686079613857109e-06,
+      "loss": 0.0235,
+      "num_input_tokens_seen": 814800,
+      "step": 119
+    },
+    {
+      "epoch": 3.0868167202572345,
+      "grad_norm": 0.9633734822273254,
+      "learning_rate": 1.6449496416858285e-06,
+      "loss": 0.0179,
+      "num_input_tokens_seen": 821536,
+      "step": 120
+    },
+    {
+      "epoch": 3.112540192926045,
+      "grad_norm": 1.0766749382019043,
+      "learning_rate": 1.6040801261367494e-06,
+      "loss": 0.0059,
+      "num_input_tokens_seen": 828128,
+      "step": 121
+    },
+    {
+      "epoch": 3.1382636655948555,
+      "grad_norm": 0.2934500277042389,
+      "learning_rate": 1.56348351646022e-06,
+      "loss": 0.0017,
+      "num_input_tokens_seen": 834608,
+      "step": 122
+    },
+    {
+      "epoch": 3.1639871382636655,
+      "grad_norm": 0.3029981553554535,
+      "learning_rate": 1.5231721787768162e-06,
+      "loss": 0.0018,
+      "num_input_tokens_seen": 841360,
+      "step": 123
+    },
+    {
+      "epoch": 3.189710610932476,
+      "grad_norm": 0.44817763566970825,
+      "learning_rate": 1.4831583923105e-06,
+      "loss": 0.0032,
+      "num_input_tokens_seen": 848496,
+      "step": 124
+    },
+    {
+      "epoch": 3.215434083601286,
+      "grad_norm": 0.21245715022087097,
+      "learning_rate": 1.443454345648252e-06,
+      "loss": 0.0019,
+      "num_input_tokens_seen": 855424,
+      "step": 125
+    },
+    {
+      "epoch": 3.2411575562700965,
+      "grad_norm": 0.22589029371738434,
+      "learning_rate": 1.4040721330273063e-06,
+      "loss": 0.0014,
+      "num_input_tokens_seen": 862224,
+      "step": 126
+    },
+    {
+      "epoch": 3.266881028938907,
+      "grad_norm": 0.5972980856895447,
+      "learning_rate": 1.3650237506511333e-06,
+      "loss": 0.0052,
+      "num_input_tokens_seen": 869312,
+      "step": 127
+    },
+    {
+      "epoch": 3.292604501607717,
+      "grad_norm": 0.06931939721107483,
+      "learning_rate": 1.3263210930352737e-06,
+      "loss": 0.0005,
+      "num_input_tokens_seen": 876272,
+      "step": 128
+    },
+    {
+      "epoch": 3.3183279742765275,
+      "grad_norm": 1.2623660564422607,
+      "learning_rate": 1.2879759493841577e-06,
+      "loss": 0.0131,
+      "num_input_tokens_seen": 883072,
+      "step": 129
+    },
+    {
+      "epoch": 3.3440514469453375,
+      "grad_norm": 0.5440672636032104,
+      "learning_rate": 1.2500000000000007e-06,
+      "loss": 0.0009,
+      "num_input_tokens_seen": 889920,
+      "step": 130
+    },
+    {
+      "epoch": 3.369774919614148,
+      "grad_norm": 0.6807874441146851,
+      "learning_rate": 1.2124048127248644e-06,
+      "loss": 0.0057,
+      "num_input_tokens_seen": 896896,
+      "step": 131
+    },
+    {
+      "epoch": 3.395498392282958,
+      "grad_norm": 0.03898243606090546,
+      "learning_rate": 1.1752018394169882e-06,
+      "loss": 0.0002,
+      "num_input_tokens_seen": 903600,
+      "step": 132
+    },
+    {
+      "epoch": 3.4212218649517685,
+      "grad_norm": 0.0243828147649765,
+      "learning_rate": 1.1384024124624324e-06,
+      "loss": 0.0002,
+      "num_input_tokens_seen": 910208,
+      "step": 133
+    },
+    {
+      "epoch": 3.446945337620579,
+      "grad_norm": 2.5577075481414795,
+      "learning_rate": 1.1020177413231334e-06,
+      "loss": 0.0145,
+      "num_input_tokens_seen": 917328,
+      "step": 134
+    },
+    {
+      "epoch": 3.472668810289389,
+      "grad_norm": 1.2599172592163086,
+      "learning_rate": 1.0660589091223854e-06,
+      "loss": 0.0034,
+      "num_input_tokens_seen": 924192,
+      "step": 135
+    },
+    {
+      "epoch": 3.4983922829581995,
+      "grad_norm": 1.3292278051376343,
+      "learning_rate": 1.0305368692688175e-06,
+      "loss": 0.0156,
+      "num_input_tokens_seen": 930784,
+      "step": 136
+    },
+    {
+      "epoch": 3.5241157556270095,
+      "grad_norm": 0.2977862060070038,
+      "learning_rate": 9.95462442119879e-07,
+      "loss": 0.0013,
+      "num_input_tokens_seen": 937856,
+      "step": 137
+    },
+    {
+      "epoch": 3.54983922829582,
+      "grad_norm": 0.3527968227863312,
+      "learning_rate": 9.608463116858544e-07,
+      "loss": 0.0007,
+      "num_input_tokens_seen": 944640,
+      "step": 138
+    },
+    {
+      "epoch": 3.57556270096463,
+      "grad_norm": 0.16851164400577545,
+      "learning_rate": 9.266990223754069e-07,
+      "loss": 0.0005,
+      "num_input_tokens_seen": 951504,
+      "step": 139
+    },
+    {
+      "epoch": 3.6012861736334405,
+      "grad_norm": 0.4394027292728424,
+      "learning_rate": 8.930309757836517e-07,
+      "loss": 0.0034,
+      "num_input_tokens_seen": 958240,
+      "step": 140
+    },
+    {
+      "epoch": 3.627009646302251,
+      "grad_norm": 0.0209315475076437,
+      "learning_rate": 8.598524275237321e-07,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 964912,
+      "step": 141
+    },
+    {
+      "epoch": 3.652733118971061,
+      "grad_norm": 0.35701486468315125,
+      "learning_rate": 8.271734841028553e-07,
+      "loss": 0.001,
+      "num_input_tokens_seen": 971872,
+      "step": 142
+    },
+    {
+      "epoch": 3.6784565916398715,
+      "grad_norm": 2.0450756549835205,
+      "learning_rate": 7.950040998437541e-07,
+      "loss": 0.0123,
+      "num_input_tokens_seen": 978640,
+      "step": 143
+    },
+    {
+      "epoch": 3.7041800643086815,
+      "grad_norm": 0.057205893099308014,
+      "learning_rate": 7.633540738525066e-07,
+      "loss": 0.0002,
+      "num_input_tokens_seen": 985328,
+      "step": 144
+    },
+    {
+      "epoch": 3.729903536977492,
+      "grad_norm": 1.7251015901565552,
+      "learning_rate": 7.322330470336314e-07,
+      "loss": 0.011,
+      "num_input_tokens_seen": 992224,
+      "step": 145
+    },
+    {
+      "epoch": 3.755627009646302,
+      "grad_norm": 0.417669415473938,
+      "learning_rate": 7.016504991533727e-07,
+      "loss": 0.0008,
+      "num_input_tokens_seen": 998688,
+      "step": 146
+    },
+    {
+      "epoch": 3.7813504823151125,
+      "grad_norm": 0.06796720623970032,
+      "learning_rate": 6.716157459520739e-07,
+      "loss": 0.0003,
+      "num_input_tokens_seen": 1006032,
+      "step": 147
+    },
+    {
+      "epoch": 3.807073954983923,
+      "grad_norm": 0.5570574998855591,
+      "learning_rate": 6.421379363065142e-07,
+      "loss": 0.0018,
+      "num_input_tokens_seen": 1012944,
+      "step": 148
+    },
+    {
+      "epoch": 3.832797427652733,
+      "grad_norm": 0.557479977607727,
+      "learning_rate": 6.1322604944307e-07,
+      "loss": 0.0016,
+      "num_input_tokens_seen": 1019856,
+      "step": 149
+    },
+    {
+      "epoch": 3.8585209003215435,
+      "grad_norm": 0.48776909708976746,
+      "learning_rate": 5.848888922025553e-07,
+      "loss": 0.0021,
+      "num_input_tokens_seen": 1026656,
+      "step": 150
+    },
+    {
+      "epoch": 3.884244372990354,
+      "grad_norm": 0.025394350290298462,
+      "learning_rate": 5.571350963575728e-07,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 1033696,
+      "step": 151
+    },
+    {
+      "epoch": 3.909967845659164,
+      "grad_norm": 0.03195232152938843,
+      "learning_rate": 5.299731159831953e-07,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 1040240,
+      "step": 152
+    },
+    {
+      "epoch": 3.935691318327974,
+      "grad_norm": 0.0676019936800003,
+      "learning_rate": 5.034112248817685e-07,
+      "loss": 0.0003,
+      "num_input_tokens_seen": 1046992,
+      "step": 153
+    },
+    {
+      "epoch": 3.9614147909967845,
+      "grad_norm": 0.04871657118201256,
+      "learning_rate": 4.774575140626317e-07,
+      "loss": 0.0002,
+      "num_input_tokens_seen": 1053936,
+      "step": 154
+    },
+    {
+      "epoch": 3.987138263665595,
+      "grad_norm": 0.08272106945514679,
+      "learning_rate": 4.5211988927752026e-07,
+      "loss": 0.0002,
+      "num_input_tokens_seen": 1060576,
+      "step": 155
+    },
+    {
+      "epoch": 4.012861736334405,
+      "grad_norm": 0.009170151315629482,
+      "learning_rate": 4.27406068612396e-07,
+      "loss": 0.0,
+      "num_input_tokens_seen": 1067648,
+      "step": 156
+    },
+    {
+      "epoch": 4.038585209003215,
+      "grad_norm": 0.10544967651367188,
+      "learning_rate": 4.033235801364402e-07,
+      "loss": 0.0003,
+      "num_input_tokens_seen": 1074512,
+      "step": 157
+    },
+    {
+      "epoch": 4.064308681672026,
+      "grad_norm": 0.04709336906671524,
+      "learning_rate": 3.798797596089351e-07,
+      "loss": 0.0002,
+      "num_input_tokens_seen": 1081296,
+      "step": 158
+    },
+    {
+      "epoch": 4.090032154340836,
+      "grad_norm": 0.042879387736320496,
+      "learning_rate": 3.5708174824471947e-07,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 1087888,
+      "step": 159
+    },
+    {
+      "epoch": 4.115755627009646,
+      "grad_norm": 0.03649509325623512,
+      "learning_rate": 3.3493649053890325e-07,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 1094960,
+      "step": 160
+    },
+    {
+      "epoch": 4.141479099678457,
+      "grad_norm": 0.01273419987410307,
+      "learning_rate": 3.134507321515107e-07,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 1101776,
+      "step": 161
+    },
+    {
+      "epoch": 4.167202572347267,
+      "grad_norm": 0.006498055998235941,
+      "learning_rate": 2.9263101785268253e-07,
+      "loss": 0.0,
+      "num_input_tokens_seen": 1108736,
+      "step": 162
+    },
+    {
+      "epoch": 4.192926045016077,
+      "grad_norm": 0.027757082134485245,
+      "learning_rate": 2.7248368952908055e-07,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 1115744,
+      "step": 163
+    },
+    {
+      "epoch": 4.218649517684887,
+      "grad_norm": 0.005632288288325071,
+      "learning_rate": 2.53014884252083e-07,
+      "loss": 0.0,
+      "num_input_tokens_seen": 1122592,
+      "step": 164
+    },
+    {
+      "epoch": 4.244372990353698,
+      "grad_norm": 0.014449907466769218,
+      "learning_rate": 2.3423053240837518e-07,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 1129136,
+      "step": 165
+    },
+    {
+      "epoch": 4.270096463022508,
+      "grad_norm": 0.013245908543467522,
+      "learning_rate": 2.1613635589349756e-07,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 1136080,
+      "step": 166
+    },
+    {
+      "epoch": 4.295819935691318,
+      "grad_norm": 0.12783974409103394,
+      "learning_rate": 1.9873786636889908e-07,
+      "loss": 0.0002,
+      "num_input_tokens_seen": 1142976,
+      "step": 167
+    },
+    {
+      "epoch": 4.321543408360129,
+      "grad_norm": 0.08086368441581726,
+      "learning_rate": 1.8204036358303173e-07,
+      "loss": 0.0002,
+      "num_input_tokens_seen": 1149712,
+      "step": 168
+    },
+    {
+      "epoch": 4.347266881028939,
+      "grad_norm": 0.018286822363734245,
+      "learning_rate": 1.6604893375699594e-07,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 1156336,
+      "step": 169
+    },
+    {
+      "epoch": 4.372990353697749,
+      "grad_norm": 0.009972570464015007,
+      "learning_rate": 1.507684480352292e-07,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 1163120,
+      "step": 170
+    },
+    {
+      "epoch": 4.39871382636656,
+      "grad_norm": 0.9909194707870483,
+      "learning_rate": 1.362035610017079e-07,
+      "loss": 0.0115,
+      "num_input_tokens_seen": 1170032,
+      "step": 171
+    },
+    {
+      "epoch": 4.42443729903537,
+      "grad_norm": 0.21059785783290863,
+      "learning_rate": 1.223587092621162e-07,
+      "loss": 0.0005,
+      "num_input_tokens_seen": 1176832,
+      "step": 172
+    },
+    {
+      "epoch": 4.45016077170418,
+      "grad_norm": 0.08872511237859726,
+      "learning_rate": 1.0923811009241142e-07,
+      "loss": 0.0003,
+      "num_input_tokens_seen": 1183856,
+      "step": 173
+    },
+    {
+      "epoch": 4.47588424437299,
+      "grad_norm": 0.7515408396720886,
+      "learning_rate": 9.684576015420277e-08,
+      "loss": 0.005,
+      "num_input_tokens_seen": 1190544,
+      "step": 174
+    },
+    {
+      "epoch": 4.501607717041801,
+      "grad_norm": 0.0772569328546524,
+      "learning_rate": 8.518543427732951e-08,
+      "loss": 0.0003,
+      "num_input_tokens_seen": 1197168,
+      "step": 175
+    },
+    {
+      "epoch": 4.527331189710611,
+      "grad_norm": 0.19207893311977386,
+      "learning_rate": 7.426068431000883e-08,
+      "loss": 0.0008,
+      "num_input_tokens_seen": 1203776,
+      "step": 176
+    },
+    {
+      "epoch": 4.553054662379421,
+      "grad_norm": 0.007837573066353798,
+      "learning_rate": 6.407483803691216e-08,
+      "loss": 0.0,
+      "num_input_tokens_seen": 1210816,
+      "step": 177
+    },
+    {
+      "epoch": 4.578778135048232,
+      "grad_norm": 0.009602434933185577,
+      "learning_rate": 5.463099816548578e-08,
+      "loss": 0.0,
+      "num_input_tokens_seen": 1217696,
+      "step": 178
+    },
+    {
+      "epoch": 4.604501607717042,
+      "grad_norm": 1.3553417921066284,
+      "learning_rate": 4.593204138084006e-08,
+      "loss": 0.0042,
+      "num_input_tokens_seen": 1224704,
+      "step": 179
+    },
+    {
+      "epoch": 4.630225080385852,
+      "grad_norm": 0.17981038987636566,
+      "learning_rate": 3.798061746947995e-08,
+      "loss": 0.0004,
+      "num_input_tokens_seen": 1231664,
+      "step": 180
+    },
+    {
+      "epoch": 4.655948553054662,
+      "grad_norm": 0.04047521948814392,
+      "learning_rate": 3.077914851215585e-08,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 1238240,
+      "step": 181
+    },
+    {
+      "epoch": 4.681672025723473,
+      "grad_norm": 0.02045159973204136,
+      "learning_rate": 2.4329828146074096e-08,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 1244832,
+      "step": 182
+    },
+    {
+      "epoch": 4.707395498392283,
+      "grad_norm": 0.10965674370527267,
+      "learning_rate": 1.8634620896695044e-08,
+      "loss": 0.0004,
+      "num_input_tokens_seen": 1251296,
+      "step": 183
+    },
+    {
+      "epoch": 4.733118971061093,
+      "grad_norm": 0.008195169270038605,
+      "learning_rate": 1.3695261579316776e-08,
+      "loss": 0.0,
+      "num_input_tokens_seen": 1257968,
+      "step": 184
+    },
+    {
+      "epoch": 4.758842443729904,
+      "grad_norm": 0.25205764174461365,
+      "learning_rate": 9.513254770636138e-09,
+      "loss": 0.0009,
+      "num_input_tokens_seen": 1264928,
+      "step": 185
+    },
+    {
+      "epoch": 4.784565916398714,
+      "grad_norm": 0.031606338918209076,
+      "learning_rate": 6.089874350439507e-09,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 1271792,
+      "step": 186
+    },
+    {
+      "epoch": 4.810289389067524,
+      "grad_norm": 0.19231323897838593,
+      "learning_rate": 3.4261631135654174e-09,
+      "loss": 0.0004,
+      "num_input_tokens_seen": 1279024,
+      "step": 187
+    },
+    {
+      "epoch": 4.836012861736334,
+      "grad_norm": 0.06766192615032196,
+      "learning_rate": 1.5229324522605949e-09,
+      "loss": 0.0002,
+      "num_input_tokens_seen": 1285792,
+      "step": 188
+    },
+    {
+      "epoch": 4.861736334405145,
+      "grad_norm": 0.005693785380572081,
+      "learning_rate": 3.8076210902182607e-10,
+      "loss": 0.0,
+      "num_input_tokens_seen": 1292576,
+      "step": 189
+    },
+    {
+      "epoch": 4.887459807073955,
+      "grad_norm": 0.02735370770096779,
+      "learning_rate": 0.0,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 1299392,
+      "step": 190
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 190,
+  "num_input_tokens_seen": 1299392,
+  "num_train_epochs": 5,
+  "save_steps": 1000,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 5.151317702790349e+16,
+  "train_batch_size": 2,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-190/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:59c20394a81d6a411e14385c1f4bccd2cbf8486e7c193698844b9070fbad87d6
+size 6584

checkpoint-190/zero_to_fp32.py

@@ -0,0 +1,604 @@
+#!/usr/bin/env python
+
+# Copyright (c) Microsoft Corporation.
+# SPDX-License-Identifier: Apache-2.0
+
+# DeepSpeed Team
+
+# This script extracts fp32 consolidated weights from a zero 1, 2 and 3 DeepSpeed checkpoints. It gets
+# copied into the top level checkpoint dir, so the user can easily do the conversion at any point in
+# the future. Once extracted, the weights don't require DeepSpeed and can be used in any
+# application.
+#
+# example: python zero_to_fp32.py . pytorch_model.bin
+
+import argparse
+import torch
+import glob
+import math
+import os
+import re
+from collections import OrderedDict
+from dataclasses import dataclass
+
+# while this script doesn't use deepspeed to recover data, since the checkpoints are pickled with
+# DeepSpeed data structures it has to be available in the current python environment.
+from deepspeed.utils import logger
+from deepspeed.checkpoint.constants import (DS_VERSION, OPTIMIZER_STATE_DICT, SINGLE_PARTITION_OF_FP32_GROUPS,
+                                            FP32_FLAT_GROUPS, ZERO_STAGE, PARTITION_COUNT, PARAM_SHAPES, BUFFER_NAMES,
+                                            FROZEN_PARAM_SHAPES, FROZEN_PARAM_FRAGMENTS)
+
+
+@dataclass
+class zero_model_state:
+    buffers: dict()
+    param_shapes: dict()
+    shared_params: list
+    ds_version: int
+    frozen_param_shapes: dict()
+    frozen_param_fragments: dict()
+
+
+debug = 0
+
+# load to cpu
+device = torch.device('cpu')
+
+
+def atoi(text):
+    return int(text) if text.isdigit() else text
+
+
+def natural_keys(text):
+    '''
+    alist.sort(key=natural_keys) sorts in human order
+    http://nedbatchelder.com/blog/200712/human_sorting.html
+    (See Toothy's implementation in the comments)
+    '''
+    return [atoi(c) for c in re.split(r'(\d+)', text)]
+
+
+def get_model_state_file(checkpoint_dir, zero_stage):
+    if not os.path.isdir(checkpoint_dir):
+        raise FileNotFoundError(f"Directory '{checkpoint_dir}' doesn't exist")
+
+    # there should be only one file
+    if zero_stage <= 2:
+        file = os.path.join(checkpoint_dir, "mp_rank_00_model_states.pt")
+    elif zero_stage == 3:
+        file = os.path.join(checkpoint_dir, "zero_pp_rank_0_mp_rank_00_model_states.pt")
+
+    if not os.path.exists(file):
+        raise FileNotFoundError(f"can't find model states file at '{file}'")
+
+    return file
+
+
+def get_checkpoint_files(checkpoint_dir, glob_pattern):
+    # XXX: need to test that this simple glob rule works for multi-node setup too
+    ckpt_files = sorted(glob.glob(os.path.join(checkpoint_dir, glob_pattern)), key=natural_keys)
+
+    if len(ckpt_files) == 0:
+        raise FileNotFoundError(f"can't find {glob_pattern} files in directory '{checkpoint_dir}'")
+
+    return ckpt_files
+
+
+def get_optim_files(checkpoint_dir):
+    return get_checkpoint_files(checkpoint_dir, "*_optim_states.pt")
+
+
+def get_model_state_files(checkpoint_dir):
+    return get_checkpoint_files(checkpoint_dir, "*_model_states.pt")
+
+
+def parse_model_states(files):
+    zero_model_states = []
+    for file in files:
+        state_dict = torch.load(file, map_location=device)
+
+        if BUFFER_NAMES not in state_dict:
+            raise ValueError(f"{file} is not a model state checkpoint")
+        buffer_names = state_dict[BUFFER_NAMES]
+        if debug:
+            print("Found buffers:", buffer_names)
+
+        # recover just the buffers while restoring them to fp32 if they were saved in fp16
+        buffers = {k: v.float() for k, v in state_dict["module"].items() if k in buffer_names}
+        param_shapes = state_dict[PARAM_SHAPES]
+
+        # collect parameters that are included in param_shapes
+        param_names = []
+        for s in param_shapes:
+            for name in s.keys():
+                param_names.append(name)
+
+        # update with frozen parameters
+        frozen_param_shapes = state_dict.get(FROZEN_PARAM_SHAPES, None)
+        if frozen_param_shapes is not None:
+            if debug:
+                print(f"Found frozen_param_shapes: {frozen_param_shapes}")
+            param_names += list(frozen_param_shapes.keys())
+
+        # handle shared params
+        shared_params = [[k, v] for k, v in state_dict["shared_params"].items()]
+
+        ds_version = state_dict.get(DS_VERSION, None)
+
+        frozen_param_fragments = state_dict.get(FROZEN_PARAM_FRAGMENTS, None)
+
+        z_model_state = zero_model_state(buffers=buffers,
+                                         param_shapes=param_shapes,
+                                         shared_params=shared_params,
+                                         ds_version=ds_version,
+                                         frozen_param_shapes=frozen_param_shapes,
+                                         frozen_param_fragments=frozen_param_fragments)
+        zero_model_states.append(z_model_state)
+
+    return zero_model_states
+
+
+def parse_optim_states(files, ds_checkpoint_dir):
+
+    total_files = len(files)
+    state_dicts = []
+    for f in files:
+        state_dict = torch.load(f, map_location=device)
+        # immediately discard the potentially huge 2 optimizer states as we only care for fp32 master weights
+        # and also handle the case where it was already removed by another helper script
+        state_dict["optimizer_state_dict"].pop("optimizer_state_dict", None)
+        state_dicts.append(state_dict)
+
+    if not ZERO_STAGE in state_dicts[0][OPTIMIZER_STATE_DICT]:
+        raise ValueError(f"{files[0]} is not a zero checkpoint")
+    zero_stage = state_dicts[0][OPTIMIZER_STATE_DICT][ZERO_STAGE]
+    world_size = state_dicts[0][OPTIMIZER_STATE_DICT][PARTITION_COUNT]
+
+    # For ZeRO-2 each param group can have different partition_count as data parallelism for expert
+    # parameters can be different from data parallelism for non-expert parameters. So we can just
+    # use the max of the partition_count to get the dp world_size.
+
+    if type(world_size) is list:
+        world_size = max(world_size)
+
+    if world_size != total_files:
+        raise ValueError(
+            f"Expected {world_size} of '*_optim_states.pt' under '{ds_checkpoint_dir}' but found {total_files} files. "
+            "Possibly due to an overwrite of an old checkpoint, or a checkpoint didn't get saved by one or more processes."
+        )
+
+    # the groups are named differently in each stage
+    if zero_stage <= 2:
+        fp32_groups_key = SINGLE_PARTITION_OF_FP32_GROUPS
+    elif zero_stage == 3:
+        fp32_groups_key = FP32_FLAT_GROUPS
+    else:
+        raise ValueError(f"unknown zero stage {zero_stage}")
+
+    if zero_stage <= 2:
+        fp32_flat_groups = [state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key] for i in range(len(state_dicts))]
+    elif zero_stage == 3:
+        # if there is more than one param group, there will be multiple flattened tensors - one
+        # flattened tensor per group - for simplicity merge them into a single tensor
+        #
+        # XXX: could make the script more memory efficient for when there are multiple groups - it
+        # will require matching the sub-lists of param_shapes for each param group flattened tensor
+
+        fp32_flat_groups = [
+            torch.cat(state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key], 0) for i in range(len(state_dicts))
+        ]
+
+    return zero_stage, world_size, fp32_flat_groups
+
+
+def _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters):
+    """
+    Returns fp32 state_dict reconstructed from ds checkpoint
+
+    Args:
+        - ``ds_checkpoint_dir``: path to the deepspeed checkpoint folder (where the optimizer files are)
+
+    """
+    print(f"Processing zero checkpoint '{ds_checkpoint_dir}'")
+
+    optim_files = get_optim_files(ds_checkpoint_dir)
+    zero_stage, world_size, fp32_flat_groups = parse_optim_states(optim_files, ds_checkpoint_dir)
+    print(f"Detected checkpoint of type zero stage {zero_stage}, world_size: {world_size}")
+
+    model_files = get_model_state_files(ds_checkpoint_dir)
+
+    zero_model_states = parse_model_states(model_files)
+    print(f'Parsing checkpoint created by deepspeed=={zero_model_states[0].ds_version}')
+
+    if zero_stage <= 2:
+        return _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states,
+                                                          exclude_frozen_parameters)
+    elif zero_stage == 3:
+        return _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states,
+                                                          exclude_frozen_parameters)
+
+
+def _zero2_merge_frozen_params(state_dict, zero_model_states):
+    if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0:
+        return
+
+    frozen_param_shapes = zero_model_states[0].frozen_param_shapes
+    frozen_param_fragments = zero_model_states[0].frozen_param_fragments
+
+    if debug:
+        num_elem = sum(s.numel() for s in frozen_param_shapes.values())
+        print(f'rank 0: {FROZEN_PARAM_SHAPES}.numel = {num_elem}')
+
+        wanted_params = len(frozen_param_shapes)
+        wanted_numel = sum(s.numel() for s in frozen_param_shapes.values())
+        avail_numel = sum([p.numel() for p in frozen_param_fragments.values()])
+        print(f'Frozen params: Have {avail_numel} numels to process.')
+        print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params')
+
+    total_params = 0
+    total_numel = 0
+    for name, shape in frozen_param_shapes.items():
+        total_params += 1
+        unpartitioned_numel = shape.numel()
+        total_numel += unpartitioned_numel
+
+        state_dict[name] = frozen_param_fragments[name]
+
+        if debug:
+            print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ")
+
+    print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements")
+
+
+def _has_callable(obj, fn):
+    attr = getattr(obj, fn, None)
+    return callable(attr)
+
+
+def _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states):
+    param_shapes = zero_model_states[0].param_shapes
+
+    # Reconstruction protocol:
+    #
+    # XXX: document this
+
+    if debug:
+        for i in range(world_size):
+            for j in range(len(fp32_flat_groups[0])):
+                print(f"{FP32_FLAT_GROUPS}[{i}][{j}].shape={fp32_flat_groups[i][j].shape}")
+
+    # XXX: memory usage doubles here (zero2)
+    num_param_groups = len(fp32_flat_groups[0])
+    merged_single_partition_of_fp32_groups = []
+    for i in range(num_param_groups):
+        merged_partitions = [sd[i] for sd in fp32_flat_groups]
+        full_single_fp32_vector = torch.cat(merged_partitions, 0)
+        merged_single_partition_of_fp32_groups.append(full_single_fp32_vector)
+    avail_numel = sum(
+        [full_single_fp32_vector.numel() for full_single_fp32_vector in merged_single_partition_of_fp32_groups])
+
+    if debug:
+        wanted_params = sum([len(shapes) for shapes in param_shapes])
+        wanted_numel = sum([sum(shape.numel() for shape in shapes.values()) for shapes in param_shapes])
+        # not asserting if there is a mismatch due to possible padding
+        print(f"Have {avail_numel} numels to process.")
+        print(f"Need {wanted_numel} numels in {wanted_params} params.")
+
+    # params
+    # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support
+    # out-of-core computing solution
+    total_numel = 0
+    total_params = 0
+    for shapes, full_single_fp32_vector in zip(param_shapes, merged_single_partition_of_fp32_groups):
+        offset = 0
+        avail_numel = full_single_fp32_vector.numel()
+        for name, shape in shapes.items():
+
+            unpartitioned_numel = shape.numel() if _has_callable(shape, 'numel') else math.prod(shape)
+            total_numel += unpartitioned_numel
+            total_params += 1
+
+            if debug:
+                print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ")
+            state_dict[name] = full_single_fp32_vector.narrow(0, offset, unpartitioned_numel).view(shape)
+            offset += unpartitioned_numel
+
+        # Z2 started to align to 2*world_size to improve nccl performance. Therefore both offset and
+        # avail_numel can differ by anywhere between 0..2*world_size. Due to two unrelated complex
+        # paddings performed in the code it's almost impossible to predict the exact numbers w/o the
+        # live optimizer object, so we are checking that the numbers are within the right range
+        align_to = 2 * world_size
+
+        def zero2_align(x):
+            return align_to * math.ceil(x / align_to)
+
+        if debug:
+            print(f"original offset={offset}, avail_numel={avail_numel}")
+
+        offset = zero2_align(offset)
+        avail_numel = zero2_align(avail_numel)
+
+        if debug:
+            print(f"aligned  offset={offset}, avail_numel={avail_numel}")
+
+        # Sanity check
+        if offset != avail_numel:
+            raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong")
+
+    print(f"Reconstructed fp32 state dict with {total_params} params {total_numel} elements")
+
+
+def _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states,
+                                               exclude_frozen_parameters):
+    state_dict = OrderedDict()
+
+    # buffers
+    buffers = zero_model_states[0].buffers
+    state_dict.update(buffers)
+    if debug:
+        print(f"added {len(buffers)} buffers")
+
+    if not exclude_frozen_parameters:
+        _zero2_merge_frozen_params(state_dict, zero_model_states)
+
+    _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states)
+
+    # recover shared parameters
+    for pair in zero_model_states[0].shared_params:
+        if pair[1] in state_dict:
+            state_dict[pair[0]] = state_dict[pair[1]]
+
+    return state_dict
+
+
+def zero3_partitioned_param_info(unpartitioned_numel, world_size):
+    remainder = unpartitioned_numel % world_size
+    padding_numel = (world_size - remainder) if remainder else 0
+    partitioned_numel = math.ceil(unpartitioned_numel / world_size)
+    return partitioned_numel, padding_numel
+
+
+def _zero3_merge_frozen_params(state_dict, world_size, zero_model_states):
+    if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0:
+        return
+
+    if debug:
+        for i in range(world_size):
+            num_elem = sum(s.numel() for s in zero_model_states[i].frozen_param_fragments.values())
+            print(f'rank {i}: {FROZEN_PARAM_SHAPES}.numel = {num_elem}')
+
+        frozen_param_shapes = zero_model_states[0].frozen_param_shapes
+        wanted_params = len(frozen_param_shapes)
+        wanted_numel = sum(s.numel() for s in frozen_param_shapes.values())
+        avail_numel = sum([p.numel() for p in zero_model_states[0].frozen_param_fragments.values()]) * world_size
+        print(f'Frozen params: Have {avail_numel} numels to process.')
+        print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params')
+
+    total_params = 0
+    total_numel = 0
+    for name, shape in zero_model_states[0].frozen_param_shapes.items():
+        total_params += 1
+        unpartitioned_numel = shape.numel()
+        total_numel += unpartitioned_numel
+
+        param_frags = tuple(model_state.frozen_param_fragments[name] for model_state in zero_model_states)
+        state_dict[name] = torch.cat(param_frags, 0).narrow(0, 0, unpartitioned_numel).view(shape)
+
+        partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size)
+
+        if debug:
+            print(
+                f"Frozen params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}"
+            )
+
+    print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements")
+
+
+def _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states):
+    param_shapes = zero_model_states[0].param_shapes
+    avail_numel = fp32_flat_groups[0].numel() * world_size
+    # Reconstruction protocol: For zero3 we need to zip the partitions together at boundary of each
+    # param, re-consolidating each param, while dealing with padding if any
+
+    # merge list of dicts, preserving order
+    param_shapes = {k: v for d in param_shapes for k, v in d.items()}
+
+    if debug:
+        for i in range(world_size):
+            print(f"{FP32_FLAT_GROUPS}[{i}].shape={fp32_flat_groups[i].shape}")
+
+        wanted_params = len(param_shapes)
+        wanted_numel = sum(shape.numel() for shape in param_shapes.values())
+        # not asserting if there is a mismatch due to possible padding
+        avail_numel = fp32_flat_groups[0].numel() * world_size
+        print(f"Trainable params: Have {avail_numel} numels to process.")
+        print(f"Trainable params: Need {wanted_numel} numels in {wanted_params} params.")
+
+    # params
+    # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support
+    # out-of-core computing solution
+    offset = 0
+    total_numel = 0
+    total_params = 0
+    for name, shape in param_shapes.items():
+
+        unpartitioned_numel = shape.numel()
+        total_numel += unpartitioned_numel
+        total_params += 1
+
+        partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size)
+
+        if debug:
+            print(
+                f"Trainable params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}"
+            )
+
+        # XXX: memory usage doubles here
+        state_dict[name] = torch.cat(
+            tuple(fp32_flat_groups[i].narrow(0, offset, partitioned_numel) for i in range(world_size)),
+            0).narrow(0, 0, unpartitioned_numel).view(shape)
+        offset += partitioned_numel
+
+    offset *= world_size
+
+    # Sanity check
+    if offset != avail_numel:
+        raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong")
+
+    print(f"Reconstructed Trainable fp32 state dict with {total_params} params {total_numel} elements")
+
+
+def _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states,
+                                               exclude_frozen_parameters):
+    state_dict = OrderedDict()
+
+    # buffers
+    buffers = zero_model_states[0].buffers
+    state_dict.update(buffers)
+    if debug:
+        print(f"added {len(buffers)} buffers")
+
+    if not exclude_frozen_parameters:
+        _zero3_merge_frozen_params(state_dict, world_size, zero_model_states)
+
+    _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states)
+
+    # recover shared parameters
+    for pair in zero_model_states[0].shared_params:
+        if pair[1] in state_dict:
+            state_dict[pair[0]] = state_dict[pair[1]]
+
+    return state_dict
+
+
+def get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag=None, exclude_frozen_parameters=False):
+    """
+    Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated state_dict that can be loaded with
+    ``load_state_dict()`` and used for training without DeepSpeed or shared with others, for example
+    via a model hub.
+
+    Args:
+        - ``checkpoint_dir``: path to the desired checkpoint folder
+        - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in 'latest' file. e.g., ``global_step14``
+        - ``exclude_frozen_parameters``: exclude frozen parameters
+
+    Returns:
+        - pytorch ``state_dict``
+
+    Note: this approach may not work if your application doesn't have sufficient free CPU memory and
+    you may need to use the offline approach using the ``zero_to_fp32.py`` script that is saved with
+    the checkpoint.
+
+    A typical usage might be ::
+
+        from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint
+        # do the training and checkpoint saving
+        state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir) # already on cpu
+        model = model.cpu() # move to cpu
+        model.load_state_dict(state_dict)
+        # submit to model hub or save the model to share with others
+
+    In this example the ``model`` will no longer be usable in the deepspeed context of the same
+    application. i.e. you will need to re-initialize the deepspeed engine, since
+    ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it.
+
+    If you want it all done for you, use ``load_state_dict_from_zero_checkpoint`` instead.
+
+    """
+    if tag is None:
+        latest_path = os.path.join(checkpoint_dir, 'latest')
+        if os.path.isfile(latest_path):
+            with open(latest_path, 'r') as fd:
+                tag = fd.read().strip()
+        else:
+            raise ValueError(f"Unable to find 'latest' file at {latest_path}")
+
+    ds_checkpoint_dir = os.path.join(checkpoint_dir, tag)
+
+    if not os.path.isdir(ds_checkpoint_dir):
+        raise FileNotFoundError(f"Directory '{ds_checkpoint_dir}' doesn't exist")
+
+    return _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters)
+
+
+def convert_zero_checkpoint_to_fp32_state_dict(checkpoint_dir, output_file, tag=None, exclude_frozen_parameters=False):
+    """
+    Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` file that can be
+    loaded with ``torch.load(file)`` + ``load_state_dict()`` and used for training without DeepSpeed.
+
+    Args:
+        - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``)
+        - ``output_file``: path to the pytorch fp32 state_dict output file (e.g. path/pytorch_model.bin)
+        - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14``
+        - ``exclude_frozen_parameters``: exclude frozen parameters
+    """
+
+    state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag, exclude_frozen_parameters)
+    print(f"Saving fp32 state dict to {output_file}")
+    torch.save(state_dict, output_file)
+
+
+def load_state_dict_from_zero_checkpoint(model, checkpoint_dir, tag=None):
+    """
+    1. Put the provided model to cpu
+    2. Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict``
+    3. Load it into the provided model
+
+    Args:
+        - ``model``: the model object to update
+        - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``)
+        - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14``
+
+    Returns:
+        - ``model`: modified model
+
+    Make sure you have plenty of CPU memory available before you call this function. If you don't
+    have enough use the ``zero_to_fp32.py`` utility to do the conversion. You will find it
+    conveniently placed for you in the checkpoint folder.
+
+    A typical usage might be ::
+
+        from deepspeed.utils.zero_to_fp32 import load_state_dict_from_zero_checkpoint
+        model = load_state_dict_from_zero_checkpoint(trainer.model, checkpoint_dir)
+        # submit to model hub or save the model to share with others
+
+    Note, that once this was run, the ``model`` will no longer be usable in the deepspeed context
+    of the same application. i.e. you will need to re-initialize the deepspeed engine, since
+    ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it.
+
+    """
+    logger.info(f"Extracting fp32 weights")
+    state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag)
+
+    logger.info(f"Overwriting model with fp32 weights")
+    model = model.cpu()
+    model.load_state_dict(state_dict, strict=False)
+
+    return model
+
+
+if __name__ == "__main__":
+
+    parser = argparse.ArgumentParser()
+    parser.add_argument("checkpoint_dir",
+                        type=str,
+                        help="path to the desired checkpoint folder, e.g., path/checkpoint-12")
+    parser.add_argument(
+        "output_file",
+        type=str,
+        help="path to the pytorch fp32 state_dict output file (e.g. path/checkpoint-12/pytorch_model.bin)")
+    parser.add_argument("-t",
+                        "--tag",
+                        type=str,
+                        default=None,
+                        help="checkpoint tag used as a unique identifier for checkpoint. e.g., global_step1")
+    parser.add_argument("--exclude_frozen_parameters", action='store_true', help="exclude frozen parameters")
+    parser.add_argument("-d", "--debug", action='store_true', help="enable debug")
+    args = parser.parse_args()
+
+    debug = args.debug
+
+    convert_zero_checkpoint_to_fp32_state_dict(args.checkpoint_dir,
+                                               args.output_file,
+                                               tag=args.tag,
+                                               exclude_frozen_parameters=args.exclude_frozen_parameters)

config.json

@@ -0,0 +1,29 @@
+{
+  "_name_or_path": "meta-llama/Llama-2-7b-chat-hf",
+  "architectures": [
+    "LlamaForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "bos_token_id": 1,
+  "eos_token_id": 2,
+  "hidden_act": "silu",
+  "hidden_size": 4096,
+  "initializer_range": 0.02,
+  "intermediate_size": 11008,
+  "max_position_embeddings": 4096,
+  "mlp_bias": false,
+  "model_type": "llama",
+  "num_attention_heads": 32,
+  "num_hidden_layers": 32,
+  "num_key_value_heads": 32,
+  "pretraining_tp": 1,
+  "rms_norm_eps": 1e-05,
+  "rope_scaling": null,
+  "rope_theta": 10000.0,
+  "tie_word_embeddings": false,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.42.3",
+  "use_cache": false,
+  "vocab_size": 32000
+}

generation_config.json

@@ -0,0 +1,10 @@
+{
+  "bos_token_id": 1,
+  "do_sample": true,
+  "eos_token_id": 2,
+  "max_length": 4096,
+  "pad_token_id": 0,
+  "temperature": 0.6,
+  "top_p": 0.9,
+  "transformers_version": "4.42.3"
+}

llamaboard_config.yaml

@@ -0,0 +1,65 @@
+top.booster: auto
+top.checkpoint_path: null
+top.finetuning_type: full
+top.model_name: LLaMA2-7B-Chat
+top.quantization_bit: none
+top.quantization_method: bitsandbytes
+top.rope_scaling: none
+top.template: llama2
+top.visual_inputs: false
+train.additional_target: ''
+train.badam_mode: layer
+train.badam_switch_interval: 50
+train.badam_switch_mode: ascending
+train.badam_update_ratio: 0.05
+train.batch_size: 2
+train.compute_type: bf16
+train.create_new_adapter: false
+train.cutoff_len: 1024
+train.dataset:
+- truth_train_0716
+train.dataset_dir: data
+train.ds_offload: false
+train.ds_stage: '2'
+train.freeze_extra_modules: ''
+train.freeze_trainable_layers: 2
+train.freeze_trainable_modules: all
+train.galore_rank: 16
+train.galore_scale: 0.25
+train.galore_target: all
+train.galore_update_interval: 200
+train.gradient_accumulation_steps: 8
+train.learning_rate: 5e-6
+train.logging_steps: 1
+train.lora_alpha: 16
+train.lora_dropout: 0
+train.lora_rank: 8
+train.lora_target: ''
+train.loraplus_lr_ratio: 0
+train.lr_scheduler_type: cosine
+train.max_grad_norm: '1.0'
+train.max_samples: '100000'
+train.neat_packing: false
+train.neftune_alpha: 0
+train.num_train_epochs: '5.0'
+train.optim: adamw_torch
+train.packing: false
+train.ppo_score_norm: false
+train.ppo_whiten_rewards: false
+train.pref_beta: 0.1
+train.pref_ftx: 0
+train.pref_loss: sigmoid
+train.report_to: false
+train.resize_vocab: false
+train.reward_model: null
+train.save_steps: 1000
+train.shift_attn: false
+train.training_stage: Supervised Fine-Tuning
+train.use_badam: false
+train.use_dora: false
+train.use_galore: false
+train.use_llama_pro: false
+train.use_pissa: false
+train.use_rslora: false
+train.val_size: 0
+train.warmup_steps: 10

model-00001-of-00003.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4ad2a4f8c844b27f40af4f7ebd576dc04d590cf97967cd141f9ff0e37e9b7f06
+size 4938985352

model-00002-of-00003.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0d6911d18e1230b44c2dcec0b8b93af000e046d3a3672425c51c3905490a01f9
+size 4947390880

model-00003-of-00003.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:38aea6ed7c69811e9b91746ce9ba805addb67a766dad6944d6b53d929f4af38e
+size 3590488816

model.safetensors.index.json

@@ -0,0 +1,298 @@
+{
+  "metadata": {
+    "total_size": 13476831232
+  },
+  "weight_map": {
+    "lm_head.weight": "model-00003-of-00003.safetensors",
+    "model.embed_tokens.weight": "model-00001-of-00003.safetensors",
+    "model.layers.0.input_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.0.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.0.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.0.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.1.input_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.1.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.1.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.1.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.1.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.1.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.1.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.10.input_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.10.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.10.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.10.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.10.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.10.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.10.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.10.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.10.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.11.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.11.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.11.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.11.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.11.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.11.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.11.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.11.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.11.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.12.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.12.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.12.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.12.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.12.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.12.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.12.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.12.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.12.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.13.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.13.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.13.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.13.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.13.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.13.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.13.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.13.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.13.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.14.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.14.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.14.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.14.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.14.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.14.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.14.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.14.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.14.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.15.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.15.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.15.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.15.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.15.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.15.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.15.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.15.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.15.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.16.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.16.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.16.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.16.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.16.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.16.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.16.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.16.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.16.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.17.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.17.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.17.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.17.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.17.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.17.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.17.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.17.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.17.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.18.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.18.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.18.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.18.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.18.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.18.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.18.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.18.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.18.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.19.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.19.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.19.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.19.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.19.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.19.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.19.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.19.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.19.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.2.input_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.2.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.2.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.2.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.2.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.2.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.2.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.2.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.2.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.20.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.20.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.20.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.20.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.20.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.20.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.20.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.20.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.20.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.21.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.21.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.21.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.21.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.21.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.21.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.21.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.21.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.21.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.22.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.22.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.22.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.22.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.22.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.22.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.22.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.22.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.22.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.23.input_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.23.mlp.down_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.23.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.23.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.23.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.23.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.23.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.23.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.23.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.24.input_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.24.mlp.down_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.24.mlp.gate_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.24.mlp.up_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.24.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.24.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.24.self_attn.o_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.24.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.24.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.25.input_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.25.mlp.down_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.25.mlp.gate_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.25.mlp.up_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.25.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.25.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.25.self_attn.o_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.25.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.25.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.26.input_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.26.mlp.down_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.26.mlp.gate_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.26.mlp.up_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.26.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.26.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.26.self_attn.o_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.26.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.26.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.27.input_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.27.mlp.down_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.27.mlp.gate_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.27.mlp.up_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.27.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.27.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.27.self_attn.o_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.27.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.27.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.28.input_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.28.mlp.down_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.28.mlp.gate_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.28.mlp.up_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.28.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.28.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.28.self_attn.o_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.28.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.28.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.29.input_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.29.mlp.down_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.29.mlp.gate_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.29.mlp.up_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.29.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.29.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.29.self_attn.o_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.29.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.29.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.3.input_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.3.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.3.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.3.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.3.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.3.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.3.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.3.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.3.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.30.input_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.30.mlp.down_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.30.mlp.gate_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.30.mlp.up_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.30.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.30.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.30.self_attn.o_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.30.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.30.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.31.input_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.31.mlp.down_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.31.mlp.gate_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.31.mlp.up_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.31.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.31.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.31.self_attn.o_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.31.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.31.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.4.input_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.4.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.4.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.4.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.4.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.4.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.4.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.4.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.4.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.5.input_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.5.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.5.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.5.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.5.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.5.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.5.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.5.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.5.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.6.input_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.6.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.6.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.6.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.6.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.6.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.6.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.6.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.6.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.7.input_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.7.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.7.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.7.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.7.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.7.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.7.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.7.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.7.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.8.input_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.8.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.8.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.8.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.8.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.8.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.8.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.8.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.8.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.9.input_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.9.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.9.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.9.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.9.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.9.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.9.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.9.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.9.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "model.norm.weight": "model-00003-of-00003.safetensors"
+  }
+}

running_log.txt

@@ -0,0 +1,631 @@
+07/16/2024 09:07:34 - INFO - llamafactory.data.template - Add pad token: </s>
+
+07/16/2024 09:07:34 - INFO - llamafactory.data.template - Add pad token: </s>
+
+07/16/2024 09:07:34 - INFO - llamafactory.hparams.parser - Process rank: 3, device: cuda:3, n_gpu: 1, distributed training: True, compute dtype: torch.bfloat16
+
+[INFO|parser.py:325] 2024-07-16 09:07:34,077 >> Process rank: 0, device: cuda:0, n_gpu: 1, distributed training: True, compute dtype: torch.bfloat16
+
+07/16/2024 09:07:34 - INFO - llamafactory.hparams.parser - Process rank: 7, device: cuda:7, n_gpu: 1, distributed training: True, compute dtype: torch.bfloat16
+
+07/16/2024 09:07:34 - INFO - llamafactory.hparams.parser - Process rank: 5, device: cuda:5, n_gpu: 1, distributed training: True, compute dtype: torch.bfloat16
+
+07/16/2024 09:07:34 - INFO - llamafactory.hparams.parser - Process rank: 2, device: cuda:2, n_gpu: 1, distributed training: True, compute dtype: torch.bfloat16
+
+[INFO|tokenization_utils_base.py:2161] 2024-07-16 09:07:34,347 >> loading file tokenizer.model from cache at /root/.cache/huggingface/hub/models--meta-llama--Llama-2-7b-chat-hf/snapshots/f5db02db724555f92da89c216ac04704f23d4590/tokenizer.model
+
+[INFO|tokenization_utils_base.py:2161] 2024-07-16 09:07:34,347 >> loading file tokenizer.json from cache at /root/.cache/huggingface/hub/models--meta-llama--Llama-2-7b-chat-hf/snapshots/f5db02db724555f92da89c216ac04704f23d4590/tokenizer.json
+
+[INFO|tokenization_utils_base.py:2161] 2024-07-16 09:07:34,348 >> loading file added_tokens.json from cache at None
+
+[INFO|tokenization_utils_base.py:2161] 2024-07-16 09:07:34,348 >> loading file special_tokens_map.json from cache at /root/.cache/huggingface/hub/models--meta-llama--Llama-2-7b-chat-hf/snapshots/f5db02db724555f92da89c216ac04704f23d4590/special_tokens_map.json
+
+[INFO|tokenization_utils_base.py:2161] 2024-07-16 09:07:34,348 >> loading file tokenizer_config.json from cache at /root/.cache/huggingface/hub/models--meta-llama--Llama-2-7b-chat-hf/snapshots/f5db02db724555f92da89c216ac04704f23d4590/tokenizer_config.json
+
+07/16/2024 09:07:34 - INFO - llamafactory.data.template - Add pad token: </s>
+
+07/16/2024 09:07:34 - INFO - llamafactory.data.template - Add pad token: </s>
+
+07/16/2024 09:07:34 - INFO - llamafactory.data.template - Add pad token: </s>
+
+07/16/2024 09:07:34 - INFO - llamafactory.data.template - Add pad token: </s>
+
+[INFO|template.py:372] 2024-07-16 09:07:34,452 >> Add pad token: </s>
+
+[INFO|loader.py:50] 2024-07-16 09:07:34,453 >> Loading dataset 0716_truthfulqa_benchmark_train.json...
+
+07/16/2024 09:07:34 - INFO - llamafactory.hparams.parser - Process rank: 6, device: cuda:6, n_gpu: 1, distributed training: True, compute dtype: torch.bfloat16
+
+07/16/2024 09:07:34 - INFO - llamafactory.data.template - Add pad token: </s>
+
+07/16/2024 09:07:36 - INFO - llamafactory.data.loader - Loading dataset 0716_truthfulqa_benchmark_train.json...
+
+07/16/2024 09:07:36 - INFO - llamafactory.data.loader - Loading dataset 0716_truthfulqa_benchmark_train.json...
+
+07/16/2024 09:07:36 - INFO - llamafactory.data.loader - Loading dataset 0716_truthfulqa_benchmark_train.json...
+
+07/16/2024 09:07:36 - INFO - llamafactory.data.loader - Loading dataset 0716_truthfulqa_benchmark_train.json...
+
+07/16/2024 09:07:36 - INFO - llamafactory.data.loader - Loading dataset 0716_truthfulqa_benchmark_train.json...
+
+07/16/2024 09:07:36 - INFO - llamafactory.data.loader - Loading dataset 0716_truthfulqa_benchmark_train.json...
+
+07/16/2024 09:07:36 - INFO - llamafactory.data.loader - Loading dataset 0716_truthfulqa_benchmark_train.json...
+
+[INFO|configuration_utils.py:733] 2024-07-16 09:07:37,470 >> loading configuration file config.json from cache at /root/.cache/huggingface/hub/models--meta-llama--Llama-2-7b-chat-hf/snapshots/f5db02db724555f92da89c216ac04704f23d4590/config.json
+
+[INFO|configuration_utils.py:800] 2024-07-16 09:07:37,473 >> Model config LlamaConfig {
+  "_name_or_path": "meta-llama/Llama-2-7b-chat-hf",
+  "architectures": [
+    "LlamaForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "bos_token_id": 1,
+  "eos_token_id": 2,
+  "hidden_act": "silu",
+  "hidden_size": 4096,
+  "initializer_range": 0.02,
+  "intermediate_size": 11008,
+  "max_position_embeddings": 4096,
+  "mlp_bias": false,
+  "model_type": "llama",
+  "num_attention_heads": 32,
+  "num_hidden_layers": 32,
+  "num_key_value_heads": 32,
+  "pretraining_tp": 1,
+  "rms_norm_eps": 1e-05,
+  "rope_scaling": null,
+  "rope_theta": 10000.0,
+  "tie_word_embeddings": false,
+  "torch_dtype": "float16",
+  "transformers_version": "4.42.3",
+  "use_cache": true,
+  "vocab_size": 32000
+}
+
+
+[INFO|modeling_utils.py:3556] 2024-07-16 09:07:37,523 >> loading weights file model.safetensors from cache at /root/.cache/huggingface/hub/models--meta-llama--Llama-2-7b-chat-hf/snapshots/f5db02db724555f92da89c216ac04704f23d4590/model.safetensors.index.json
+
+[INFO|modeling_utils.py:1531] 2024-07-16 09:07:37,524 >> Instantiating LlamaForCausalLM model under default dtype torch.bfloat16.
+
+[INFO|configuration_utils.py:1000] 2024-07-16 09:07:37,526 >> Generate config GenerationConfig {
+  "bos_token_id": 1,
+  "eos_token_id": 2
+}
+
+
+[INFO|modeling_utils.py:4364] 2024-07-16 09:07:54,870 >> All model checkpoint weights were used when initializing LlamaForCausalLM.
+
+
+[INFO|modeling_utils.py:4372] 2024-07-16 09:07:54,870 >> All the weights of LlamaForCausalLM were initialized from the model checkpoint at meta-llama/Llama-2-7b-chat-hf.
+If your task is similar to the task the model of the checkpoint was trained on, you can already use LlamaForCausalLM for predictions without further training.
+
+07/16/2024 09:07:55 - INFO - llamafactory.model.model_utils.checkpointing - Gradient checkpointing enabled.
+
+07/16/2024 09:07:55 - INFO - llamafactory.model.model_utils.attention - Using torch SDPA for faster training and inference.
+
+07/16/2024 09:07:55 - INFO - llamafactory.model.adapter - Upcasting trainable params to float32.
+
+07/16/2024 09:07:55 - INFO - llamafactory.model.adapter - Fine-tuning method: Full
+
+07/16/2024 09:07:55 - INFO - llamafactory.model.model_utils.checkpointing - Gradient checkpointing enabled.
+
+07/16/2024 09:07:55 - INFO - llamafactory.model.model_utils.attention - Using torch SDPA for faster training and inference.
+
+07/16/2024 09:07:55 - INFO - llamafactory.model.adapter - Upcasting trainable params to float32.
+
+07/16/2024 09:07:55 - INFO - llamafactory.model.adapter - Fine-tuning method: Full
+
+[INFO|configuration_utils.py:955] 2024-07-16 09:07:55,055 >> loading configuration file generation_config.json from cache at /root/.cache/huggingface/hub/models--meta-llama--Llama-2-7b-chat-hf/snapshots/f5db02db724555f92da89c216ac04704f23d4590/generation_config.json
+
+[INFO|configuration_utils.py:1000] 2024-07-16 09:07:55,055 >> Generate config GenerationConfig {
+  "bos_token_id": 1,
+  "do_sample": true,
+  "eos_token_id": 2,
+  "max_length": 4096,
+  "pad_token_id": 0,
+  "temperature": 0.6,
+  "top_p": 0.9
+}
+
+
+07/16/2024 09:07:55 - INFO - llamafactory.model.model_utils.checkpointing - Gradient checkpointing enabled.
+
+07/16/2024 09:07:55 - INFO - llamafactory.model.model_utils.attention - Using torch SDPA for faster training and inference.
+
+07/16/2024 09:07:55 - INFO - llamafactory.model.adapter - Upcasting trainable params to float32.
+
+07/16/2024 09:07:55 - INFO - llamafactory.model.adapter - Fine-tuning method: Full
+
+[INFO|checkpointing.py:103] 2024-07-16 09:07:55,062 >> Gradient checkpointing enabled.
+
+[INFO|attention.py:80] 2024-07-16 09:07:55,062 >> Using torch SDPA for faster training and inference.
+
+[INFO|adapter.py:302] 2024-07-16 09:07:55,062 >> Upcasting trainable params to float32.
+
+[INFO|adapter.py:48] 2024-07-16 09:07:55,062 >> Fine-tuning method: Full
+
+07/16/2024 09:07:55 - INFO - llamafactory.model.model_utils.checkpointing - Gradient checkpointing enabled.
+
+07/16/2024 09:07:55 - INFO - llamafactory.model.model_utils.checkpointing - Gradient checkpointing enabled.
+
+07/16/2024 09:07:55 - INFO - llamafactory.model.model_utils.attention - Using torch SDPA for faster training and inference.
+
+07/16/2024 09:07:55 - INFO - llamafactory.model.model_utils.attention - Using torch SDPA for faster training and inference.
+
+07/16/2024 09:07:55 - INFO - llamafactory.model.adapter - Upcasting trainable params to float32.
+
+07/16/2024 09:07:55 - INFO - llamafactory.model.adapter - Upcasting trainable params to float32.
+
+07/16/2024 09:07:55 - INFO - llamafactory.model.adapter - Fine-tuning method: Full
+
+07/16/2024 09:07:55 - INFO - llamafactory.model.adapter - Fine-tuning method: Full
+
+07/16/2024 09:07:55 - INFO - llamafactory.model.loader - trainable params: 6,738,415,616 || all params: 6,738,415,616 || trainable%: 100.0000
+
+07/16/2024 09:07:55 - INFO - llamafactory.model.loader - trainable params: 6,738,415,616 || all params: 6,738,415,616 || trainable%: 100.0000
+
+07/16/2024 09:07:55 - INFO - llamafactory.model.loader - trainable params: 6,738,415,616 || all params: 6,738,415,616 || trainable%: 100.0000
+
+[INFO|loader.py:196] 2024-07-16 09:07:55,174 >> trainable params: 6,738,415,616 || all params: 6,738,415,616 || trainable%: 100.0000
+
+07/16/2024 09:07:55 - INFO - llamafactory.model.loader - trainable params: 6,738,415,616 || all params: 6,738,415,616 || trainable%: 100.0000
+
+07/16/2024 09:07:55 - INFO - llamafactory.model.loader - trainable params: 6,738,415,616 || all params: 6,738,415,616 || trainable%: 100.0000
+
+07/16/2024 09:07:55 - INFO - llamafactory.model.model_utils.checkpointing - Gradient checkpointing enabled.
+
+07/16/2024 09:07:55 - INFO - llamafactory.model.model_utils.attention - Using torch SDPA for faster training and inference.
+
+07/16/2024 09:07:55 - INFO - llamafactory.model.adapter - Upcasting trainable params to float32.
+
+07/16/2024 09:07:55 - INFO - llamafactory.model.adapter - Fine-tuning method: Full
+
+07/16/2024 09:07:55 - INFO - llamafactory.model.loader - trainable params: 6,738,415,616 || all params: 6,738,415,616 || trainable%: 100.0000
+
+[INFO|trainer.py:642] 2024-07-16 09:07:55,179 >> Using auto half precision backend
+
+07/16/2024 09:07:55 - INFO - llamafactory.model.model_utils.checkpointing - Gradient checkpointing enabled.
+
+07/16/2024 09:07:55 - INFO - llamafactory.model.model_utils.attention - Using torch SDPA for faster training and inference.
+
+07/16/2024 09:07:55 - INFO - llamafactory.model.adapter - Upcasting trainable params to float32.
+
+07/16/2024 09:07:55 - INFO - llamafactory.model.adapter - Fine-tuning method: Full
+
+07/16/2024 09:07:55 - INFO - llamafactory.model.loader - trainable params: 6,738,415,616 || all params: 6,738,415,616 || trainable%: 100.0000
+
+[INFO|trainer.py:2128] 2024-07-16 09:08:14,231 >> ***** Running training *****
+
+[INFO|trainer.py:2129] 2024-07-16 09:08:14,231 >>   Num examples = 4,968
+
+[INFO|trainer.py:2130] 2024-07-16 09:08:14,231 >>   Num Epochs = 5
+
+[INFO|trainer.py:2131] 2024-07-16 09:08:14,231 >>   Instantaneous batch size per device = 2
+
+[INFO|trainer.py:2134] 2024-07-16 09:08:14,231 >>   Total train batch size (w. parallel, distributed & accumulation) = 128
+
+[INFO|trainer.py:2135] 2024-07-16 09:08:14,231 >>   Gradient Accumulation steps = 8
+
+[INFO|trainer.py:2136] 2024-07-16 09:08:14,231 >>   Total optimization steps = 190
+
+[INFO|trainer.py:2137] 2024-07-16 09:08:14,233 >>   Number of trainable parameters = 6,738,415,616
+
+[INFO|callbacks.py:310] 2024-07-16 09:08:27,214 >> {'loss': 8.3599, 'learning_rate': 5.0000e-07, 'epoch': 0.03, 'throughput': 548.54}
+
+[INFO|callbacks.py:310] 2024-07-16 09:08:38,345 >> {'loss': 8.1891, 'learning_rate': 1.0000e-06, 'epoch': 0.05, 'throughput': 575.99}
+
+[INFO|callbacks.py:310] 2024-07-16 09:08:49,459 >> {'loss': 8.0792, 'learning_rate': 1.5000e-06, 'epoch': 0.08, 'throughput': 586.40}
+
+[INFO|callbacks.py:310] 2024-07-16 09:09:00,554 >> {'loss': 7.9682, 'learning_rate': 2.0000e-06, 'epoch': 0.10, 'throughput': 586.87}
+
+[INFO|callbacks.py:310] 2024-07-16 09:09:11,650 >> {'loss': 6.9482, 'learning_rate': 2.5000e-06, 'epoch': 0.13, 'throughput': 599.42}
+
+[INFO|callbacks.py:310] 2024-07-16 09:09:22,743 >> {'loss': 5.1505, 'learning_rate': 3.0000e-06, 'epoch': 0.15, 'throughput': 599.28}
+
+[INFO|callbacks.py:310] 2024-07-16 09:09:33,861 >> {'loss': 4.7491, 'learning_rate': 3.5000e-06, 'epoch': 0.18, 'throughput': 596.99}
+
+[INFO|callbacks.py:310] 2024-07-16 09:09:44,975 >> {'loss': 3.2164, 'learning_rate': 4.0000e-06, 'epoch': 0.21, 'throughput': 600.21}
+
+[INFO|callbacks.py:310] 2024-07-16 09:09:56,098 >> {'loss': 2.7761, 'learning_rate': 4.5000e-06, 'epoch': 0.23, 'throughput': 603.94}
+
+[INFO|callbacks.py:310] 2024-07-16 09:10:07,219 >> {'loss': 0.6703, 'learning_rate': 5.0000e-06, 'epoch': 0.26, 'throughput': 605.96}
+
+[INFO|callbacks.py:310] 2024-07-16 09:10:18,354 >> {'loss': 0.3255, 'learning_rate': 4.9996e-06, 'epoch': 0.28, 'throughput': 605.48}
+
+[INFO|callbacks.py:310] 2024-07-16 09:10:29,443 >> {'loss': 0.3301, 'learning_rate': 4.9985e-06, 'epoch': 0.31, 'throughput': 605.64}
+
+[INFO|callbacks.py:310] 2024-07-16 09:10:40,548 >> {'loss': 0.2121, 'learning_rate': 4.9966e-06, 'epoch': 0.33, 'throughput': 606.15}
+
+[INFO|callbacks.py:310] 2024-07-16 09:10:51,649 >> {'loss': 1.1565, 'learning_rate': 4.9939e-06, 'epoch': 0.36, 'throughput': 607.41}
+
+[INFO|callbacks.py:310] 2024-07-16 09:11:02,736 >> {'loss': 0.8054, 'learning_rate': 4.9905e-06, 'epoch': 0.39, 'throughput': 606.19}
+
+[INFO|callbacks.py:310] 2024-07-16 09:11:13,851 >> {'loss': 0.2386, 'learning_rate': 4.9863e-06, 'epoch': 0.41, 'throughput': 607.52}
+
+[INFO|callbacks.py:310] 2024-07-16 09:11:24,984 >> {'loss': 0.3161, 'learning_rate': 4.9814e-06, 'epoch': 0.44, 'throughput': 606.78}
+
+[INFO|callbacks.py:310] 2024-07-16 09:11:36,116 >> {'loss': 0.2773, 'learning_rate': 4.9757e-06, 'epoch': 0.46, 'throughput': 607.80}
+
+[INFO|callbacks.py:310] 2024-07-16 09:11:47,247 >> {'loss': 0.2062, 'learning_rate': 4.9692e-06, 'epoch': 0.49, 'throughput': 608.19}
+
+[INFO|callbacks.py:310] 2024-07-16 09:11:58,362 >> {'loss': 0.1837, 'learning_rate': 4.9620e-06, 'epoch': 0.51, 'throughput': 609.22}
+
+[INFO|callbacks.py:310] 2024-07-16 09:12:09,457 >> {'loss': 0.1735, 'learning_rate': 4.9541e-06, 'epoch': 0.54, 'throughput': 610.01}
+
+[INFO|callbacks.py:310] 2024-07-16 09:12:20,540 >> {'loss': 0.1588, 'learning_rate': 4.9454e-06, 'epoch': 0.57, 'throughput': 609.91}
+
+[INFO|callbacks.py:310] 2024-07-16 09:12:31,647 >> {'loss': 0.1443, 'learning_rate': 4.9359e-06, 'epoch': 0.59, 'throughput': 610.82}
+
+[INFO|callbacks.py:310] 2024-07-16 09:12:42,733 >> {'loss': 0.1570, 'learning_rate': 4.9257e-06, 'epoch': 0.62, 'throughput': 609.97}
+
+[INFO|callbacks.py:310] 2024-07-16 09:12:53,861 >> {'loss': 0.1199, 'learning_rate': 4.9148e-06, 'epoch': 0.64, 'throughput': 609.21}
+
+[INFO|callbacks.py:310] 2024-07-16 09:13:04,974 >> {'loss': 0.1539, 'learning_rate': 4.9032e-06, 'epoch': 0.67, 'throughput': 609.20}
+
+[INFO|callbacks.py:310] 2024-07-16 09:13:16,096 >> {'loss': 0.1208, 'learning_rate': 4.8908e-06, 'epoch': 0.69, 'throughput': 609.87}
+
+[INFO|callbacks.py:310] 2024-07-16 09:13:27,217 >> {'loss': 0.0954, 'learning_rate': 4.8776e-06, 'epoch': 0.72, 'throughput': 610.39}
+
+[INFO|callbacks.py:310] 2024-07-16 09:13:38,328 >> {'loss': 0.1387, 'learning_rate': 4.8638e-06, 'epoch': 0.75, 'throughput': 611.13}
+
+[INFO|callbacks.py:310] 2024-07-16 09:13:49,415 >> {'loss': 0.1484, 'learning_rate': 4.8492e-06, 'epoch': 0.77, 'throughput': 612.02}
+
+[INFO|callbacks.py:310] 2024-07-16 09:14:00,513 >> {'loss': 0.0998, 'learning_rate': 4.8340e-06, 'epoch': 0.80, 'throughput': 612.22}
+
+[INFO|callbacks.py:310] 2024-07-16 09:14:11,593 >> {'loss': 0.1068, 'learning_rate': 4.8180e-06, 'epoch': 0.82, 'throughput': 612.05}
+
+[INFO|callbacks.py:310] 2024-07-16 09:14:22,685 >> {'loss': 0.0801, 'learning_rate': 4.8013e-06, 'epoch': 0.85, 'throughput': 612.99}
+
+[INFO|callbacks.py:310] 2024-07-16 09:14:33,813 >> {'loss': 0.1066, 'learning_rate': 4.7839e-06, 'epoch': 0.87, 'throughput': 612.89}
+
+[INFO|callbacks.py:310] 2024-07-16 09:14:44,935 >> {'loss': 0.1038, 'learning_rate': 4.7658e-06, 'epoch': 0.90, 'throughput': 613.01}
+
+[INFO|callbacks.py:310] 2024-07-16 09:14:56,047 >> {'loss': 0.1060, 'learning_rate': 4.7470e-06, 'epoch': 0.93, 'throughput': 612.94}
+
+[INFO|callbacks.py:310] 2024-07-16 09:15:07,172 >> {'loss': 0.1107, 'learning_rate': 4.7275e-06, 'epoch': 0.95, 'throughput': 613.01}
+
+[INFO|callbacks.py:310] 2024-07-16 09:15:18,265 >> {'loss': 0.1372, 'learning_rate': 4.7074e-06, 'epoch': 0.98, 'throughput': 613.54}
+
+[INFO|callbacks.py:310] 2024-07-16 09:15:29,366 >> {'loss': 0.0816, 'learning_rate': 4.6865e-06, 'epoch': 1.00, 'throughput': 613.88}
+
+[INFO|callbacks.py:310] 2024-07-16 09:15:40,449 >> {'loss': 0.0743, 'learning_rate': 4.6651e-06, 'epoch': 1.03, 'throughput': 614.30}
+
+[INFO|callbacks.py:310] 2024-07-16 09:15:51,540 >> {'loss': 0.0720, 'learning_rate': 4.6429e-06, 'epoch': 1.05, 'throughput': 614.77}
+
+[INFO|callbacks.py:310] 2024-07-16 09:16:02,629 >> {'loss': 0.0596, 'learning_rate': 4.6201e-06, 'epoch': 1.08, 'throughput': 614.97}
+
+[INFO|callbacks.py:310] 2024-07-16 09:16:13,746 >> {'loss': 0.0544, 'learning_rate': 4.5967e-06, 'epoch': 1.11, 'throughput': 615.46}
+
+[INFO|callbacks.py:310] 2024-07-16 09:16:24,855 >> {'loss': 0.0342, 'learning_rate': 4.5726e-06, 'epoch': 1.13, 'throughput': 615.55}
+
+[INFO|callbacks.py:310] 2024-07-16 09:16:35,985 >> {'loss': 0.0394, 'learning_rate': 4.5479e-06, 'epoch': 1.16, 'throughput': 615.19}
+
+[INFO|callbacks.py:310] 2024-07-16 09:16:47,103 >> {'loss': 0.0196, 'learning_rate': 4.5225e-06, 'epoch': 1.18, 'throughput': 615.36}
+
+[INFO|callbacks.py:310] 2024-07-16 09:16:58,199 >> {'loss': 0.0411, 'learning_rate': 4.4966e-06, 'epoch': 1.21, 'throughput': 615.43}
+
+[INFO|callbacks.py:310] 2024-07-16 09:17:09,282 >> {'loss': 0.0257, 'learning_rate': 4.4700e-06, 'epoch': 1.23, 'throughput': 614.94}
+
+[INFO|callbacks.py:310] 2024-07-16 09:17:20,373 >> {'loss': 0.0289, 'learning_rate': 4.4429e-06, 'epoch': 1.26, 'throughput': 615.29}
+
+[INFO|callbacks.py:310] 2024-07-16 09:17:31,470 >> {'loss': 0.1193, 'learning_rate': 4.4151e-06, 'epoch': 1.29, 'throughput': 615.01}
+
+[INFO|callbacks.py:310] 2024-07-16 09:17:42,559 >> {'loss': 0.0883, 'learning_rate': 4.3868e-06, 'epoch': 1.31, 'throughput': 614.92}
+
+[INFO|callbacks.py:310] 2024-07-16 09:17:53,670 >> {'loss': 0.0377, 'learning_rate': 4.3579e-06, 'epoch': 1.34, 'throughput': 614.86}
+
+[INFO|callbacks.py:310] 2024-07-16 09:18:04,800 >> {'loss': 0.0602, 'learning_rate': 4.3284e-06, 'epoch': 1.36, 'throughput': 614.73}
+
+[INFO|callbacks.py:310] 2024-07-16 09:18:15,923 >> {'loss': 0.0830, 'learning_rate': 4.2983e-06, 'epoch': 1.39, 'throughput': 614.38}
+
+[INFO|callbacks.py:310] 2024-07-16 09:18:27,039 >> {'loss': 0.0358, 'learning_rate': 4.2678e-06, 'epoch': 1.41, 'throughput': 614.72}
+
+[INFO|callbacks.py:310] 2024-07-16 09:18:38,136 >> {'loss': 0.0321, 'learning_rate': 4.2366e-06, 'epoch': 1.44, 'throughput': 614.84}
+
+[INFO|callbacks.py:310] 2024-07-16 09:18:49,231 >> {'loss': 0.0452, 'learning_rate': 4.2050e-06, 'epoch': 1.47, 'throughput': 615.11}
+
+[INFO|callbacks.py:310] 2024-07-16 09:19:00,331 >> {'loss': 0.0915, 'learning_rate': 4.1728e-06, 'epoch': 1.49, 'throughput': 615.02}
+
+[INFO|callbacks.py:310] 2024-07-16 09:19:11,424 >> {'loss': 0.0651, 'learning_rate': 4.1401e-06, 'epoch': 1.52, 'throughput': 614.81}
+
+[INFO|callbacks.py:310] 2024-07-16 09:19:22,545 >> {'loss': 0.0868, 'learning_rate': 4.1070e-06, 'epoch': 1.54, 'throughput': 614.92}
+
+[INFO|callbacks.py:310] 2024-07-16 09:19:33,666 >> {'loss': 0.0554, 'learning_rate': 4.0733e-06, 'epoch': 1.57, 'throughput': 615.06}
+
+[INFO|callbacks.py:310] 2024-07-16 09:19:44,774 >> {'loss': 0.0336, 'learning_rate': 4.0392e-06, 'epoch': 1.59, 'throughput': 615.29}
+
+[INFO|callbacks.py:310] 2024-07-16 09:19:55,885 >> {'loss': 0.0455, 'learning_rate': 4.0045e-06, 'epoch': 1.62, 'throughput': 615.69}
+
+[INFO|callbacks.py:310] 2024-07-16 09:20:07,002 >> {'loss': 0.0406, 'learning_rate': 3.9695e-06, 'epoch': 1.65, 'throughput': 615.45}
+
+[INFO|callbacks.py:310] 2024-07-16 09:20:18,095 >> {'loss': 0.0461, 'learning_rate': 3.9339e-06, 'epoch': 1.67, 'throughput': 615.37}
+
+[INFO|callbacks.py:310] 2024-07-16 09:20:29,180 >> {'loss': 0.0466, 'learning_rate': 3.8980e-06, 'epoch': 1.70, 'throughput': 615.10}
+
+[INFO|callbacks.py:310] 2024-07-16 09:20:40,282 >> {'loss': 0.0382, 'learning_rate': 3.8616e-06, 'epoch': 1.72, 'throughput': 615.23}
+
+[INFO|callbacks.py:310] 2024-07-16 09:20:51,381 >> {'loss': 0.0426, 'learning_rate': 3.8248e-06, 'epoch': 1.75, 'throughput': 614.90}
+
+[INFO|callbacks.py:310] 2024-07-16 09:21:02,489 >> {'loss': 0.0264, 'learning_rate': 3.7876e-06, 'epoch': 1.77, 'throughput': 615.03}
+
+[INFO|callbacks.py:310] 2024-07-16 09:21:13,594 >> {'loss': 0.0567, 'learning_rate': 3.7500e-06, 'epoch': 1.80, 'throughput': 615.11}
+
+[INFO|callbacks.py:310] 2024-07-16 09:21:24,706 >> {'loss': 0.0688, 'learning_rate': 3.7120e-06, 'epoch': 1.83, 'throughput': 615.35}
+
+[INFO|callbacks.py:310] 2024-07-16 09:21:35,842 >> {'loss': 0.0351, 'learning_rate': 3.6737e-06, 'epoch': 1.85, 'throughput': 614.88}
+
+[INFO|callbacks.py:310] 2024-07-16 09:21:46,947 >> {'loss': 0.0246, 'learning_rate': 3.6350e-06, 'epoch': 1.88, 'throughput': 614.93}
+
+[INFO|callbacks.py:310] 2024-07-16 09:21:58,021 >> {'loss': 0.0364, 'learning_rate': 3.5959e-06, 'epoch': 1.90, 'throughput': 615.23}
+
+[INFO|callbacks.py:310] 2024-07-16 09:22:09,127 >> {'loss': 0.0352, 'learning_rate': 3.5565e-06, 'epoch': 1.93, 'throughput': 615.23}
+
+[INFO|callbacks.py:310] 2024-07-16 09:22:20,219 >> {'loss': 0.0915, 'learning_rate': 3.5168e-06, 'epoch': 1.95, 'throughput': 615.24}
+
+[INFO|callbacks.py:310] 2024-07-16 09:22:31,310 >> {'loss': 0.0327, 'learning_rate': 3.4768e-06, 'epoch': 1.98, 'throughput': 614.95}
+
+[INFO|callbacks.py:310] 2024-07-16 09:22:42,417 >> {'loss': 0.0448, 'learning_rate': 3.4365e-06, 'epoch': 2.01, 'throughput': 615.21}
+
+[INFO|callbacks.py:310] 2024-07-16 09:22:53,536 >> {'loss': 0.0186, 'learning_rate': 3.3959e-06, 'epoch': 2.03, 'throughput': 615.29}
+
+[INFO|callbacks.py:310] 2024-07-16 09:23:04,675 >> {'loss': 0.0342, 'learning_rate': 3.3551e-06, 'epoch': 2.06, 'throughput': 615.30}
+
+[INFO|callbacks.py:310] 2024-07-16 09:23:15,801 >> {'loss': 0.0079, 'learning_rate': 3.3139e-06, 'epoch': 2.08, 'throughput': 615.14}
+
+[INFO|callbacks.py:310] 2024-07-16 09:23:26,896 >> {'loss': 0.0177, 'learning_rate': 3.2725e-06, 'epoch': 2.11, 'throughput': 615.01}
+
+[INFO|callbacks.py:310] 2024-07-16 09:23:37,991 >> {'loss': 0.0139, 'learning_rate': 3.2309e-06, 'epoch': 2.14, 'throughput': 614.74}
+
+[INFO|callbacks.py:310] 2024-07-16 09:23:49,080 >> {'loss': 0.0103, 'learning_rate': 3.1891e-06, 'epoch': 2.16, 'throughput': 615.15}
+
+[INFO|callbacks.py:310] 2024-07-16 09:24:00,169 >> {'loss': 0.0221, 'learning_rate': 3.1470e-06, 'epoch': 2.19, 'throughput': 615.35}
+
+[INFO|callbacks.py:310] 2024-07-16 09:24:11,255 >> {'loss': 0.0021, 'learning_rate': 3.1048e-06, 'epoch': 2.21, 'throughput': 615.26}
+
+[INFO|callbacks.py:310] 2024-07-16 09:24:22,375 >> {'loss': 0.0110, 'learning_rate': 3.0624e-06, 'epoch': 2.24, 'throughput': 615.65}
+
+[INFO|callbacks.py:310] 2024-07-16 09:24:33,470 >> {'loss': 0.0081, 'learning_rate': 3.0198e-06, 'epoch': 2.26, 'throughput': 615.45}
+
+[INFO|callbacks.py:310] 2024-07-16 09:24:44,602 >> {'loss': 0.0149, 'learning_rate': 2.9770e-06, 'epoch': 2.29, 'throughput': 615.35}
+
+[INFO|callbacks.py:310] 2024-07-16 09:24:55,725 >> {'loss': 0.0010, 'learning_rate': 2.9341e-06, 'epoch': 2.32, 'throughput': 615.53}
+
+[INFO|callbacks.py:310] 2024-07-16 09:25:06,826 >> {'loss': 0.0070, 'learning_rate': 2.8911e-06, 'epoch': 2.34, 'throughput': 615.54}
+
+[INFO|callbacks.py:310] 2024-07-16 09:25:17,934 >> {'loss': 0.0089, 'learning_rate': 2.8479e-06, 'epoch': 2.37, 'throughput': 615.48}
+
+[INFO|callbacks.py:310] 2024-07-16 09:25:29,026 >> {'loss': 0.0013, 'learning_rate': 2.8047e-06, 'epoch': 2.39, 'throughput': 615.67}
+
+[INFO|callbacks.py:310] 2024-07-16 09:25:40,116 >> {'loss': 0.0267, 'learning_rate': 2.7613e-06, 'epoch': 2.42, 'throughput': 615.82}
+
+[INFO|callbacks.py:310] 2024-07-16 09:25:51,214 >> {'loss': 0.0171, 'learning_rate': 2.7179e-06, 'epoch': 2.44, 'throughput': 615.76}
+
+[INFO|callbacks.py:310] 2024-07-16 09:26:02,342 >> {'loss': 0.0375, 'learning_rate': 2.6744e-06, 'epoch': 2.47, 'throughput': 615.50}
+
+[INFO|callbacks.py:310] 2024-07-16 09:26:13,469 >> {'loss': 0.0101, 'learning_rate': 2.6308e-06, 'epoch': 2.50, 'throughput': 615.37}
+
+[INFO|callbacks.py:310] 2024-07-16 09:26:24,600 >> {'loss': 0.0282, 'learning_rate': 2.5872e-06, 'epoch': 2.52, 'throughput': 615.50}
+
+[INFO|callbacks.py:310] 2024-07-16 09:26:35,708 >> {'loss': 0.0069, 'learning_rate': 2.5436e-06, 'epoch': 2.55, 'throughput': 615.47}
+
+[INFO|callbacks.py:310] 2024-07-16 09:26:46,803 >> {'loss': 0.0135, 'learning_rate': 2.5000e-06, 'epoch': 2.57, 'throughput': 615.66}
+
+[INFO|callbacks.py:310] 2024-07-16 09:26:57,903 >> {'loss': 0.0062, 'learning_rate': 2.4564e-06, 'epoch': 2.60, 'throughput': 615.71}
+
+[INFO|callbacks.py:310] 2024-07-16 09:27:08,991 >> {'loss': 0.0050, 'learning_rate': 2.4128e-06, 'epoch': 2.62, 'throughput': 615.56}
+
+[INFO|callbacks.py:310] 2024-07-16 09:27:20,085 >> {'loss': 0.0285, 'learning_rate': 2.3692e-06, 'epoch': 2.65, 'throughput': 615.65}
+
+[INFO|callbacks.py:310] 2024-07-16 09:27:31,191 >> {'loss': 0.0225, 'learning_rate': 2.3256e-06, 'epoch': 2.68, 'throughput': 615.86}
+
+[INFO|callbacks.py:310] 2024-07-16 09:27:42,299 >> {'loss': 0.0280, 'learning_rate': 2.2821e-06, 'epoch': 2.70, 'throughput': 615.69}
+
+[INFO|callbacks.py:310] 2024-07-16 09:27:53,416 >> {'loss': 0.0176, 'learning_rate': 2.2387e-06, 'epoch': 2.73, 'throughput': 615.60}
+
+[INFO|callbacks.py:310] 2024-07-16 09:28:04,554 >> {'loss': 0.0047, 'learning_rate': 2.1953e-06, 'epoch': 2.75, 'throughput': 615.36}
+
+[INFO|callbacks.py:310] 2024-07-16 09:28:15,674 >> {'loss': 0.0135, 'learning_rate': 2.1521e-06, 'epoch': 2.78, 'throughput': 615.25}
+
+[INFO|callbacks.py:310] 2024-07-16 09:28:26,766 >> {'loss': 0.0044, 'learning_rate': 2.1089e-06, 'epoch': 2.80, 'throughput': 615.51}
+
+[INFO|callbacks.py:310] 2024-07-16 09:28:37,852 >> {'loss': 0.0252, 'learning_rate': 2.0659e-06, 'epoch': 2.83, 'throughput': 615.50}
+
+[INFO|callbacks.py:310] 2024-07-16 09:28:48,945 >> {'loss': 0.0249, 'learning_rate': 2.0230e-06, 'epoch': 2.86, 'throughput': 615.61}
+
+[INFO|callbacks.py:310] 2024-07-16 09:29:00,043 >> {'loss': 0.0146, 'learning_rate': 1.9802e-06, 'epoch': 2.88, 'throughput': 615.75}
+
+[INFO|callbacks.py:310] 2024-07-16 09:29:11,162 >> {'loss': 0.0044, 'learning_rate': 1.9376e-06, 'epoch': 2.91, 'throughput': 615.69}
+
+[INFO|callbacks.py:310] 2024-07-16 09:29:22,253 >> {'loss': 0.0054, 'learning_rate': 1.8952e-06, 'epoch': 2.93, 'throughput': 615.71}
+
+[INFO|callbacks.py:310] 2024-07-16 09:29:33,390 >> {'loss': 0.0106, 'learning_rate': 1.8530e-06, 'epoch': 2.96, 'throughput': 615.58}
+
+[INFO|callbacks.py:310] 2024-07-16 09:29:44,539 >> {'loss': 0.0167, 'learning_rate': 1.8109e-06, 'epoch': 2.98, 'throughput': 615.53}
+
+[INFO|callbacks.py:310] 2024-07-16 09:29:55,648 >> {'loss': 0.0090, 'learning_rate': 1.7691e-06, 'epoch': 3.01, 'throughput': 615.55}
+
+[INFO|callbacks.py:310] 2024-07-16 09:30:06,727 >> {'loss': 0.0024, 'learning_rate': 1.7275e-06, 'epoch': 3.04, 'throughput': 615.66}
+
+[INFO|callbacks.py:310] 2024-07-16 09:30:17,830 >> {'loss': 0.0235, 'learning_rate': 1.6861e-06, 'epoch': 3.06, 'throughput': 615.60}
+
+[INFO|callbacks.py:310] 2024-07-16 09:30:28,918 >> {'loss': 0.0179, 'learning_rate': 1.6449e-06, 'epoch': 3.09, 'throughput': 615.53}
+
+[INFO|callbacks.py:310] 2024-07-16 09:30:40,012 >> {'loss': 0.0059, 'learning_rate': 1.6041e-06, 'epoch': 3.11, 'throughput': 615.35}
+
+[INFO|callbacks.py:310] 2024-07-16 09:30:51,146 >> {'loss': 0.0017, 'learning_rate': 1.5635e-06, 'epoch': 3.14, 'throughput': 615.08}
+
+[INFO|callbacks.py:310] 2024-07-16 09:31:02,257 >> {'loss': 0.0018, 'learning_rate': 1.5232e-06, 'epoch': 3.16, 'throughput': 615.02}
+
+[INFO|callbacks.py:310] 2024-07-16 09:31:13,381 >> {'loss': 0.0032, 'learning_rate': 1.4832e-06, 'epoch': 3.19, 'throughput': 615.23}
+
+[INFO|callbacks.py:310] 2024-07-16 09:31:24,512 >> {'loss': 0.0019, 'learning_rate': 1.4435e-06, 'epoch': 3.22, 'throughput': 615.29}
+
+[INFO|callbacks.py:310] 2024-07-16 09:31:35,616 >> {'loss': 0.0014, 'learning_rate': 1.4041e-06, 'epoch': 3.24, 'throughput': 615.27}
+
+[INFO|callbacks.py:310] 2024-07-16 09:31:46,705 >> {'loss': 0.0052, 'learning_rate': 1.3650e-06, 'epoch': 3.27, 'throughput': 615.45}
+
+[INFO|callbacks.py:310] 2024-07-16 09:31:57,796 >> {'loss': 0.0005, 'learning_rate': 1.3263e-06, 'epoch': 3.29, 'throughput': 615.55}
+
+[INFO|callbacks.py:310] 2024-07-16 09:32:08,900 >> {'loss': 0.0131, 'learning_rate': 1.2880e-06, 'epoch': 3.32, 'throughput': 615.52}
+
+[INFO|callbacks.py:310] 2024-07-16 09:32:19,991 >> {'loss': 0.0009, 'learning_rate': 1.2500e-06, 'epoch': 3.34, 'throughput': 615.54}
+
+[INFO|callbacks.py:310] 2024-07-16 09:32:31,110 >> {'loss': 0.0057, 'learning_rate': 1.2124e-06, 'epoch': 3.37, 'throughput': 615.63}
+
+[INFO|callbacks.py:310] 2024-07-16 09:32:42,232 >> {'loss': 0.0002, 'learning_rate': 1.1752e-06, 'epoch': 3.40, 'throughput': 615.53}
+
+[INFO|callbacks.py:310] 2024-07-16 09:32:53,354 >> {'loss': 0.0002, 'learning_rate': 1.1384e-06, 'epoch': 3.42, 'throughput': 615.37}
+
+[INFO|callbacks.py:310] 2024-07-16 09:33:04,458 >> {'loss': 0.0145, 'learning_rate': 1.1020e-06, 'epoch': 3.45, 'throughput': 615.56}
+
+[INFO|callbacks.py:310] 2024-07-16 09:33:15,548 >> {'loss': 0.0034, 'learning_rate': 1.0661e-06, 'epoch': 3.47, 'throughput': 615.59}
+
+[INFO|callbacks.py:310] 2024-07-16 09:33:26,645 >> {'loss': 0.0156, 'learning_rate': 1.0305e-06, 'epoch': 3.50, 'throughput': 615.43}
+
+[INFO|callbacks.py:310] 2024-07-16 09:33:37,738 >> {'loss': 0.0013, 'learning_rate': 9.9546e-07, 'epoch': 3.52, 'throughput': 615.59}
+
+[INFO|callbacks.py:310] 2024-07-16 09:33:48,828 >> {'loss': 0.0007, 'learning_rate': 9.6085e-07, 'epoch': 3.55, 'throughput': 615.56}
+
+[INFO|callbacks.py:310] 2024-07-16 09:33:59,926 >> {'loss': 0.0005, 'learning_rate': 9.2670e-07, 'epoch': 3.58, 'throughput': 615.58}
+
+[INFO|callbacks.py:310] 2024-07-16 09:34:11,043 >> {'loss': 0.0034, 'learning_rate': 8.9303e-07, 'epoch': 3.60, 'throughput': 615.52}
+
+[INFO|callbacks.py:310] 2024-07-16 09:34:22,149 >> {'loss': 0.0001, 'learning_rate': 8.5985e-07, 'epoch': 3.63, 'throughput': 615.41}
+
+[INFO|callbacks.py:310] 2024-07-16 09:34:33,264 >> {'loss': 0.0010, 'learning_rate': 8.2717e-07, 'epoch': 3.65, 'throughput': 615.49}
+
+[INFO|callbacks.py:310] 2024-07-16 09:34:44,385 >> {'loss': 0.0123, 'learning_rate': 7.9500e-07, 'epoch': 3.68, 'throughput': 615.44}
+
+[INFO|callbacks.py:310] 2024-07-16 09:34:55,486 >> {'loss': 0.0002, 'learning_rate': 7.6335e-07, 'epoch': 3.70, 'throughput': 615.35}
+
+[INFO|callbacks.py:310] 2024-07-16 09:35:06,571 >> {'loss': 0.0110, 'learning_rate': 7.3223e-07, 'epoch': 3.73, 'throughput': 615.39}
+
+[INFO|callbacks.py:310] 2024-07-16 09:35:17,657 >> {'loss': 0.0008, 'learning_rate': 7.0165e-07, 'epoch': 3.76, 'throughput': 615.17}
+
+[INFO|callbacks.py:310] 2024-07-16 09:35:28,737 >> {'loss': 0.0003, 'learning_rate': 6.7162e-07, 'epoch': 3.78, 'throughput': 615.50}
+
+[INFO|callbacks.py:310] 2024-07-16 09:35:39,839 >> {'loss': 0.0018, 'learning_rate': 6.4214e-07, 'epoch': 3.81, 'throughput': 615.55}
+
+[INFO|callbacks.py:310] 2024-07-16 09:35:50,950 >> {'loss': 0.0016, 'learning_rate': 6.1323e-07, 'epoch': 3.83, 'throughput': 615.59}
+
+[INFO|callbacks.py:310] 2024-07-16 09:36:02,073 >> {'loss': 0.0021, 'learning_rate': 5.8489e-07, 'epoch': 3.86, 'throughput': 615.56}
+
+[INFO|callbacks.py:310] 2024-07-16 09:36:13,188 >> {'loss': 0.0001, 'learning_rate': 5.5714e-07, 'epoch': 3.88, 'throughput': 615.68}
+
+[INFO|callbacks.py:310] 2024-07-16 09:36:24,312 >> {'loss': 0.0001, 'learning_rate': 5.2997e-07, 'epoch': 3.91, 'throughput': 615.50}
+
+[INFO|callbacks.py:310] 2024-07-16 09:36:35,410 >> {'loss': 0.0003, 'learning_rate': 5.0341e-07, 'epoch': 3.94, 'throughput': 615.45}
+
+[INFO|callbacks.py:310] 2024-07-16 09:36:46,505 >> {'loss': 0.0002, 'learning_rate': 4.7746e-07, 'epoch': 3.96, 'throughput': 615.52}
+
+[INFO|callbacks.py:310] 2024-07-16 09:36:57,590 >> {'loss': 0.0002, 'learning_rate': 4.5212e-07, 'epoch': 3.99, 'throughput': 615.41}
+
+[INFO|callbacks.py:310] 2024-07-16 09:37:08,665 >> {'loss': 0.0000, 'learning_rate': 4.2741e-07, 'epoch': 4.01, 'throughput': 615.56}
+
+[INFO|callbacks.py:310] 2024-07-16 09:37:19,763 >> {'loss': 0.0003, 'learning_rate': 4.0332e-07, 'epoch': 4.04, 'throughput': 615.58}
+
+[INFO|callbacks.py:310] 2024-07-16 09:37:30,878 >> {'loss': 0.0002, 'learning_rate': 3.7988e-07, 'epoch': 4.06, 'throughput': 615.55}
+
+[INFO|callbacks.py:310] 2024-07-16 09:37:41,999 >> {'loss': 0.0001, 'learning_rate': 3.5708e-07, 'epoch': 4.09, 'throughput': 615.40}
+
+[INFO|callbacks.py:310] 2024-07-16 09:37:53,113 >> {'loss': 0.0001, 'learning_rate': 3.3494e-07, 'epoch': 4.12, 'throughput': 615.53}
+
+[INFO|callbacks.py:310] 2024-07-16 09:38:04,234 >> {'loss': 0.0001, 'learning_rate': 3.1345e-07, 'epoch': 4.14, 'throughput': 615.52}
+
+[INFO|callbacks.py:310] 2024-07-16 09:38:15,321 >> {'loss': 0.0000, 'learning_rate': 2.9263e-07, 'epoch': 4.17, 'throughput': 615.59}
+
+[INFO|callbacks.py:310] 2024-07-16 09:38:26,408 >> {'loss': 0.0001, 'learning_rate': 2.7248e-07, 'epoch': 4.19, 'throughput': 615.69}
+
+[INFO|callbacks.py:310] 2024-07-16 09:38:37,489 >> {'loss': 0.0000, 'learning_rate': 2.5301e-07, 'epoch': 4.22, 'throughput': 615.71}
+
+[INFO|callbacks.py:310] 2024-07-16 09:38:48,575 >> {'loss': 0.0001, 'learning_rate': 2.3423e-07, 'epoch': 4.24, 'throughput': 615.55}
+
+[INFO|callbacks.py:310] 2024-07-16 09:38:59,677 >> {'loss': 0.0001, 'learning_rate': 2.1614e-07, 'epoch': 4.27, 'throughput': 615.61}
+
+[INFO|callbacks.py:310] 2024-07-16 09:39:10,799 >> {'loss': 0.0002, 'learning_rate': 1.9874e-07, 'epoch': 4.30, 'throughput': 615.64}
+
+[INFO|callbacks.py:310] 2024-07-16 09:39:21,928 >> {'loss': 0.0002, 'learning_rate': 1.8204e-07, 'epoch': 4.32, 'throughput': 615.58}
+
+[INFO|callbacks.py:310] 2024-07-16 09:39:33,061 >> {'loss': 0.0001, 'learning_rate': 1.6605e-07, 'epoch': 4.35, 'throughput': 615.46}
+
+[INFO|callbacks.py:310] 2024-07-16 09:39:44,166 >> {'loss': 0.0001, 'learning_rate': 1.5077e-07, 'epoch': 4.37, 'throughput': 615.43}
+
+[INFO|callbacks.py:310] 2024-07-16 09:39:55,251 >> {'loss': 0.0115, 'learning_rate': 1.3620e-07, 'epoch': 4.40, 'throughput': 615.48}
+
+[INFO|callbacks.py:310] 2024-07-16 09:40:06,330 >> {'loss': 0.0005, 'learning_rate': 1.2236e-07, 'epoch': 4.42, 'throughput': 615.47}
+
+[INFO|callbacks.py:310] 2024-07-16 09:40:17,431 >> {'loss': 0.0003, 'learning_rate': 1.0924e-07, 'epoch': 4.45, 'throughput': 615.57}
+
+[INFO|callbacks.py:310] 2024-07-16 09:40:28,525 >> {'loss': 0.0050, 'learning_rate': 9.6846e-08, 'epoch': 4.48, 'throughput': 615.49}
+
+[INFO|callbacks.py:310] 2024-07-16 09:40:39,640 >> {'loss': 0.0003, 'learning_rate': 8.5185e-08, 'epoch': 4.50, 'throughput': 615.38}
+
+[INFO|callbacks.py:310] 2024-07-16 09:40:50,749 >> {'loss': 0.0008, 'learning_rate': 7.4261e-08, 'epoch': 4.53, 'throughput': 615.27}
+
+[INFO|callbacks.py:310] 2024-07-16 09:41:01,862 >> {'loss': 0.0000, 'learning_rate': 6.4075e-08, 'epoch': 4.55, 'throughput': 615.37}
+
+[INFO|callbacks.py:310] 2024-07-16 09:41:12,986 >> {'loss': 0.0000, 'learning_rate': 5.4631e-08, 'epoch': 4.58, 'throughput': 615.39}
+
+[INFO|callbacks.py:310] 2024-07-16 09:41:24,079 >> {'loss': 0.0042, 'learning_rate': 4.5932e-08, 'epoch': 4.60, 'throughput': 615.48}
+
+[INFO|callbacks.py:310] 2024-07-16 09:41:35,169 >> {'loss': 0.0004, 'learning_rate': 3.7981e-08, 'epoch': 4.63, 'throughput': 615.54}
+
+[INFO|callbacks.py:310] 2024-07-16 09:41:46,249 >> {'loss': 0.0001, 'learning_rate': 3.0779e-08, 'epoch': 4.66, 'throughput': 615.42}
+
+[INFO|callbacks.py:310] 2024-07-16 09:41:57,352 >> {'loss': 0.0001, 'learning_rate': 2.4330e-08, 'epoch': 4.68, 'throughput': 615.30}
+
+[INFO|callbacks.py:310] 2024-07-16 09:42:08,449 >> {'loss': 0.0004, 'learning_rate': 1.8635e-08, 'epoch': 4.71, 'throughput': 615.12}
+
+[INFO|callbacks.py:310] 2024-07-16 09:42:19,548 >> {'loss': 0.0000, 'learning_rate': 1.3695e-08, 'epoch': 4.73, 'throughput': 615.05}
+
+[INFO|callbacks.py:310] 2024-07-16 09:42:30,662 >> {'loss': 0.0009, 'learning_rate': 9.5133e-09, 'epoch': 4.76, 'throughput': 615.11}
+
+[INFO|callbacks.py:310] 2024-07-16 09:42:41,790 >> {'loss': 0.0001, 'learning_rate': 6.0899e-09, 'epoch': 4.78, 'throughput': 615.12}
+
+[INFO|callbacks.py:310] 2024-07-16 09:42:52,921 >> {'loss': 0.0004, 'learning_rate': 3.4262e-09, 'epoch': 4.81, 'throughput': 615.30}
+
+[INFO|callbacks.py:310] 2024-07-16 09:43:04,012 >> {'loss': 0.0002, 'learning_rate': 1.5229e-09, 'epoch': 4.84, 'throughput': 615.28}
+
+[INFO|callbacks.py:310] 2024-07-16 09:43:15,108 >> {'loss': 0.0000, 'learning_rate': 3.8076e-10, 'epoch': 4.86, 'throughput': 615.26}
+
+[INFO|callbacks.py:310] 2024-07-16 09:43:26,201 >> {'loss': 0.0001, 'learning_rate': 0.0000e+00, 'epoch': 4.89, 'throughput': 615.25}
+
+[INFO|trainer.py:3478] 2024-07-16 09:43:32,570 >> Saving model checkpoint to saves/LLaMA2-7B-Chat/full/train_2024-07-16-09-05-28_llama2/checkpoint-190
+
+[INFO|configuration_utils.py:472] 2024-07-16 09:43:32,573 >> Configuration saved in saves/LLaMA2-7B-Chat/full/train_2024-07-16-09-05-28_llama2/checkpoint-190/config.json
+
+[INFO|configuration_utils.py:769] 2024-07-16 09:43:32,573 >> Configuration saved in saves/LLaMA2-7B-Chat/full/train_2024-07-16-09-05-28_llama2/checkpoint-190/generation_config.json
+
+[INFO|modeling_utils.py:2698] 2024-07-16 09:43:46,233 >> The model is bigger than the maximum size per checkpoint (5GB) and is going to be split in 3 checkpoint shards. You can find where each parameters has been saved in the index located at saves/LLaMA2-7B-Chat/full/train_2024-07-16-09-05-28_llama2/checkpoint-190/model.safetensors.index.json.
+
+[INFO|tokenization_utils_base.py:2574] 2024-07-16 09:43:46,233 >> tokenizer config file saved in saves/LLaMA2-7B-Chat/full/train_2024-07-16-09-05-28_llama2/checkpoint-190/tokenizer_config.json
+
+[INFO|tokenization_utils_base.py:2583] 2024-07-16 09:43:46,234 >> Special tokens file saved in saves/LLaMA2-7B-Chat/full/train_2024-07-16-09-05-28_llama2/checkpoint-190/special_tokens_map.json
+
+[INFO|trainer.py:2383] 2024-07-16 09:44:16,328 >> 
+
+Training completed. Do not forget to share your model on huggingface.co/models =)
+
+
+
+[INFO|trainer.py:3478] 2024-07-16 09:44:22,736 >> Saving model checkpoint to saves/LLaMA2-7B-Chat/full/train_2024-07-16-09-05-28_llama2
+
+[INFO|configuration_utils.py:472] 2024-07-16 09:44:22,738 >> Configuration saved in saves/LLaMA2-7B-Chat/full/train_2024-07-16-09-05-28_llama2/config.json
+
+[INFO|configuration_utils.py:769] 2024-07-16 09:44:22,739 >> Configuration saved in saves/LLaMA2-7B-Chat/full/train_2024-07-16-09-05-28_llama2/generation_config.json
+
+[INFO|modeling_utils.py:2698] 2024-07-16 09:44:36,499 >> The model is bigger than the maximum size per checkpoint (5GB) and is going to be split in 3 checkpoint shards. You can find where each parameters has been saved in the index located at saves/LLaMA2-7B-Chat/full/train_2024-07-16-09-05-28_llama2/model.safetensors.index.json.
+
+[INFO|tokenization_utils_base.py:2574] 2024-07-16 09:44:36,499 >> tokenizer config file saved in saves/LLaMA2-7B-Chat/full/train_2024-07-16-09-05-28_llama2/tokenizer_config.json
+
+[INFO|tokenization_utils_base.py:2583] 2024-07-16 09:44:36,499 >> Special tokens file saved in saves/LLaMA2-7B-Chat/full/train_2024-07-16-09-05-28_llama2/special_tokens_map.json
+
+[WARNING|ploting.py:89] 2024-07-16 09:44:37,565 >> No metric eval_loss to plot.
+
+[WARNING|ploting.py:89] 2024-07-16 09:44:37,565 >> No metric eval_accuracy to plot.
+
+[INFO|modelcard.py:449] 2024-07-16 09:44:37,565 >> Dropping the following result as it does not have all the necessary fields:
+{'task': {'name': 'Causal Language Modeling', 'type': 'text-generation'}}
+

special_tokens_map.json

@@ -0,0 +1,24 @@
+{
+  "bos_token": {
+    "content": "<s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "</s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "</s>",
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer.model

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

tokenizer_config.json

@@ -0,0 +1,44 @@
+{
+  "add_bos_token": true,
+  "add_eos_token": false,
+  "add_prefix_space": null,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<s>",
+  "chat_template": "{% if messages[0]['role'] == 'system' %}{% set system_message = messages[0]['content'] %}{% endif %}{% for message in messages %}{% set content = message['content'] %}{% if loop.index0 == 0 and system_message is defined %}{% set content = '<<SYS>>\n' + system_message + '\n<</SYS>>\n\n' + message['content'] %}{% endif %}{% if message['role'] == 'user' %}{{ '<s>' + '[INST] ' + content + ' [/INST]' }}{% elif message['role'] == 'assistant' %}{{ content + '</s>' }}{% endif %}{% endfor %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "</s>",
+  "legacy": false,
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "</s>",
+  "padding_side": "right",
+  "sp_model_kwargs": {},
+  "split_special_tokens": false,
+  "tokenizer_class": "LlamaTokenizer",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false
+}

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 4.887459807073955,
+    "num_input_tokens_seen": 1299392,
+    "total_flos": 5.151317702790349e+16,
+    "train_loss": 0.3433768034317166,
+    "train_runtime": 2162.0959,
+    "train_samples_per_second": 11.489,
+    "train_steps_per_second": 0.088
+}

trainer_log.jsonl

@@ -0,0 +1,191 @@
+{"current_steps": 1, "total_steps": 190, "loss": 8.3599, "learning_rate": 5.000000000000001e-07, "epoch": 0.02572347266881029, "percentage": 0.53, "elapsed_time": "0:00:12", "remaining_time": "0:40:53", "throughput": "548.54", "total_tokens": 7120}
+{"current_steps": 2, "total_steps": 190, "loss": 8.1891, "learning_rate": 1.0000000000000002e-06, "epoch": 0.05144694533762058, "percentage": 1.05, "elapsed_time": "0:00:24", "remaining_time": "0:37:46", "throughput": "575.99", "total_tokens": 13888}
+{"current_steps": 3, "total_steps": 190, "loss": 8.0792, "learning_rate": 1.5e-06, "epoch": 0.07717041800643087, "percentage": 1.58, "elapsed_time": "0:00:35", "remaining_time": "0:36:35", "throughput": "586.40", "total_tokens": 20656}
+{"current_steps": 4, "total_steps": 190, "loss": 7.9682, "learning_rate": 2.0000000000000003e-06, "epoch": 0.10289389067524116, "percentage": 2.11, "elapsed_time": "0:00:46", "remaining_time": "0:35:53", "throughput": "586.87", "total_tokens": 27184}
+{"current_steps": 5, "total_steps": 190, "loss": 6.9482, "learning_rate": 2.5e-06, "epoch": 0.12861736334405144, "percentage": 2.63, "elapsed_time": "0:00:57", "remaining_time": "0:35:24", "throughput": "599.42", "total_tokens": 34416}
+{"current_steps": 6, "total_steps": 190, "loss": 5.1505, "learning_rate": 3e-06, "epoch": 0.15434083601286175, "percentage": 3.16, "elapsed_time": "0:01:08", "remaining_time": "0:35:00", "throughput": "599.28", "total_tokens": 41056}
+{"current_steps": 7, "total_steps": 190, "loss": 4.7491, "learning_rate": 3.5e-06, "epoch": 0.18006430868167203, "percentage": 3.68, "elapsed_time": "0:01:19", "remaining_time": "0:34:41", "throughput": "596.99", "total_tokens": 47536}
+{"current_steps": 8, "total_steps": 190, "loss": 3.2164, "learning_rate": 4.000000000000001e-06, "epoch": 0.2057877813504823, "percentage": 4.21, "elapsed_time": "0:01:30", "remaining_time": "0:34:24", "throughput": "600.21", "total_tokens": 54464}
+{"current_steps": 9, "total_steps": 190, "loss": 2.7761, "learning_rate": 4.5e-06, "epoch": 0.2315112540192926, "percentage": 4.74, "elapsed_time": "0:01:41", "remaining_time": "0:34:08", "throughput": "603.94", "total_tokens": 61520}
+{"current_steps": 10, "total_steps": 190, "loss": 0.6703, "learning_rate": 5e-06, "epoch": 0.2572347266881029, "percentage": 5.26, "elapsed_time": "0:01:52", "remaining_time": "0:33:53", "throughput": "605.96", "total_tokens": 68464}
+{"current_steps": 11, "total_steps": 190, "loss": 0.3255, "learning_rate": 4.9996192378909785e-06, "epoch": 0.2829581993569132, "percentage": 5.79, "elapsed_time": "0:02:04", "remaining_time": "0:33:39", "throughput": "605.48", "total_tokens": 75152}
+{"current_steps": 12, "total_steps": 190, "loss": 0.3301, "learning_rate": 4.99847706754774e-06, "epoch": 0.3086816720257235, "percentage": 6.32, "elapsed_time": "0:02:15", "remaining_time": "0:33:25", "throughput": "605.64", "total_tokens": 81888}
+{"current_steps": 13, "total_steps": 190, "loss": 0.2121, "learning_rate": 4.9965738368864345e-06, "epoch": 0.33440514469453375, "percentage": 6.84, "elapsed_time": "0:02:26", "remaining_time": "0:33:12", "throughput": "606.15", "total_tokens": 88688}
+{"current_steps": 14, "total_steps": 190, "loss": 1.1565, "learning_rate": 4.993910125649561e-06, "epoch": 0.36012861736334406, "percentage": 7.37, "elapsed_time": "0:02:37", "remaining_time": "0:32:58", "throughput": "607.41", "total_tokens": 95616}
+{"current_steps": 15, "total_steps": 190, "loss": 0.8054, "learning_rate": 4.990486745229364e-06, "epoch": 0.3858520900321543, "percentage": 7.89, "elapsed_time": "0:02:48", "remaining_time": "0:32:45", "throughput": "606.19", "total_tokens": 102144}
+{"current_steps": 16, "total_steps": 190, "loss": 0.2386, "learning_rate": 4.986304738420684e-06, "epoch": 0.4115755627009646, "percentage": 8.42, "elapsed_time": "0:02:59", "remaining_time": "0:32:33", "throughput": "607.52", "total_tokens": 109120}
+{"current_steps": 17, "total_steps": 190, "loss": 0.3161, "learning_rate": 4.981365379103306e-06, "epoch": 0.43729903536977494, "percentage": 8.95, "elapsed_time": "0:03:10", "remaining_time": "0:32:21", "throughput": "606.78", "total_tokens": 115744}
+{"current_steps": 18, "total_steps": 190, "loss": 0.2773, "learning_rate": 4.975670171853926e-06, "epoch": 0.4630225080385852, "percentage": 9.47, "elapsed_time": "0:03:21", "remaining_time": "0:32:09", "throughput": "607.80", "total_tokens": 122704}
+{"current_steps": 19, "total_steps": 190, "loss": 0.2062, "learning_rate": 4.9692208514878445e-06, "epoch": 0.4887459807073955, "percentage": 10.0, "elapsed_time": "0:03:33", "remaining_time": "0:31:57", "throughput": "608.19", "total_tokens": 129552}
+{"current_steps": 20, "total_steps": 190, "loss": 0.1837, "learning_rate": 4.962019382530521e-06, "epoch": 0.5144694533762058, "percentage": 10.53, "elapsed_time": "0:03:44", "remaining_time": "0:31:45", "throughput": "609.22", "total_tokens": 136544}
+{"current_steps": 21, "total_steps": 190, "loss": 0.1735, "learning_rate": 4.9540679586191605e-06, "epoch": 0.5401929260450161, "percentage": 11.05, "elapsed_time": "0:03:55", "remaining_time": "0:31:32", "throughput": "610.01", "total_tokens": 143488}
+{"current_steps": 22, "total_steps": 190, "loss": 0.1588, "learning_rate": 4.9453690018345144e-06, "epoch": 0.5659163987138264, "percentage": 11.58, "elapsed_time": "0:04:06", "remaining_time": "0:31:20", "throughput": "609.91", "total_tokens": 150224}
+{"current_steps": 23, "total_steps": 190, "loss": 0.1443, "learning_rate": 4.935925161963089e-06, "epoch": 0.5916398713826366, "percentage": 12.11, "elapsed_time": "0:04:17", "remaining_time": "0:31:09", "throughput": "610.82", "total_tokens": 157232}
+{"current_steps": 24, "total_steps": 190, "loss": 0.157, "learning_rate": 4.925739315689991e-06, "epoch": 0.617363344051447, "percentage": 12.63, "elapsed_time": "0:04:28", "remaining_time": "0:30:57", "throughput": "609.97", "total_tokens": 163776}
+{"current_steps": 25, "total_steps": 190, "loss": 0.1199, "learning_rate": 4.914814565722671e-06, "epoch": 0.6430868167202572, "percentage": 13.16, "elapsed_time": "0:04:39", "remaining_time": "0:30:45", "throughput": "609.21", "total_tokens": 170352}
+{"current_steps": 26, "total_steps": 190, "loss": 0.1539, "learning_rate": 4.903154239845798e-06, "epoch": 0.6688102893890675, "percentage": 13.68, "elapsed_time": "0:04:50", "remaining_time": "0:30:33", "throughput": "609.20", "total_tokens": 177120}
+{"current_steps": 27, "total_steps": 190, "loss": 0.1208, "learning_rate": 4.890761889907589e-06, "epoch": 0.6945337620578779, "percentage": 14.21, "elapsed_time": "0:05:01", "remaining_time": "0:30:22", "throughput": "609.87", "total_tokens": 184096}
+{"current_steps": 28, "total_steps": 190, "loss": 0.0954, "learning_rate": 4.8776412907378845e-06, "epoch": 0.7202572347266881, "percentage": 14.74, "elapsed_time": "0:05:12", "remaining_time": "0:30:10", "throughput": "610.39", "total_tokens": 191040}
+{"current_steps": 29, "total_steps": 190, "loss": 0.1387, "learning_rate": 4.863796438998293e-06, "epoch": 0.7459807073954984, "percentage": 15.26, "elapsed_time": "0:05:24", "remaining_time": "0:29:59", "throughput": "611.13", "total_tokens": 198064}
+{"current_steps": 30, "total_steps": 190, "loss": 0.1484, "learning_rate": 4.849231551964771e-06, "epoch": 0.7717041800643086, "percentage": 15.79, "elapsed_time": "0:05:35", "remaining_time": "0:29:47", "throughput": "612.02", "total_tokens": 205136}
+{"current_steps": 31, "total_steps": 190, "loss": 0.0998, "learning_rate": 4.833951066243004e-06, "epoch": 0.797427652733119, "percentage": 16.32, "elapsed_time": "0:05:46", "remaining_time": "0:29:36", "throughput": "612.22", "total_tokens": 212000}
+{"current_steps": 32, "total_steps": 190, "loss": 0.1068, "learning_rate": 4.817959636416969e-06, "epoch": 0.8231511254019293, "percentage": 16.84, "elapsed_time": "0:05:57", "remaining_time": "0:29:24", "throughput": "612.05", "total_tokens": 218720}
+{"current_steps": 33, "total_steps": 190, "loss": 0.0801, "learning_rate": 4.801262133631101e-06, "epoch": 0.8488745980707395, "percentage": 17.37, "elapsed_time": "0:06:08", "remaining_time": "0:29:12", "throughput": "612.99", "total_tokens": 225856}
+{"current_steps": 34, "total_steps": 190, "loss": 0.1066, "learning_rate": 4.783863644106502e-06, "epoch": 0.8745980707395499, "percentage": 17.89, "elapsed_time": "0:06:19", "remaining_time": "0:29:01", "throughput": "612.89", "total_tokens": 232640}
+{"current_steps": 35, "total_steps": 190, "loss": 0.1038, "learning_rate": 4.765769467591626e-06, "epoch": 0.9003215434083601, "percentage": 18.42, "elapsed_time": "0:06:30", "remaining_time": "0:28:50", "throughput": "613.01", "total_tokens": 239504}
+{"current_steps": 36, "total_steps": 190, "loss": 0.106, "learning_rate": 4.746985115747918e-06, "epoch": 0.9260450160771704, "percentage": 18.95, "elapsed_time": "0:06:41", "remaining_time": "0:28:38", "throughput": "612.94", "total_tokens": 246288}
+{"current_steps": 37, "total_steps": 190, "loss": 0.1107, "learning_rate": 4.72751631047092e-06, "epoch": 0.9517684887459807, "percentage": 19.47, "elapsed_time": "0:06:52", "remaining_time": "0:28:27", "throughput": "613.01", "total_tokens": 253136}
+{"current_steps": 38, "total_steps": 190, "loss": 0.1372, "learning_rate": 4.707368982147318e-06, "epoch": 0.977491961414791, "percentage": 20.0, "elapsed_time": "0:07:04", "remaining_time": "0:28:16", "throughput": "613.54", "total_tokens": 260160}
+{"current_steps": 39, "total_steps": 190, "loss": 0.0816, "learning_rate": 4.68654926784849e-06, "epoch": 1.0032154340836013, "percentage": 20.53, "elapsed_time": "0:07:15", "remaining_time": "0:28:04", "throughput": "613.88", "total_tokens": 267120}
+{"current_steps": 40, "total_steps": 190, "loss": 0.0743, "learning_rate": 4.665063509461098e-06, "epoch": 1.0289389067524115, "percentage": 21.05, "elapsed_time": "0:07:26", "remaining_time": "0:27:53", "throughput": "614.30", "total_tokens": 274112}
+{"current_steps": 41, "total_steps": 190, "loss": 0.072, "learning_rate": 4.642918251755281e-06, "epoch": 1.0546623794212218, "percentage": 21.58, "elapsed_time": "0:07:37", "remaining_time": "0:27:41", "throughput": "614.77", "total_tokens": 281136}
+{"current_steps": 42, "total_steps": 190, "loss": 0.0596, "learning_rate": 4.620120240391065e-06, "epoch": 1.0803858520900322, "percentage": 22.11, "elapsed_time": "0:07:48", "remaining_time": "0:27:30", "throughput": "614.97", "total_tokens": 288048}
+{"current_steps": 43, "total_steps": 190, "loss": 0.0544, "learning_rate": 4.596676419863561e-06, "epoch": 1.1061093247588425, "percentage": 22.63, "elapsed_time": "0:07:59", "remaining_time": "0:27:19", "throughput": "615.46", "total_tokens": 295120}
+{"current_steps": 44, "total_steps": 190, "loss": 0.0342, "learning_rate": 4.572593931387604e-06, "epoch": 1.1318327974276527, "percentage": 23.16, "elapsed_time": "0:08:10", "remaining_time": "0:27:07", "throughput": "615.55", "total_tokens": 302000}
+{"current_steps": 45, "total_steps": 190, "loss": 0.0394, "learning_rate": 4.54788011072248e-06, "epoch": 1.157556270096463, "percentage": 23.68, "elapsed_time": "0:08:21", "remaining_time": "0:26:56", "throughput": "615.19", "total_tokens": 308672}
+{"current_steps": 46, "total_steps": 190, "loss": 0.0196, "learning_rate": 4.522542485937369e-06, "epoch": 1.1832797427652733, "percentage": 24.21, "elapsed_time": "0:08:32", "remaining_time": "0:26:45", "throughput": "615.36", "total_tokens": 315600}
+{"current_steps": 47, "total_steps": 190, "loss": 0.0411, "learning_rate": 4.496588775118232e-06, "epoch": 1.2090032154340835, "percentage": 24.74, "elapsed_time": "0:08:43", "remaining_time": "0:26:34", "throughput": "615.43", "total_tokens": 322464}
+{"current_steps": 48, "total_steps": 190, "loss": 0.0257, "learning_rate": 4.470026884016805e-06, "epoch": 1.234726688102894, "percentage": 25.26, "elapsed_time": "0:08:55", "remaining_time": "0:26:22", "throughput": "614.94", "total_tokens": 329024}
+{"current_steps": 49, "total_steps": 190, "loss": 0.0289, "learning_rate": 4.442864903642428e-06, "epoch": 1.2604501607717042, "percentage": 25.79, "elapsed_time": "0:09:06", "remaining_time": "0:26:11", "throughput": "615.29", "total_tokens": 336032}
+{"current_steps": 50, "total_steps": 190, "loss": 0.1193, "learning_rate": 4.415111107797445e-06, "epoch": 1.2861736334405145, "percentage": 26.32, "elapsed_time": "0:09:17", "remaining_time": "0:26:00", "throughput": "615.01", "total_tokens": 342704}
+{"current_steps": 51, "total_steps": 190, "loss": 0.0883, "learning_rate": 4.386773950556931e-06, "epoch": 1.3118971061093248, "percentage": 26.84, "elapsed_time": "0:09:28", "remaining_time": "0:25:48", "throughput": "614.92", "total_tokens": 349472}
+{"current_steps": 52, "total_steps": 190, "loss": 0.0377, "learning_rate": 4.357862063693486e-06, "epoch": 1.337620578778135, "percentage": 27.37, "elapsed_time": "0:09:39", "remaining_time": "0:25:37", "throughput": "614.86", "total_tokens": 356272}
+{"current_steps": 53, "total_steps": 190, "loss": 0.0602, "learning_rate": 4.328384254047927e-06, "epoch": 1.3633440514469453, "percentage": 27.89, "elapsed_time": "0:09:50", "remaining_time": "0:25:26", "throughput": "614.73", "total_tokens": 363040}
+{"current_steps": 54, "total_steps": 190, "loss": 0.083, "learning_rate": 4.2983495008466285e-06, "epoch": 1.3890675241157555, "percentage": 28.42, "elapsed_time": "0:10:01", "remaining_time": "0:25:15", "throughput": "614.38", "total_tokens": 369664}
+{"current_steps": 55, "total_steps": 190, "loss": 0.0358, "learning_rate": 4.267766952966369e-06, "epoch": 1.414790996784566, "percentage": 28.95, "elapsed_time": "0:10:12", "remaining_time": "0:25:04", "throughput": "614.72", "total_tokens": 376704}
+{"current_steps": 56, "total_steps": 190, "loss": 0.0321, "learning_rate": 4.236645926147493e-06, "epoch": 1.4405144694533762, "percentage": 29.47, "elapsed_time": "0:10:23", "remaining_time": "0:24:52", "throughput": "614.84", "total_tokens": 383600}
+{"current_steps": 57, "total_steps": 190, "loss": 0.0452, "learning_rate": 4.204995900156247e-06, "epoch": 1.4662379421221865, "percentage": 30.0, "elapsed_time": "0:10:34", "remaining_time": "0:24:41", "throughput": "615.11", "total_tokens": 390592}
+{"current_steps": 58, "total_steps": 190, "loss": 0.0915, "learning_rate": 4.172826515897146e-06, "epoch": 1.4919614147909968, "percentage": 30.53, "elapsed_time": "0:10:46", "remaining_time": "0:24:30", "throughput": "615.02", "total_tokens": 397360}
+{"current_steps": 59, "total_steps": 190, "loss": 0.0651, "learning_rate": 4.140147572476269e-06, "epoch": 1.517684887459807, "percentage": 31.05, "elapsed_time": "0:10:57", "remaining_time": "0:24:19", "throughput": "614.81", "total_tokens": 404048}
+{"current_steps": 60, "total_steps": 190, "loss": 0.0868, "learning_rate": 4.106969024216348e-06, "epoch": 1.5434083601286175, "percentage": 31.58, "elapsed_time": "0:11:08", "remaining_time": "0:24:08", "throughput": "614.92", "total_tokens": 410960}
+{"current_steps": 61, "total_steps": 190, "loss": 0.0554, "learning_rate": 4.073300977624594e-06, "epoch": 1.5691318327974275, "percentage": 32.11, "elapsed_time": "0:11:19", "remaining_time": "0:23:56", "throughput": "615.06", "total_tokens": 417888}
+{"current_steps": 62, "total_steps": 190, "loss": 0.0336, "learning_rate": 4.039153688314146e-06, "epoch": 1.594855305466238, "percentage": 32.63, "elapsed_time": "0:11:30", "remaining_time": "0:23:45", "throughput": "615.29", "total_tokens": 424880}
+{"current_steps": 63, "total_steps": 190, "loss": 0.0455, "learning_rate": 4.0045375578801216e-06, "epoch": 1.6205787781350482, "percentage": 33.16, "elapsed_time": "0:11:41", "remaining_time": "0:23:34", "throughput": "615.69", "total_tokens": 432000}
+{"current_steps": 64, "total_steps": 190, "loss": 0.0406, "learning_rate": 3.969463130731183e-06, "epoch": 1.6463022508038585, "percentage": 33.68, "elapsed_time": "0:11:52", "remaining_time": "0:23:23", "throughput": "615.45", "total_tokens": 438672}
+{"current_steps": 65, "total_steps": 190, "loss": 0.0461, "learning_rate": 3.933941090877615e-06, "epoch": 1.6720257234726688, "percentage": 34.21, "elapsed_time": "0:12:03", "remaining_time": "0:23:12", "throughput": "615.37", "total_tokens": 445440}
+{"current_steps": 66, "total_steps": 190, "loss": 0.0466, "learning_rate": 3.897982258676867e-06, "epoch": 1.697749196141479, "percentage": 34.74, "elapsed_time": "0:12:14", "remaining_time": "0:23:00", "throughput": "615.10", "total_tokens": 452064}
+{"current_steps": 67, "total_steps": 190, "loss": 0.0382, "learning_rate": 3.861597587537568e-06, "epoch": 1.7234726688102895, "percentage": 35.26, "elapsed_time": "0:12:26", "remaining_time": "0:22:49", "throughput": "615.23", "total_tokens": 458992}
+{"current_steps": 68, "total_steps": 190, "loss": 0.0426, "learning_rate": 3.824798160583012e-06, "epoch": 1.7491961414790995, "percentage": 35.79, "elapsed_time": "0:12:37", "remaining_time": "0:22:38", "throughput": "614.90", "total_tokens": 465568}
+{"current_steps": 69, "total_steps": 190, "loss": 0.0264, "learning_rate": 3.787595187275136e-06, "epoch": 1.77491961414791, "percentage": 36.32, "elapsed_time": "0:12:48", "remaining_time": "0:22:27", "throughput": "615.03", "total_tokens": 472496}
+{"current_steps": 70, "total_steps": 190, "loss": 0.0567, "learning_rate": 3.7500000000000005e-06, "epoch": 1.8006430868167203, "percentage": 36.84, "elapsed_time": "0:12:59", "remaining_time": "0:22:16", "throughput": "615.11", "total_tokens": 479392}
+{"current_steps": 71, "total_steps": 190, "loss": 0.0688, "learning_rate": 3.7120240506158433e-06, "epoch": 1.8263665594855305, "percentage": 37.37, "elapsed_time": "0:13:10", "remaining_time": "0:22:04", "throughput": "615.35", "total_tokens": 486416}
+{"current_steps": 72, "total_steps": 190, "loss": 0.0351, "learning_rate": 3.6736789069647273e-06, "epoch": 1.852090032154341, "percentage": 37.89, "elapsed_time": "0:13:21", "remaining_time": "0:21:53", "throughput": "614.88", "total_tokens": 492896}
+{"current_steps": 73, "total_steps": 190, "loss": 0.0246, "learning_rate": 3.634976249348867e-06, "epoch": 1.877813504823151, "percentage": 38.42, "elapsed_time": "0:13:32", "remaining_time": "0:21:42", "throughput": "614.93", "total_tokens": 499760}
+{"current_steps": 74, "total_steps": 190, "loss": 0.0364, "learning_rate": 3.595927866972694e-06, "epoch": 1.9035369774919615, "percentage": 38.95, "elapsed_time": "0:13:43", "remaining_time": "0:21:31", "throughput": "615.23", "total_tokens": 506816}
+{"current_steps": 75, "total_steps": 190, "loss": 0.0352, "learning_rate": 3.556545654351749e-06, "epoch": 1.9292604501607717, "percentage": 39.47, "elapsed_time": "0:13:54", "remaining_time": "0:21:20", "throughput": "615.23", "total_tokens": 513648}
+{"current_steps": 76, "total_steps": 190, "loss": 0.0915, "learning_rate": 3.516841607689501e-06, "epoch": 1.954983922829582, "percentage": 40.0, "elapsed_time": "0:14:05", "remaining_time": "0:21:08", "throughput": "615.24", "total_tokens": 520480}
+{"current_steps": 77, "total_steps": 190, "loss": 0.0327, "learning_rate": 3.476827821223184e-06, "epoch": 1.9807073954983923, "percentage": 40.53, "elapsed_time": "0:14:17", "remaining_time": "0:20:57", "throughput": "614.95", "total_tokens": 527056}
+{"current_steps": 78, "total_steps": 190, "loss": 0.0448, "learning_rate": 3.436516483539781e-06, "epoch": 2.0064308681672025, "percentage": 41.05, "elapsed_time": "0:14:28", "remaining_time": "0:20:46", "throughput": "615.21", "total_tokens": 534112}
+{"current_steps": 79, "total_steps": 190, "loss": 0.0186, "learning_rate": 3.39591987386325e-06, "epoch": 2.032154340836013, "percentage": 41.58, "elapsed_time": "0:14:39", "remaining_time": "0:20:35", "throughput": "615.29", "total_tokens": 541024}
+{"current_steps": 80, "total_steps": 190, "loss": 0.0342, "learning_rate": 3.3550503583141726e-06, "epoch": 2.057877813504823, "percentage": 42.11, "elapsed_time": "0:14:50", "remaining_time": "0:20:24", "throughput": "615.30", "total_tokens": 547888}
+{"current_steps": 81, "total_steps": 190, "loss": 0.0079, "learning_rate": 3.313920386142892e-06, "epoch": 2.0836012861736335, "percentage": 42.63, "elapsed_time": "0:15:01", "remaining_time": "0:20:13", "throughput": "615.14", "total_tokens": 554592}
+{"current_steps": 82, "total_steps": 190, "loss": 0.0177, "learning_rate": 3.272542485937369e-06, "epoch": 2.1093247588424435, "percentage": 43.16, "elapsed_time": "0:15:12", "remaining_time": "0:20:02", "throughput": "615.01", "total_tokens": 561296}
+{"current_steps": 83, "total_steps": 190, "loss": 0.0139, "learning_rate": 3.230929261806842e-06, "epoch": 2.135048231511254, "percentage": 43.68, "elapsed_time": "0:15:23", "remaining_time": "0:19:50", "throughput": "614.74", "total_tokens": 567872}
+{"current_steps": 84, "total_steps": 190, "loss": 0.0103, "learning_rate": 3.189093389542498e-06, "epoch": 2.1607717041800645, "percentage": 44.21, "elapsed_time": "0:15:34", "remaining_time": "0:19:39", "throughput": "615.15", "total_tokens": 575072}
+{"current_steps": 85, "total_steps": 190, "loss": 0.0221, "learning_rate": 3.147047612756302e-06, "epoch": 2.1864951768488745, "percentage": 44.74, "elapsed_time": "0:15:45", "remaining_time": "0:19:28", "throughput": "615.35", "total_tokens": 582080}
+{"current_steps": 86, "total_steps": 190, "loss": 0.0021, "learning_rate": 3.1048047389991693e-06, "epoch": 2.212218649517685, "percentage": 45.26, "elapsed_time": "0:15:57", "remaining_time": "0:19:17", "throughput": "615.26", "total_tokens": 588816}
+{"current_steps": 87, "total_steps": 190, "loss": 0.011, "learning_rate": 3.062377635859663e-06, "epoch": 2.237942122186495, "percentage": 45.79, "elapsed_time": "0:16:08", "remaining_time": "0:19:06", "throughput": "615.65", "total_tokens": 596032}
+{"current_steps": 88, "total_steps": 190, "loss": 0.0081, "learning_rate": 3.019779227044398e-06, "epoch": 2.2636655948553055, "percentage": 46.32, "elapsed_time": "0:16:19", "remaining_time": "0:18:55", "throughput": "615.45", "total_tokens": 602672}
+{"current_steps": 89, "total_steps": 190, "loss": 0.0149, "learning_rate": 2.9770224884413625e-06, "epoch": 2.289389067524116, "percentage": 46.84, "elapsed_time": "0:16:30", "remaining_time": "0:18:43", "throughput": "615.35", "total_tokens": 609424}
+{"current_steps": 90, "total_steps": 190, "loss": 0.001, "learning_rate": 2.9341204441673267e-06, "epoch": 2.315112540192926, "percentage": 47.37, "elapsed_time": "0:16:41", "remaining_time": "0:18:32", "throughput": "615.53", "total_tokens": 616448}
+{"current_steps": 91, "total_steps": 190, "loss": 0.007, "learning_rate": 2.8910861626005774e-06, "epoch": 2.3408360128617365, "percentage": 47.89, "elapsed_time": "0:16:52", "remaining_time": "0:18:21", "throughput": "615.54", "total_tokens": 623296}
+{"current_steps": 92, "total_steps": 190, "loss": 0.0089, "learning_rate": 2.847932752400164e-06, "epoch": 2.3665594855305465, "percentage": 48.42, "elapsed_time": "0:17:03", "remaining_time": "0:18:10", "throughput": "615.48", "total_tokens": 630064}
+{"current_steps": 93, "total_steps": 190, "loss": 0.0013, "learning_rate": 2.804673358512869e-06, "epoch": 2.392282958199357, "percentage": 48.95, "elapsed_time": "0:17:14", "remaining_time": "0:17:59", "throughput": "615.67", "total_tokens": 637088}
+{"current_steps": 94, "total_steps": 190, "loss": 0.0267, "learning_rate": 2.761321158169134e-06, "epoch": 2.418006430868167, "percentage": 49.47, "elapsed_time": "0:17:25", "remaining_time": "0:17:48", "throughput": "615.82", "total_tokens": 644080}
+{"current_steps": 95, "total_steps": 190, "loss": 0.0171, "learning_rate": 2.717889356869146e-06, "epoch": 2.4437299035369775, "percentage": 50.0, "elapsed_time": "0:17:36", "remaining_time": "0:17:36", "throughput": "615.76", "total_tokens": 650848}
+{"current_steps": 96, "total_steps": 190, "loss": 0.0375, "learning_rate": 2.6743911843603134e-06, "epoch": 2.469453376205788, "percentage": 50.53, "elapsed_time": "0:17:48", "remaining_time": "0:17:25", "throughput": "615.50", "total_tokens": 657424}
+{"current_steps": 97, "total_steps": 190, "loss": 0.0101, "learning_rate": 2.6308398906073603e-06, "epoch": 2.495176848874598, "percentage": 51.05, "elapsed_time": "0:17:59", "remaining_time": "0:17:14", "throughput": "615.37", "total_tokens": 664128}
+{"current_steps": 98, "total_steps": 190, "loss": 0.0282, "learning_rate": 2.587248741756253e-06, "epoch": 2.5209003215434085, "percentage": 51.58, "elapsed_time": "0:18:10", "remaining_time": "0:17:03", "throughput": "615.50", "total_tokens": 671120}
+{"current_steps": 99, "total_steps": 190, "loss": 0.0069, "learning_rate": 2.543631016093209e-06, "epoch": 2.5466237942122185, "percentage": 52.11, "elapsed_time": "0:18:21", "remaining_time": "0:16:52", "throughput": "615.47", "total_tokens": 677920}
+{"current_steps": 100, "total_steps": 190, "loss": 0.0135, "learning_rate": 2.5e-06, "epoch": 2.572347266881029, "percentage": 52.63, "elapsed_time": "0:18:32", "remaining_time": "0:16:41", "throughput": "615.66", "total_tokens": 684960}
+{"current_steps": 101, "total_steps": 190, "loss": 0.0062, "learning_rate": 2.4563689839067913e-06, "epoch": 2.598070739549839, "percentage": 53.16, "elapsed_time": "0:18:43", "remaining_time": "0:16:30", "throughput": "615.71", "total_tokens": 691856}
+{"current_steps": 102, "total_steps": 190, "loss": 0.005, "learning_rate": 2.4127512582437486e-06, "epoch": 2.6237942122186495, "percentage": 53.68, "elapsed_time": "0:18:54", "remaining_time": "0:16:19", "throughput": "615.56", "total_tokens": 698512}
+{"current_steps": 103, "total_steps": 190, "loss": 0.0285, "learning_rate": 2.3691601093926406e-06, "epoch": 2.64951768488746, "percentage": 54.21, "elapsed_time": "0:19:05", "remaining_time": "0:16:07", "throughput": "615.65", "total_tokens": 705440}
+{"current_steps": 104, "total_steps": 190, "loss": 0.0225, "learning_rate": 2.325608815639687e-06, "epoch": 2.67524115755627, "percentage": 54.74, "elapsed_time": "0:19:16", "remaining_time": "0:15:56", "throughput": "615.86", "total_tokens": 712528}
+{"current_steps": 105, "total_steps": 190, "loss": 0.028, "learning_rate": 2.2821106431308546e-06, "epoch": 2.7009646302250805, "percentage": 55.26, "elapsed_time": "0:19:28", "remaining_time": "0:15:45", "throughput": "615.69", "total_tokens": 719168}
+{"current_steps": 106, "total_steps": 190, "loss": 0.0176, "learning_rate": 2.238678841830867e-06, "epoch": 2.7266881028938905, "percentage": 55.79, "elapsed_time": "0:19:39", "remaining_time": "0:15:34", "throughput": "615.60", "total_tokens": 725904}
+{"current_steps": 107, "total_steps": 190, "loss": 0.0047, "learning_rate": 2.195326641487132e-06, "epoch": 2.752411575562701, "percentage": 56.32, "elapsed_time": "0:19:50", "remaining_time": "0:15:23", "throughput": "615.36", "total_tokens": 732480}
+{"current_steps": 108, "total_steps": 190, "loss": 0.0135, "learning_rate": 2.1520672475998374e-06, "epoch": 2.778135048231511, "percentage": 56.84, "elapsed_time": "0:20:01", "remaining_time": "0:15:12", "throughput": "615.25", "total_tokens": 739184}
+{"current_steps": 109, "total_steps": 190, "loss": 0.0044, "learning_rate": 2.1089138373994226e-06, "epoch": 2.8038585209003215, "percentage": 57.37, "elapsed_time": "0:20:12", "remaining_time": "0:15:01", "throughput": "615.51", "total_tokens": 746320}
+{"current_steps": 110, "total_steps": 190, "loss": 0.0252, "learning_rate": 2.0658795558326745e-06, "epoch": 2.829581993569132, "percentage": 57.89, "elapsed_time": "0:20:23", "remaining_time": "0:14:49", "throughput": "615.50", "total_tokens": 753136}
+{"current_steps": 111, "total_steps": 190, "loss": 0.0249, "learning_rate": 2.022977511558638e-06, "epoch": 2.855305466237942, "percentage": 58.42, "elapsed_time": "0:20:34", "remaining_time": "0:14:38", "throughput": "615.61", "total_tokens": 760096}
+{"current_steps": 112, "total_steps": 190, "loss": 0.0146, "learning_rate": 1.9802207729556023e-06, "epoch": 2.8810289389067525, "percentage": 58.95, "elapsed_time": "0:20:45", "remaining_time": "0:14:27", "throughput": "615.75", "total_tokens": 767104}
+{"current_steps": 113, "total_steps": 190, "loss": 0.0044, "learning_rate": 1.937622364140338e-06, "epoch": 2.906752411575563, "percentage": 59.47, "elapsed_time": "0:20:56", "remaining_time": "0:14:16", "throughput": "615.69", "total_tokens": 773872}
+{"current_steps": 114, "total_steps": 190, "loss": 0.0054, "learning_rate": 1.895195261000831e-06, "epoch": 2.932475884244373, "percentage": 60.0, "elapsed_time": "0:21:08", "remaining_time": "0:14:05", "throughput": "615.71", "total_tokens": 780736}
+{"current_steps": 115, "total_steps": 190, "loss": 0.0106, "learning_rate": 1.852952387243698e-06, "epoch": 2.958199356913183, "percentage": 60.53, "elapsed_time": "0:21:19", "remaining_time": "0:13:54", "throughput": "615.58", "total_tokens": 787424}
+{"current_steps": 116, "total_steps": 190, "loss": 0.0167, "learning_rate": 1.8109066104575023e-06, "epoch": 2.9839228295819935, "percentage": 61.05, "elapsed_time": "0:21:30", "remaining_time": "0:13:43", "throughput": "615.53", "total_tokens": 794224}
+{"current_steps": 117, "total_steps": 190, "loss": 0.009, "learning_rate": 1.7690707381931585e-06, "epoch": 3.009646302250804, "percentage": 61.58, "elapsed_time": "0:21:41", "remaining_time": "0:13:31", "throughput": "615.55", "total_tokens": 801088}
+{"current_steps": 118, "total_steps": 190, "loss": 0.0024, "learning_rate": 1.7274575140626318e-06, "epoch": 3.035369774919614, "percentage": 62.11, "elapsed_time": "0:21:52", "remaining_time": "0:13:20", "throughput": "615.66", "total_tokens": 808048}
+{"current_steps": 119, "total_steps": 190, "loss": 0.0235, "learning_rate": 1.686079613857109e-06, "epoch": 3.0610932475884245, "percentage": 62.63, "elapsed_time": "0:22:03", "remaining_time": "0:13:09", "throughput": "615.60", "total_tokens": 814800}
+{"current_steps": 120, "total_steps": 190, "loss": 0.0179, "learning_rate": 1.6449496416858285e-06, "epoch": 3.0868167202572345, "percentage": 63.16, "elapsed_time": "0:22:14", "remaining_time": "0:12:58", "throughput": "615.53", "total_tokens": 821536}
+{"current_steps": 121, "total_steps": 190, "loss": 0.0059, "learning_rate": 1.6040801261367494e-06, "epoch": 3.112540192926045, "percentage": 63.68, "elapsed_time": "0:22:25", "remaining_time": "0:12:47", "throughput": "615.35", "total_tokens": 828128}
+{"current_steps": 122, "total_steps": 190, "loss": 0.0017, "learning_rate": 1.56348351646022e-06, "epoch": 3.1382636655948555, "percentage": 64.21, "elapsed_time": "0:22:36", "remaining_time": "0:12:36", "throughput": "615.08", "total_tokens": 834608}
+{"current_steps": 123, "total_steps": 190, "loss": 0.0018, "learning_rate": 1.5231721787768162e-06, "epoch": 3.1639871382636655, "percentage": 64.74, "elapsed_time": "0:22:48", "remaining_time": "0:12:25", "throughput": "615.02", "total_tokens": 841360}
+{"current_steps": 124, "total_steps": 190, "loss": 0.0032, "learning_rate": 1.4831583923105e-06, "epoch": 3.189710610932476, "percentage": 65.26, "elapsed_time": "0:22:59", "remaining_time": "0:12:14", "throughput": "615.23", "total_tokens": 848496}
+{"current_steps": 125, "total_steps": 190, "loss": 0.0019, "learning_rate": 1.443454345648252e-06, "epoch": 3.215434083601286, "percentage": 65.79, "elapsed_time": "0:23:10", "remaining_time": "0:12:02", "throughput": "615.29", "total_tokens": 855424}
+{"current_steps": 126, "total_steps": 190, "loss": 0.0014, "learning_rate": 1.4040721330273063e-06, "epoch": 3.2411575562700965, "percentage": 66.32, "elapsed_time": "0:23:21", "remaining_time": "0:11:51", "throughput": "615.27", "total_tokens": 862224}
+{"current_steps": 127, "total_steps": 190, "loss": 0.0052, "learning_rate": 1.3650237506511333e-06, "epoch": 3.266881028938907, "percentage": 66.84, "elapsed_time": "0:23:32", "remaining_time": "0:11:40", "throughput": "615.45", "total_tokens": 869312}
+{"current_steps": 128, "total_steps": 190, "loss": 0.0005, "learning_rate": 1.3263210930352737e-06, "epoch": 3.292604501607717, "percentage": 67.37, "elapsed_time": "0:23:43", "remaining_time": "0:11:29", "throughput": "615.55", "total_tokens": 876272}
+{"current_steps": 129, "total_steps": 190, "loss": 0.0131, "learning_rate": 1.2879759493841577e-06, "epoch": 3.3183279742765275, "percentage": 67.89, "elapsed_time": "0:23:54", "remaining_time": "0:11:18", "throughput": "615.52", "total_tokens": 883072}
+{"current_steps": 130, "total_steps": 190, "loss": 0.0009, "learning_rate": 1.2500000000000007e-06, "epoch": 3.3440514469453375, "percentage": 68.42, "elapsed_time": "0:24:05", "remaining_time": "0:11:07", "throughput": "615.54", "total_tokens": 889920}
+{"current_steps": 131, "total_steps": 190, "loss": 0.0057, "learning_rate": 1.2124048127248644e-06, "epoch": 3.369774919614148, "percentage": 68.95, "elapsed_time": "0:24:16", "remaining_time": "0:10:56", "throughput": "615.63", "total_tokens": 896896}
+{"current_steps": 132, "total_steps": 190, "loss": 0.0002, "learning_rate": 1.1752018394169882e-06, "epoch": 3.395498392282958, "percentage": 69.47, "elapsed_time": "0:24:27", "remaining_time": "0:10:45", "throughput": "615.53", "total_tokens": 903600}
+{"current_steps": 133, "total_steps": 190, "loss": 0.0002, "learning_rate": 1.1384024124624324e-06, "epoch": 3.4212218649517685, "percentage": 70.0, "elapsed_time": "0:24:39", "remaining_time": "0:10:33", "throughput": "615.37", "total_tokens": 910208}
+{"current_steps": 134, "total_steps": 190, "loss": 0.0145, "learning_rate": 1.1020177413231334e-06, "epoch": 3.446945337620579, "percentage": 70.53, "elapsed_time": "0:24:50", "remaining_time": "0:10:22", "throughput": "615.56", "total_tokens": 917328}
+{"current_steps": 135, "total_steps": 190, "loss": 0.0034, "learning_rate": 1.0660589091223854e-06, "epoch": 3.472668810289389, "percentage": 71.05, "elapsed_time": "0:25:01", "remaining_time": "0:10:11", "throughput": "615.59", "total_tokens": 924192}
+{"current_steps": 136, "total_steps": 190, "loss": 0.0156, "learning_rate": 1.0305368692688175e-06, "epoch": 3.4983922829581995, "percentage": 71.58, "elapsed_time": "0:25:12", "remaining_time": "0:10:00", "throughput": "615.43", "total_tokens": 930784}
+{"current_steps": 137, "total_steps": 190, "loss": 0.0013, "learning_rate": 9.95462442119879e-07, "epoch": 3.5241157556270095, "percentage": 72.11, "elapsed_time": "0:25:23", "remaining_time": "0:09:49", "throughput": "615.59", "total_tokens": 937856}
+{"current_steps": 138, "total_steps": 190, "loss": 0.0007, "learning_rate": 9.608463116858544e-07, "epoch": 3.54983922829582, "percentage": 72.63, "elapsed_time": "0:25:34", "remaining_time": "0:09:38", "throughput": "615.56", "total_tokens": 944640}
+{"current_steps": 139, "total_steps": 190, "loss": 0.0005, "learning_rate": 9.266990223754069e-07, "epoch": 3.57556270096463, "percentage": 73.16, "elapsed_time": "0:25:45", "remaining_time": "0:09:27", "throughput": "615.58", "total_tokens": 951504}
+{"current_steps": 140, "total_steps": 190, "loss": 0.0034, "learning_rate": 8.930309757836517e-07, "epoch": 3.6012861736334405, "percentage": 73.68, "elapsed_time": "0:25:56", "remaining_time": "0:09:16", "throughput": "615.52", "total_tokens": 958240}
+{"current_steps": 141, "total_steps": 190, "loss": 0.0001, "learning_rate": 8.598524275237321e-07, "epoch": 3.627009646302251, "percentage": 74.21, "elapsed_time": "0:26:07", "remaining_time": "0:09:04", "throughput": "615.41", "total_tokens": 964912}
+{"current_steps": 142, "total_steps": 190, "loss": 0.001, "learning_rate": 8.271734841028553e-07, "epoch": 3.652733118971061, "percentage": 74.74, "elapsed_time": "0:26:19", "remaining_time": "0:08:53", "throughput": "615.49", "total_tokens": 971872}
+{"current_steps": 143, "total_steps": 190, "loss": 0.0123, "learning_rate": 7.950040998437541e-07, "epoch": 3.6784565916398715, "percentage": 75.26, "elapsed_time": "0:26:30", "remaining_time": "0:08:42", "throughput": "615.44", "total_tokens": 978640}
+{"current_steps": 144, "total_steps": 190, "loss": 0.0002, "learning_rate": 7.633540738525066e-07, "epoch": 3.7041800643086815, "percentage": 75.79, "elapsed_time": "0:26:41", "remaining_time": "0:08:31", "throughput": "615.35", "total_tokens": 985328}
+{"current_steps": 145, "total_steps": 190, "loss": 0.011, "learning_rate": 7.322330470336314e-07, "epoch": 3.729903536977492, "percentage": 76.32, "elapsed_time": "0:26:52", "remaining_time": "0:08:20", "throughput": "615.39", "total_tokens": 992224}
+{"current_steps": 146, "total_steps": 190, "loss": 0.0008, "learning_rate": 7.016504991533727e-07, "epoch": 3.755627009646302, "percentage": 76.84, "elapsed_time": "0:27:03", "remaining_time": "0:08:09", "throughput": "615.17", "total_tokens": 998688}
+{"current_steps": 147, "total_steps": 190, "loss": 0.0003, "learning_rate": 6.716157459520739e-07, "epoch": 3.7813504823151125, "percentage": 77.37, "elapsed_time": "0:27:14", "remaining_time": "0:07:58", "throughput": "615.50", "total_tokens": 1006032}
+{"current_steps": 148, "total_steps": 190, "loss": 0.0018, "learning_rate": 6.421379363065142e-07, "epoch": 3.807073954983923, "percentage": 77.89, "elapsed_time": "0:27:25", "remaining_time": "0:07:46", "throughput": "615.55", "total_tokens": 1012944}
+{"current_steps": 149, "total_steps": 190, "loss": 0.0016, "learning_rate": 6.1322604944307e-07, "epoch": 3.832797427652733, "percentage": 78.42, "elapsed_time": "0:27:36", "remaining_time": "0:07:35", "throughput": "615.59", "total_tokens": 1019856}
+{"current_steps": 150, "total_steps": 190, "loss": 0.0021, "learning_rate": 5.848888922025553e-07, "epoch": 3.8585209003215435, "percentage": 78.95, "elapsed_time": "0:27:47", "remaining_time": "0:07:24", "throughput": "615.56", "total_tokens": 1026656}
+{"current_steps": 151, "total_steps": 190, "loss": 0.0001, "learning_rate": 5.571350963575728e-07, "epoch": 3.884244372990354, "percentage": 79.47, "elapsed_time": "0:27:58", "remaining_time": "0:07:13", "throughput": "615.68", "total_tokens": 1033696}
+{"current_steps": 152, "total_steps": 190, "loss": 0.0001, "learning_rate": 5.299731159831953e-07, "epoch": 3.909967845659164, "percentage": 80.0, "elapsed_time": "0:28:10", "remaining_time": "0:07:02", "throughput": "615.50", "total_tokens": 1040240}
+{"current_steps": 153, "total_steps": 190, "loss": 0.0003, "learning_rate": 5.034112248817685e-07, "epoch": 3.935691318327974, "percentage": 80.53, "elapsed_time": "0:28:21", "remaining_time": "0:06:51", "throughput": "615.45", "total_tokens": 1046992}
+{"current_steps": 154, "total_steps": 190, "loss": 0.0002, "learning_rate": 4.774575140626317e-07, "epoch": 3.9614147909967845, "percentage": 81.05, "elapsed_time": "0:28:32", "remaining_time": "0:06:40", "throughput": "615.52", "total_tokens": 1053936}
+{"current_steps": 155, "total_steps": 190, "loss": 0.0002, "learning_rate": 4.5211988927752026e-07, "epoch": 3.987138263665595, "percentage": 81.58, "elapsed_time": "0:28:43", "remaining_time": "0:06:29", "throughput": "615.41", "total_tokens": 1060576}
+{"current_steps": 156, "total_steps": 190, "loss": 0.0, "learning_rate": 4.27406068612396e-07, "epoch": 4.012861736334405, "percentage": 82.11, "elapsed_time": "0:28:54", "remaining_time": "0:06:18", "throughput": "615.56", "total_tokens": 1067648}
+{"current_steps": 157, "total_steps": 190, "loss": 0.0003, "learning_rate": 4.033235801364402e-07, "epoch": 4.038585209003215, "percentage": 82.63, "elapsed_time": "0:29:05", "remaining_time": "0:06:06", "throughput": "615.58", "total_tokens": 1074512}
+{"current_steps": 158, "total_steps": 190, "loss": 0.0002, "learning_rate": 3.798797596089351e-07, "epoch": 4.064308681672026, "percentage": 83.16, "elapsed_time": "0:29:16", "remaining_time": "0:05:55", "throughput": "615.55", "total_tokens": 1081296}
+{"current_steps": 159, "total_steps": 190, "loss": 0.0001, "learning_rate": 3.5708174824471947e-07, "epoch": 4.090032154340836, "percentage": 83.68, "elapsed_time": "0:29:27", "remaining_time": "0:05:44", "throughput": "615.40", "total_tokens": 1087888}
+{"current_steps": 160, "total_steps": 190, "loss": 0.0001, "learning_rate": 3.3493649053890325e-07, "epoch": 4.115755627009646, "percentage": 84.21, "elapsed_time": "0:29:38", "remaining_time": "0:05:33", "throughput": "615.53", "total_tokens": 1094960}
+{"current_steps": 161, "total_steps": 190, "loss": 0.0001, "learning_rate": 3.134507321515107e-07, "epoch": 4.141479099678457, "percentage": 84.74, "elapsed_time": "0:29:50", "remaining_time": "0:05:22", "throughput": "615.52", "total_tokens": 1101776}
+{"current_steps": 162, "total_steps": 190, "loss": 0.0, "learning_rate": 2.9263101785268253e-07, "epoch": 4.167202572347267, "percentage": 85.26, "elapsed_time": "0:30:01", "remaining_time": "0:05:11", "throughput": "615.59", "total_tokens": 1108736}
+{"current_steps": 163, "total_steps": 190, "loss": 0.0001, "learning_rate": 2.7248368952908055e-07, "epoch": 4.192926045016077, "percentage": 85.79, "elapsed_time": "0:30:12", "remaining_time": "0:05:00", "throughput": "615.69", "total_tokens": 1115744}
+{"current_steps": 164, "total_steps": 190, "loss": 0.0, "learning_rate": 2.53014884252083e-07, "epoch": 4.218649517684887, "percentage": 86.32, "elapsed_time": "0:30:23", "remaining_time": "0:04:49", "throughput": "615.71", "total_tokens": 1122592}
+{"current_steps": 165, "total_steps": 190, "loss": 0.0001, "learning_rate": 2.3423053240837518e-07, "epoch": 4.244372990353698, "percentage": 86.84, "elapsed_time": "0:30:34", "remaining_time": "0:04:37", "throughput": "615.55", "total_tokens": 1129136}
+{"current_steps": 166, "total_steps": 190, "loss": 0.0001, "learning_rate": 2.1613635589349756e-07, "epoch": 4.270096463022508, "percentage": 87.37, "elapsed_time": "0:30:45", "remaining_time": "0:04:26", "throughput": "615.61", "total_tokens": 1136080}
+{"current_steps": 167, "total_steps": 190, "loss": 0.0002, "learning_rate": 1.9873786636889908e-07, "epoch": 4.295819935691318, "percentage": 87.89, "elapsed_time": "0:30:56", "remaining_time": "0:04:15", "throughput": "615.64", "total_tokens": 1142976}
+{"current_steps": 168, "total_steps": 190, "loss": 0.0002, "learning_rate": 1.8204036358303173e-07, "epoch": 4.321543408360129, "percentage": 88.42, "elapsed_time": "0:31:07", "remaining_time": "0:04:04", "throughput": "615.58", "total_tokens": 1149712}
+{"current_steps": 169, "total_steps": 190, "loss": 0.0001, "learning_rate": 1.6604893375699594e-07, "epoch": 4.347266881028939, "percentage": 88.95, "elapsed_time": "0:31:18", "remaining_time": "0:03:53", "throughput": "615.46", "total_tokens": 1156336}
+{"current_steps": 170, "total_steps": 190, "loss": 0.0001, "learning_rate": 1.507684480352292e-07, "epoch": 4.372990353697749, "percentage": 89.47, "elapsed_time": "0:31:29", "remaining_time": "0:03:42", "throughput": "615.43", "total_tokens": 1163120}
+{"current_steps": 171, "total_steps": 190, "loss": 0.0115, "learning_rate": 1.362035610017079e-07, "epoch": 4.39871382636656, "percentage": 90.0, "elapsed_time": "0:31:41", "remaining_time": "0:03:31", "throughput": "615.48", "total_tokens": 1170032}
+{"current_steps": 172, "total_steps": 190, "loss": 0.0005, "learning_rate": 1.223587092621162e-07, "epoch": 4.42443729903537, "percentage": 90.53, "elapsed_time": "0:31:52", "remaining_time": "0:03:20", "throughput": "615.47", "total_tokens": 1176832}
+{"current_steps": 173, "total_steps": 190, "loss": 0.0003, "learning_rate": 1.0923811009241142e-07, "epoch": 4.45016077170418, "percentage": 91.05, "elapsed_time": "0:32:03", "remaining_time": "0:03:08", "throughput": "615.57", "total_tokens": 1183856}
+{"current_steps": 174, "total_steps": 190, "loss": 0.005, "learning_rate": 9.684576015420277e-08, "epoch": 4.47588424437299, "percentage": 91.58, "elapsed_time": "0:32:14", "remaining_time": "0:02:57", "throughput": "615.49", "total_tokens": 1190544}
+{"current_steps": 175, "total_steps": 190, "loss": 0.0003, "learning_rate": 8.518543427732951e-08, "epoch": 4.501607717041801, "percentage": 92.11, "elapsed_time": "0:32:25", "remaining_time": "0:02:46", "throughput": "615.38", "total_tokens": 1197168}
+{"current_steps": 176, "total_steps": 190, "loss": 0.0008, "learning_rate": 7.426068431000883e-08, "epoch": 4.527331189710611, "percentage": 92.63, "elapsed_time": "0:32:36", "remaining_time": "0:02:35", "throughput": "615.27", "total_tokens": 1203776}
+{"current_steps": 177, "total_steps": 190, "loss": 0.0, "learning_rate": 6.407483803691216e-08, "epoch": 4.553054662379421, "percentage": 93.16, "elapsed_time": "0:32:47", "remaining_time": "0:02:24", "throughput": "615.37", "total_tokens": 1210816}
+{"current_steps": 178, "total_steps": 190, "loss": 0.0, "learning_rate": 5.463099816548578e-08, "epoch": 4.578778135048232, "percentage": 93.68, "elapsed_time": "0:32:58", "remaining_time": "0:02:13", "throughput": "615.39", "total_tokens": 1217696}
+{"current_steps": 179, "total_steps": 190, "loss": 0.0042, "learning_rate": 4.593204138084006e-08, "epoch": 4.604501607717042, "percentage": 94.21, "elapsed_time": "0:33:09", "remaining_time": "0:02:02", "throughput": "615.48", "total_tokens": 1224704}
+{"current_steps": 180, "total_steps": 190, "loss": 0.0004, "learning_rate": 3.798061746947995e-08, "epoch": 4.630225080385852, "percentage": 94.74, "elapsed_time": "0:33:20", "remaining_time": "0:01:51", "throughput": "615.54", "total_tokens": 1231664}
+{"current_steps": 181, "total_steps": 190, "loss": 0.0001, "learning_rate": 3.077914851215585e-08, "epoch": 4.655948553054662, "percentage": 95.26, "elapsed_time": "0:33:32", "remaining_time": "0:01:40", "throughput": "615.42", "total_tokens": 1238240}
+{"current_steps": 182, "total_steps": 190, "loss": 0.0001, "learning_rate": 2.4329828146074096e-08, "epoch": 4.681672025723473, "percentage": 95.79, "elapsed_time": "0:33:43", "remaining_time": "0:01:28", "throughput": "615.30", "total_tokens": 1244832}
+{"current_steps": 183, "total_steps": 190, "loss": 0.0004, "learning_rate": 1.8634620896695044e-08, "epoch": 4.707395498392283, "percentage": 96.32, "elapsed_time": "0:33:54", "remaining_time": "0:01:17", "throughput": "615.12", "total_tokens": 1251296}
+{"current_steps": 184, "total_steps": 190, "loss": 0.0, "learning_rate": 1.3695261579316776e-08, "epoch": 4.733118971061093, "percentage": 96.84, "elapsed_time": "0:34:05", "remaining_time": "0:01:06", "throughput": "615.05", "total_tokens": 1257968}
+{"current_steps": 185, "total_steps": 190, "loss": 0.0009, "learning_rate": 9.513254770636138e-09, "epoch": 4.758842443729904, "percentage": 97.37, "elapsed_time": "0:34:16", "remaining_time": "0:00:55", "throughput": "615.11", "total_tokens": 1264928}
+{"current_steps": 186, "total_steps": 190, "loss": 0.0001, "learning_rate": 6.089874350439507e-09, "epoch": 4.784565916398714, "percentage": 97.89, "elapsed_time": "0:34:27", "remaining_time": "0:00:44", "throughput": "615.12", "total_tokens": 1271792}
+{"current_steps": 187, "total_steps": 190, "loss": 0.0004, "learning_rate": 3.4261631135654174e-09, "epoch": 4.810289389067524, "percentage": 98.42, "elapsed_time": "0:34:38", "remaining_time": "0:00:33", "throughput": "615.30", "total_tokens": 1279024}
+{"current_steps": 188, "total_steps": 190, "loss": 0.0002, "learning_rate": 1.5229324522605949e-09, "epoch": 4.836012861736334, "percentage": 98.95, "elapsed_time": "0:34:49", "remaining_time": "0:00:22", "throughput": "615.28", "total_tokens": 1285792}
+{"current_steps": 189, "total_steps": 190, "loss": 0.0, "learning_rate": 3.8076210902182607e-10, "epoch": 4.861736334405145, "percentage": 99.47, "elapsed_time": "0:35:00", "remaining_time": "0:00:11", "throughput": "615.26", "total_tokens": 1292576}
+{"current_steps": 190, "total_steps": 190, "loss": 0.0001, "learning_rate": 0.0, "epoch": 4.887459807073955, "percentage": 100.0, "elapsed_time": "0:35:11", "remaining_time": "0:00:00", "throughput": "615.25", "total_tokens": 1299392}
+{"current_steps": 190, "total_steps": 190, "epoch": 4.887459807073955, "percentage": 100.0, "elapsed_time": "0:36:02", "remaining_time": "0:00:00", "throughput": "600.99", "total_tokens": 1299392}

trainer_state.json

@@ -0,0 +1,1563 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 4.887459807073955,
+  "eval_steps": 500,
+  "global_step": 190,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.02572347266881029,
+      "grad_norm": 268.0186767578125,
+      "learning_rate": 5.000000000000001e-07,
+      "loss": 8.3599,
+      "num_input_tokens_seen": 7120,
+      "step": 1
+    },
+    {
+      "epoch": 0.05144694533762058,
+      "grad_norm": 277.9050598144531,
+      "learning_rate": 1.0000000000000002e-06,
+      "loss": 8.1891,
+      "num_input_tokens_seen": 13888,
+      "step": 2
+    },
+    {
+      "epoch": 0.07717041800643087,
+      "grad_norm": 277.69873046875,
+      "learning_rate": 1.5e-06,
+      "loss": 8.0792,
+      "num_input_tokens_seen": 20656,
+      "step": 3
+    },
+    {
+      "epoch": 0.10289389067524116,
+      "grad_norm": 267.486328125,
+      "learning_rate": 2.0000000000000003e-06,
+      "loss": 7.9682,
+      "num_input_tokens_seen": 27184,
+      "step": 4
+    },
+    {
+      "epoch": 0.12861736334405144,
+      "grad_norm": 301.225830078125,
+      "learning_rate": 2.5e-06,
+      "loss": 6.9482,
+      "num_input_tokens_seen": 34416,
+      "step": 5
+    },
+    {
+      "epoch": 0.15434083601286175,
+      "grad_norm": 137.74415588378906,
+      "learning_rate": 3e-06,
+      "loss": 5.1505,
+      "num_input_tokens_seen": 41056,
+      "step": 6
+    },
+    {
+      "epoch": 0.18006430868167203,
+      "grad_norm": 113.7622299194336,
+      "learning_rate": 3.5e-06,
+      "loss": 4.7491,
+      "num_input_tokens_seen": 47536,
+      "step": 7
+    },
+    {
+      "epoch": 0.2057877813504823,
+      "grad_norm": 109.39883422851562,
+      "learning_rate": 4.000000000000001e-06,
+      "loss": 3.2164,
+      "num_input_tokens_seen": 54464,
+      "step": 8
+    },
+    {
+      "epoch": 0.2315112540192926,
+      "grad_norm": 119.0561294555664,
+      "learning_rate": 4.5e-06,
+      "loss": 2.7761,
+      "num_input_tokens_seen": 61520,
+      "step": 9
+    },
+    {
+      "epoch": 0.2572347266881029,
+      "grad_norm": 111.31031036376953,
+      "learning_rate": 5e-06,
+      "loss": 0.6703,
+      "num_input_tokens_seen": 68464,
+      "step": 10
+    },
+    {
+      "epoch": 0.2829581993569132,
+      "grad_norm": 39.81684494018555,
+      "learning_rate": 4.9996192378909785e-06,
+      "loss": 0.3255,
+      "num_input_tokens_seen": 75152,
+      "step": 11
+    },
+    {
+      "epoch": 0.3086816720257235,
+      "grad_norm": 41.08443069458008,
+      "learning_rate": 4.99847706754774e-06,
+      "loss": 0.3301,
+      "num_input_tokens_seen": 81888,
+      "step": 12
+    },
+    {
+      "epoch": 0.33440514469453375,
+      "grad_norm": 17.01146125793457,
+      "learning_rate": 4.9965738368864345e-06,
+      "loss": 0.2121,
+      "num_input_tokens_seen": 88688,
+      "step": 13
+    },
+    {
+      "epoch": 0.36012861736334406,
+      "grad_norm": 73.99099731445312,
+      "learning_rate": 4.993910125649561e-06,
+      "loss": 1.1565,
+      "num_input_tokens_seen": 95616,
+      "step": 14
+    },
+    {
+      "epoch": 0.3858520900321543,
+      "grad_norm": 114.83367919921875,
+      "learning_rate": 4.990486745229364e-06,
+      "loss": 0.8054,
+      "num_input_tokens_seen": 102144,
+      "step": 15
+    },
+    {
+      "epoch": 0.4115755627009646,
+      "grad_norm": 18.247591018676758,
+      "learning_rate": 4.986304738420684e-06,
+      "loss": 0.2386,
+      "num_input_tokens_seen": 109120,
+      "step": 16
+    },
+    {
+      "epoch": 0.43729903536977494,
+      "grad_norm": 27.782032012939453,
+      "learning_rate": 4.981365379103306e-06,
+      "loss": 0.3161,
+      "num_input_tokens_seen": 115744,
+      "step": 17
+    },
+    {
+      "epoch": 0.4630225080385852,
+      "grad_norm": 21.16512107849121,
+      "learning_rate": 4.975670171853926e-06,
+      "loss": 0.2773,
+      "num_input_tokens_seen": 122704,
+      "step": 18
+    },
+    {
+      "epoch": 0.4887459807073955,
+      "grad_norm": 6.738564491271973,
+      "learning_rate": 4.9692208514878445e-06,
+      "loss": 0.2062,
+      "num_input_tokens_seen": 129552,
+      "step": 19
+    },
+    {
+      "epoch": 0.5144694533762058,
+      "grad_norm": 3.6603872776031494,
+      "learning_rate": 4.962019382530521e-06,
+      "loss": 0.1837,
+      "num_input_tokens_seen": 136544,
+      "step": 20
+    },
+    {
+      "epoch": 0.5401929260450161,
+      "grad_norm": 10.339789390563965,
+      "learning_rate": 4.9540679586191605e-06,
+      "loss": 0.1735,
+      "num_input_tokens_seen": 143488,
+      "step": 21
+    },
+    {
+      "epoch": 0.5659163987138264,
+      "grad_norm": 4.833702087402344,
+      "learning_rate": 4.9453690018345144e-06,
+      "loss": 0.1588,
+      "num_input_tokens_seen": 150224,
+      "step": 22
+    },
+    {
+      "epoch": 0.5916398713826366,
+      "grad_norm": 4.9161152839660645,
+      "learning_rate": 4.935925161963089e-06,
+      "loss": 0.1443,
+      "num_input_tokens_seen": 157232,
+      "step": 23
+    },
+    {
+      "epoch": 0.617363344051447,
+      "grad_norm": 9.033807754516602,
+      "learning_rate": 4.925739315689991e-06,
+      "loss": 0.157,
+      "num_input_tokens_seen": 163776,
+      "step": 24
+    },
+    {
+      "epoch": 0.6430868167202572,
+      "grad_norm": 4.518654823303223,
+      "learning_rate": 4.914814565722671e-06,
+      "loss": 0.1199,
+      "num_input_tokens_seen": 170352,
+      "step": 25
+    },
+    {
+      "epoch": 0.6688102893890675,
+      "grad_norm": 10.43909740447998,
+      "learning_rate": 4.903154239845798e-06,
+      "loss": 0.1539,
+      "num_input_tokens_seen": 177120,
+      "step": 26
+    },
+    {
+      "epoch": 0.6945337620578779,
+      "grad_norm": 9.000858306884766,
+      "learning_rate": 4.890761889907589e-06,
+      "loss": 0.1208,
+      "num_input_tokens_seen": 184096,
+      "step": 27
+    },
+    {
+      "epoch": 0.7202572347266881,
+      "grad_norm": 3.3685858249664307,
+      "learning_rate": 4.8776412907378845e-06,
+      "loss": 0.0954,
+      "num_input_tokens_seen": 191040,
+      "step": 28
+    },
+    {
+      "epoch": 0.7459807073954984,
+      "grad_norm": 6.876232147216797,
+      "learning_rate": 4.863796438998293e-06,
+      "loss": 0.1387,
+      "num_input_tokens_seen": 198064,
+      "step": 29
+    },
+    {
+      "epoch": 0.7717041800643086,
+      "grad_norm": 11.019433975219727,
+      "learning_rate": 4.849231551964771e-06,
+      "loss": 0.1484,
+      "num_input_tokens_seen": 205136,
+      "step": 30
+    },
+    {
+      "epoch": 0.797427652733119,
+      "grad_norm": 4.780524730682373,
+      "learning_rate": 4.833951066243004e-06,
+      "loss": 0.0998,
+      "num_input_tokens_seen": 212000,
+      "step": 31
+    },
+    {
+      "epoch": 0.8231511254019293,
+      "grad_norm": 3.613060235977173,
+      "learning_rate": 4.817959636416969e-06,
+      "loss": 0.1068,
+      "num_input_tokens_seen": 218720,
+      "step": 32
+    },
+    {
+      "epoch": 0.8488745980707395,
+      "grad_norm": 5.576949596405029,
+      "learning_rate": 4.801262133631101e-06,
+      "loss": 0.0801,
+      "num_input_tokens_seen": 225856,
+      "step": 33
+    },
+    {
+      "epoch": 0.8745980707395499,
+      "grad_norm": 3.3204915523529053,
+      "learning_rate": 4.783863644106502e-06,
+      "loss": 0.1066,
+      "num_input_tokens_seen": 232640,
+      "step": 34
+    },
+    {
+      "epoch": 0.9003215434083601,
+      "grad_norm": 3.090853452682495,
+      "learning_rate": 4.765769467591626e-06,
+      "loss": 0.1038,
+      "num_input_tokens_seen": 239504,
+      "step": 35
+    },
+    {
+      "epoch": 0.9260450160771704,
+      "grad_norm": 6.979610443115234,
+      "learning_rate": 4.746985115747918e-06,
+      "loss": 0.106,
+      "num_input_tokens_seen": 246288,
+      "step": 36
+    },
+    {
+      "epoch": 0.9517684887459807,
+      "grad_norm": 3.491868495941162,
+      "learning_rate": 4.72751631047092e-06,
+      "loss": 0.1107,
+      "num_input_tokens_seen": 253136,
+      "step": 37
+    },
+    {
+      "epoch": 0.977491961414791,
+      "grad_norm": 6.108020782470703,
+      "learning_rate": 4.707368982147318e-06,
+      "loss": 0.1372,
+      "num_input_tokens_seen": 260160,
+      "step": 38
+    },
+    {
+      "epoch": 1.0032154340836013,
+      "grad_norm": 1.9764081239700317,
+      "learning_rate": 4.68654926784849e-06,
+      "loss": 0.0816,
+      "num_input_tokens_seen": 267120,
+      "step": 39
+    },
+    {
+      "epoch": 1.0289389067524115,
+      "grad_norm": 2.5603554248809814,
+      "learning_rate": 4.665063509461098e-06,
+      "loss": 0.0743,
+      "num_input_tokens_seen": 274112,
+      "step": 40
+    },
+    {
+      "epoch": 1.0546623794212218,
+      "grad_norm": 5.457398414611816,
+      "learning_rate": 4.642918251755281e-06,
+      "loss": 0.072,
+      "num_input_tokens_seen": 281136,
+      "step": 41
+    },
+    {
+      "epoch": 1.0803858520900322,
+      "grad_norm": 4.488720417022705,
+      "learning_rate": 4.620120240391065e-06,
+      "loss": 0.0596,
+      "num_input_tokens_seen": 288048,
+      "step": 42
+    },
+    {
+      "epoch": 1.1061093247588425,
+      "grad_norm": 2.8057987689971924,
+      "learning_rate": 4.596676419863561e-06,
+      "loss": 0.0544,
+      "num_input_tokens_seen": 295120,
+      "step": 43
+    },
+    {
+      "epoch": 1.1318327974276527,
+      "grad_norm": 2.921278476715088,
+      "learning_rate": 4.572593931387604e-06,
+      "loss": 0.0342,
+      "num_input_tokens_seen": 302000,
+      "step": 44
+    },
+    {
+      "epoch": 1.157556270096463,
+      "grad_norm": 2.9281697273254395,
+      "learning_rate": 4.54788011072248e-06,
+      "loss": 0.0394,
+      "num_input_tokens_seen": 308672,
+      "step": 45
+    },
+    {
+      "epoch": 1.1832797427652733,
+      "grad_norm": 4.087917327880859,
+      "learning_rate": 4.522542485937369e-06,
+      "loss": 0.0196,
+      "num_input_tokens_seen": 315600,
+      "step": 46
+    },
+    {
+      "epoch": 1.2090032154340835,
+      "grad_norm": 3.390829086303711,
+      "learning_rate": 4.496588775118232e-06,
+      "loss": 0.0411,
+      "num_input_tokens_seen": 322464,
+      "step": 47
+    },
+    {
+      "epoch": 1.234726688102894,
+      "grad_norm": 4.550433158874512,
+      "learning_rate": 4.470026884016805e-06,
+      "loss": 0.0257,
+      "num_input_tokens_seen": 329024,
+      "step": 48
+    },
+    {
+      "epoch": 1.2604501607717042,
+      "grad_norm": 6.6677446365356445,
+      "learning_rate": 4.442864903642428e-06,
+      "loss": 0.0289,
+      "num_input_tokens_seen": 336032,
+      "step": 49
+    },
+    {
+      "epoch": 1.2861736334405145,
+      "grad_norm": 7.874734878540039,
+      "learning_rate": 4.415111107797445e-06,
+      "loss": 0.1193,
+      "num_input_tokens_seen": 342704,
+      "step": 50
+    },
+    {
+      "epoch": 1.3118971061093248,
+      "grad_norm": 5.971132278442383,
+      "learning_rate": 4.386773950556931e-06,
+      "loss": 0.0883,
+      "num_input_tokens_seen": 349472,
+      "step": 51
+    },
+    {
+      "epoch": 1.337620578778135,
+      "grad_norm": 4.0566725730896,
+      "learning_rate": 4.357862063693486e-06,
+      "loss": 0.0377,
+      "num_input_tokens_seen": 356272,
+      "step": 52
+    },
+    {
+      "epoch": 1.3633440514469453,
+      "grad_norm": 5.443573474884033,
+      "learning_rate": 4.328384254047927e-06,
+      "loss": 0.0602,
+      "num_input_tokens_seen": 363040,
+      "step": 53
+    },
+    {
+      "epoch": 1.3890675241157555,
+      "grad_norm": 6.038252353668213,
+      "learning_rate": 4.2983495008466285e-06,
+      "loss": 0.083,
+      "num_input_tokens_seen": 369664,
+      "step": 54
+    },
+    {
+      "epoch": 1.414790996784566,
+      "grad_norm": 2.8046696186065674,
+      "learning_rate": 4.267766952966369e-06,
+      "loss": 0.0358,
+      "num_input_tokens_seen": 376704,
+      "step": 55
+    },
+    {
+      "epoch": 1.4405144694533762,
+      "grad_norm": 2.451719284057617,
+      "learning_rate": 4.236645926147493e-06,
+      "loss": 0.0321,
+      "num_input_tokens_seen": 383600,
+      "step": 56
+    },
+    {
+      "epoch": 1.4662379421221865,
+      "grad_norm": 3.4111475944519043,
+      "learning_rate": 4.204995900156247e-06,
+      "loss": 0.0452,
+      "num_input_tokens_seen": 390592,
+      "step": 57
+    },
+    {
+      "epoch": 1.4919614147909968,
+      "grad_norm": 3.065139055252075,
+      "learning_rate": 4.172826515897146e-06,
+      "loss": 0.0915,
+      "num_input_tokens_seen": 397360,
+      "step": 58
+    },
+    {
+      "epoch": 1.517684887459807,
+      "grad_norm": 3.0692198276519775,
+      "learning_rate": 4.140147572476269e-06,
+      "loss": 0.0651,
+      "num_input_tokens_seen": 404048,
+      "step": 59
+    },
+    {
+      "epoch": 1.5434083601286175,
+      "grad_norm": 3.6201350688934326,
+      "learning_rate": 4.106969024216348e-06,
+      "loss": 0.0868,
+      "num_input_tokens_seen": 410960,
+      "step": 60
+    },
+    {
+      "epoch": 1.5691318327974275,
+      "grad_norm": 2.289900779724121,
+      "learning_rate": 4.073300977624594e-06,
+      "loss": 0.0554,
+      "num_input_tokens_seen": 417888,
+      "step": 61
+    },
+    {
+      "epoch": 1.594855305466238,
+      "grad_norm": 1.4743808507919312,
+      "learning_rate": 4.039153688314146e-06,
+      "loss": 0.0336,
+      "num_input_tokens_seen": 424880,
+      "step": 62
+    },
+    {
+      "epoch": 1.6205787781350482,
+      "grad_norm": 2.1891098022460938,
+      "learning_rate": 4.0045375578801216e-06,
+      "loss": 0.0455,
+      "num_input_tokens_seen": 432000,
+      "step": 63
+    },
+    {
+      "epoch": 1.6463022508038585,
+      "grad_norm": 1.8203191757202148,
+      "learning_rate": 3.969463130731183e-06,
+      "loss": 0.0406,
+      "num_input_tokens_seen": 438672,
+      "step": 64
+    },
+    {
+      "epoch": 1.6720257234726688,
+      "grad_norm": 2.453317165374756,
+      "learning_rate": 3.933941090877615e-06,
+      "loss": 0.0461,
+      "num_input_tokens_seen": 445440,
+      "step": 65
+    },
+    {
+      "epoch": 1.697749196141479,
+      "grad_norm": 2.035806894302368,
+      "learning_rate": 3.897982258676867e-06,
+      "loss": 0.0466,
+      "num_input_tokens_seen": 452064,
+      "step": 66
+    },
+    {
+      "epoch": 1.7234726688102895,
+      "grad_norm": 3.169175386428833,
+      "learning_rate": 3.861597587537568e-06,
+      "loss": 0.0382,
+      "num_input_tokens_seen": 458992,
+      "step": 67
+    },
+    {
+      "epoch": 1.7491961414790995,
+      "grad_norm": 2.0320234298706055,
+      "learning_rate": 3.824798160583012e-06,
+      "loss": 0.0426,
+      "num_input_tokens_seen": 465568,
+      "step": 68
+    },
+    {
+      "epoch": 1.77491961414791,
+      "grad_norm": 4.094840049743652,
+      "learning_rate": 3.787595187275136e-06,
+      "loss": 0.0264,
+      "num_input_tokens_seen": 472496,
+      "step": 69
+    },
+    {
+      "epoch": 1.8006430868167203,
+      "grad_norm": 4.245150089263916,
+      "learning_rate": 3.7500000000000005e-06,
+      "loss": 0.0567,
+      "num_input_tokens_seen": 479392,
+      "step": 70
+    },
+    {
+      "epoch": 1.8263665594855305,
+      "grad_norm": 3.226271152496338,
+      "learning_rate": 3.7120240506158433e-06,
+      "loss": 0.0688,
+      "num_input_tokens_seen": 486416,
+      "step": 71
+    },
+    {
+      "epoch": 1.852090032154341,
+      "grad_norm": 2.070322275161743,
+      "learning_rate": 3.6736789069647273e-06,
+      "loss": 0.0351,
+      "num_input_tokens_seen": 492896,
+      "step": 72
+    },
+    {
+      "epoch": 1.877813504823151,
+      "grad_norm": 2.4622035026550293,
+      "learning_rate": 3.634976249348867e-06,
+      "loss": 0.0246,
+      "num_input_tokens_seen": 499760,
+      "step": 73
+    },
+    {
+      "epoch": 1.9035369774919615,
+      "grad_norm": 3.902181625366211,
+      "learning_rate": 3.595927866972694e-06,
+      "loss": 0.0364,
+      "num_input_tokens_seen": 506816,
+      "step": 74
+    },
+    {
+      "epoch": 1.9292604501607717,
+      "grad_norm": 4.0147480964660645,
+      "learning_rate": 3.556545654351749e-06,
+      "loss": 0.0352,
+      "num_input_tokens_seen": 513648,
+      "step": 75
+    },
+    {
+      "epoch": 1.954983922829582,
+      "grad_norm": 3.6791763305664062,
+      "learning_rate": 3.516841607689501e-06,
+      "loss": 0.0915,
+      "num_input_tokens_seen": 520480,
+      "step": 76
+    },
+    {
+      "epoch": 1.9807073954983923,
+      "grad_norm": 1.641350269317627,
+      "learning_rate": 3.476827821223184e-06,
+      "loss": 0.0327,
+      "num_input_tokens_seen": 527056,
+      "step": 77
+    },
+    {
+      "epoch": 2.0064308681672025,
+      "grad_norm": 3.6227402687072754,
+      "learning_rate": 3.436516483539781e-06,
+      "loss": 0.0448,
+      "num_input_tokens_seen": 534112,
+      "step": 78
+    },
+    {
+      "epoch": 2.032154340836013,
+      "grad_norm": 1.654789924621582,
+      "learning_rate": 3.39591987386325e-06,
+      "loss": 0.0186,
+      "num_input_tokens_seen": 541024,
+      "step": 79
+    },
+    {
+      "epoch": 2.057877813504823,
+      "grad_norm": 2.0418519973754883,
+      "learning_rate": 3.3550503583141726e-06,
+      "loss": 0.0342,
+      "num_input_tokens_seen": 547888,
+      "step": 80
+    },
+    {
+      "epoch": 2.0836012861736335,
+      "grad_norm": 1.0242717266082764,
+      "learning_rate": 3.313920386142892e-06,
+      "loss": 0.0079,
+      "num_input_tokens_seen": 554592,
+      "step": 81
+    },
+    {
+      "epoch": 2.1093247588424435,
+      "grad_norm": 1.59933602809906,
+      "learning_rate": 3.272542485937369e-06,
+      "loss": 0.0177,
+      "num_input_tokens_seen": 561296,
+      "step": 82
+    },
+    {
+      "epoch": 2.135048231511254,
+      "grad_norm": 1.83909273147583,
+      "learning_rate": 3.230929261806842e-06,
+      "loss": 0.0139,
+      "num_input_tokens_seen": 567872,
+      "step": 83
+    },
+    {
+      "epoch": 2.1607717041800645,
+      "grad_norm": 1.933048963546753,
+      "learning_rate": 3.189093389542498e-06,
+      "loss": 0.0103,
+      "num_input_tokens_seen": 575072,
+      "step": 84
+    },
+    {
+      "epoch": 2.1864951768488745,
+      "grad_norm": 2.1397433280944824,
+      "learning_rate": 3.147047612756302e-06,
+      "loss": 0.0221,
+      "num_input_tokens_seen": 582080,
+      "step": 85
+    },
+    {
+      "epoch": 2.212218649517685,
+      "grad_norm": 0.7958673238754272,
+      "learning_rate": 3.1048047389991693e-06,
+      "loss": 0.0021,
+      "num_input_tokens_seen": 588816,
+      "step": 86
+    },
+    {
+      "epoch": 2.237942122186495,
+      "grad_norm": 1.8060617446899414,
+      "learning_rate": 3.062377635859663e-06,
+      "loss": 0.011,
+      "num_input_tokens_seen": 596032,
+      "step": 87
+    },
+    {
+      "epoch": 2.2636655948553055,
+      "grad_norm": 1.7997971773147583,
+      "learning_rate": 3.019779227044398e-06,
+      "loss": 0.0081,
+      "num_input_tokens_seen": 602672,
+      "step": 88
+    },
+    {
+      "epoch": 2.289389067524116,
+      "grad_norm": 2.0229434967041016,
+      "learning_rate": 2.9770224884413625e-06,
+      "loss": 0.0149,
+      "num_input_tokens_seen": 609424,
+      "step": 89
+    },
+    {
+      "epoch": 2.315112540192926,
+      "grad_norm": 0.5248342156410217,
+      "learning_rate": 2.9341204441673267e-06,
+      "loss": 0.001,
+      "num_input_tokens_seen": 616448,
+      "step": 90
+    },
+    {
+      "epoch": 2.3408360128617365,
+      "grad_norm": 1.1899443864822388,
+      "learning_rate": 2.8910861626005774e-06,
+      "loss": 0.007,
+      "num_input_tokens_seen": 623296,
+      "step": 91
+    },
+    {
+      "epoch": 2.3665594855305465,
+      "grad_norm": 3.1920082569122314,
+      "learning_rate": 2.847932752400164e-06,
+      "loss": 0.0089,
+      "num_input_tokens_seen": 630064,
+      "step": 92
+    },
+    {
+      "epoch": 2.392282958199357,
+      "grad_norm": 0.37337368726730347,
+      "learning_rate": 2.804673358512869e-06,
+      "loss": 0.0013,
+      "num_input_tokens_seen": 637088,
+      "step": 93
+    },
+    {
+      "epoch": 2.418006430868167,
+      "grad_norm": 4.124266147613525,
+      "learning_rate": 2.761321158169134e-06,
+      "loss": 0.0267,
+      "num_input_tokens_seen": 644080,
+      "step": 94
+    },
+    {
+      "epoch": 2.4437299035369775,
+      "grad_norm": 1.9108864068984985,
+      "learning_rate": 2.717889356869146e-06,
+      "loss": 0.0171,
+      "num_input_tokens_seen": 650848,
+      "step": 95
+    },
+    {
+      "epoch": 2.469453376205788,
+      "grad_norm": 3.225116729736328,
+      "learning_rate": 2.6743911843603134e-06,
+      "loss": 0.0375,
+      "num_input_tokens_seen": 657424,
+      "step": 96
+    },
+    {
+      "epoch": 2.495176848874598,
+      "grad_norm": 1.8133978843688965,
+      "learning_rate": 2.6308398906073603e-06,
+      "loss": 0.0101,
+      "num_input_tokens_seen": 664128,
+      "step": 97
+    },
+    {
+      "epoch": 2.5209003215434085,
+      "grad_norm": 3.179337739944458,
+      "learning_rate": 2.587248741756253e-06,
+      "loss": 0.0282,
+      "num_input_tokens_seen": 671120,
+      "step": 98
+    },
+    {
+      "epoch": 2.5466237942122185,
+      "grad_norm": 1.5852500200271606,
+      "learning_rate": 2.543631016093209e-06,
+      "loss": 0.0069,
+      "num_input_tokens_seen": 677920,
+      "step": 99
+    },
+    {
+      "epoch": 2.572347266881029,
+      "grad_norm": 2.1428685188293457,
+      "learning_rate": 2.5e-06,
+      "loss": 0.0135,
+      "num_input_tokens_seen": 684960,
+      "step": 100
+    },
+    {
+      "epoch": 2.598070739549839,
+      "grad_norm": 1.0775537490844727,
+      "learning_rate": 2.4563689839067913e-06,
+      "loss": 0.0062,
+      "num_input_tokens_seen": 691856,
+      "step": 101
+    },
+    {
+      "epoch": 2.6237942122186495,
+      "grad_norm": 1.5073256492614746,
+      "learning_rate": 2.4127512582437486e-06,
+      "loss": 0.005,
+      "num_input_tokens_seen": 698512,
+      "step": 102
+    },
+    {
+      "epoch": 2.64951768488746,
+      "grad_norm": 1.7251828908920288,
+      "learning_rate": 2.3691601093926406e-06,
+      "loss": 0.0285,
+      "num_input_tokens_seen": 705440,
+      "step": 103
+    },
+    {
+      "epoch": 2.67524115755627,
+      "grad_norm": 2.271348237991333,
+      "learning_rate": 2.325608815639687e-06,
+      "loss": 0.0225,
+      "num_input_tokens_seen": 712528,
+      "step": 104
+    },
+    {
+      "epoch": 2.7009646302250805,
+      "grad_norm": 0.9558768272399902,
+      "learning_rate": 2.2821106431308546e-06,
+      "loss": 0.028,
+      "num_input_tokens_seen": 719168,
+      "step": 105
+    },
+    {
+      "epoch": 2.7266881028938905,
+      "grad_norm": 2.1378886699676514,
+      "learning_rate": 2.238678841830867e-06,
+      "loss": 0.0176,
+      "num_input_tokens_seen": 725904,
+      "step": 106
+    },
+    {
+      "epoch": 2.752411575562701,
+      "grad_norm": 0.9674012660980225,
+      "learning_rate": 2.195326641487132e-06,
+      "loss": 0.0047,
+      "num_input_tokens_seen": 732480,
+      "step": 107
+    },
+    {
+      "epoch": 2.778135048231511,
+      "grad_norm": 0.6068861484527588,
+      "learning_rate": 2.1520672475998374e-06,
+      "loss": 0.0135,
+      "num_input_tokens_seen": 739184,
+      "step": 108
+    },
+    {
+      "epoch": 2.8038585209003215,
+      "grad_norm": 0.833519458770752,
+      "learning_rate": 2.1089138373994226e-06,
+      "loss": 0.0044,
+      "num_input_tokens_seen": 746320,
+      "step": 109
+    },
+    {
+      "epoch": 2.829581993569132,
+      "grad_norm": 1.4018948078155518,
+      "learning_rate": 2.0658795558326745e-06,
+      "loss": 0.0252,
+      "num_input_tokens_seen": 753136,
+      "step": 110
+    },
+    {
+      "epoch": 2.855305466237942,
+      "grad_norm": 1.5698492527008057,
+      "learning_rate": 2.022977511558638e-06,
+      "loss": 0.0249,
+      "num_input_tokens_seen": 760096,
+      "step": 111
+    },
+    {
+      "epoch": 2.8810289389067525,
+      "grad_norm": 2.051377534866333,
+      "learning_rate": 1.9802207729556023e-06,
+      "loss": 0.0146,
+      "num_input_tokens_seen": 767104,
+      "step": 112
+    },
+    {
+      "epoch": 2.906752411575563,
+      "grad_norm": 0.48683854937553406,
+      "learning_rate": 1.937622364140338e-06,
+      "loss": 0.0044,
+      "num_input_tokens_seen": 773872,
+      "step": 113
+    },
+    {
+      "epoch": 2.932475884244373,
+      "grad_norm": 0.9895896911621094,
+      "learning_rate": 1.895195261000831e-06,
+      "loss": 0.0054,
+      "num_input_tokens_seen": 780736,
+      "step": 114
+    },
+    {
+      "epoch": 2.958199356913183,
+      "grad_norm": 0.8025338053703308,
+      "learning_rate": 1.852952387243698e-06,
+      "loss": 0.0106,
+      "num_input_tokens_seen": 787424,
+      "step": 115
+    },
+    {
+      "epoch": 2.9839228295819935,
+      "grad_norm": 1.7318856716156006,
+      "learning_rate": 1.8109066104575023e-06,
+      "loss": 0.0167,
+      "num_input_tokens_seen": 794224,
+      "step": 116
+    },
+    {
+      "epoch": 3.009646302250804,
+      "grad_norm": 1.5407195091247559,
+      "learning_rate": 1.7690707381931585e-06,
+      "loss": 0.009,
+      "num_input_tokens_seen": 801088,
+      "step": 117
+    },
+    {
+      "epoch": 3.035369774919614,
+      "grad_norm": 0.25947141647338867,
+      "learning_rate": 1.7274575140626318e-06,
+      "loss": 0.0024,
+      "num_input_tokens_seen": 808048,
+      "step": 118
+    },
+    {
+      "epoch": 3.0610932475884245,
+      "grad_norm": 2.6516857147216797,
+      "learning_rate": 1.686079613857109e-06,
+      "loss": 0.0235,
+      "num_input_tokens_seen": 814800,
+      "step": 119
+    },
+    {
+      "epoch": 3.0868167202572345,
+      "grad_norm": 0.9633734822273254,
+      "learning_rate": 1.6449496416858285e-06,
+      "loss": 0.0179,
+      "num_input_tokens_seen": 821536,
+      "step": 120
+    },
+    {
+      "epoch": 3.112540192926045,
+      "grad_norm": 1.0766749382019043,
+      "learning_rate": 1.6040801261367494e-06,
+      "loss": 0.0059,
+      "num_input_tokens_seen": 828128,
+      "step": 121
+    },
+    {
+      "epoch": 3.1382636655948555,
+      "grad_norm": 0.2934500277042389,
+      "learning_rate": 1.56348351646022e-06,
+      "loss": 0.0017,
+      "num_input_tokens_seen": 834608,
+      "step": 122
+    },
+    {
+      "epoch": 3.1639871382636655,
+      "grad_norm": 0.3029981553554535,
+      "learning_rate": 1.5231721787768162e-06,
+      "loss": 0.0018,
+      "num_input_tokens_seen": 841360,
+      "step": 123
+    },
+    {
+      "epoch": 3.189710610932476,
+      "grad_norm": 0.44817763566970825,
+      "learning_rate": 1.4831583923105e-06,
+      "loss": 0.0032,
+      "num_input_tokens_seen": 848496,
+      "step": 124
+    },
+    {
+      "epoch": 3.215434083601286,
+      "grad_norm": 0.21245715022087097,
+      "learning_rate": 1.443454345648252e-06,
+      "loss": 0.0019,
+      "num_input_tokens_seen": 855424,
+      "step": 125
+    },
+    {
+      "epoch": 3.2411575562700965,
+      "grad_norm": 0.22589029371738434,
+      "learning_rate": 1.4040721330273063e-06,
+      "loss": 0.0014,
+      "num_input_tokens_seen": 862224,
+      "step": 126
+    },
+    {
+      "epoch": 3.266881028938907,
+      "grad_norm": 0.5972980856895447,
+      "learning_rate": 1.3650237506511333e-06,
+      "loss": 0.0052,
+      "num_input_tokens_seen": 869312,
+      "step": 127
+    },
+    {
+      "epoch": 3.292604501607717,
+      "grad_norm": 0.06931939721107483,
+      "learning_rate": 1.3263210930352737e-06,
+      "loss": 0.0005,
+      "num_input_tokens_seen": 876272,
+      "step": 128
+    },
+    {
+      "epoch": 3.3183279742765275,
+      "grad_norm": 1.2623660564422607,
+      "learning_rate": 1.2879759493841577e-06,
+      "loss": 0.0131,
+      "num_input_tokens_seen": 883072,
+      "step": 129
+    },
+    {
+      "epoch": 3.3440514469453375,
+      "grad_norm": 0.5440672636032104,
+      "learning_rate": 1.2500000000000007e-06,
+      "loss": 0.0009,
+      "num_input_tokens_seen": 889920,
+      "step": 130
+    },
+    {
+      "epoch": 3.369774919614148,
+      "grad_norm": 0.6807874441146851,
+      "learning_rate": 1.2124048127248644e-06,
+      "loss": 0.0057,
+      "num_input_tokens_seen": 896896,
+      "step": 131
+    },
+    {
+      "epoch": 3.395498392282958,
+      "grad_norm": 0.03898243606090546,
+      "learning_rate": 1.1752018394169882e-06,
+      "loss": 0.0002,
+      "num_input_tokens_seen": 903600,
+      "step": 132
+    },
+    {
+      "epoch": 3.4212218649517685,
+      "grad_norm": 0.0243828147649765,
+      "learning_rate": 1.1384024124624324e-06,
+      "loss": 0.0002,
+      "num_input_tokens_seen": 910208,
+      "step": 133
+    },
+    {
+      "epoch": 3.446945337620579,
+      "grad_norm": 2.5577075481414795,
+      "learning_rate": 1.1020177413231334e-06,
+      "loss": 0.0145,
+      "num_input_tokens_seen": 917328,
+      "step": 134
+    },
+    {
+      "epoch": 3.472668810289389,
+      "grad_norm": 1.2599172592163086,
+      "learning_rate": 1.0660589091223854e-06,
+      "loss": 0.0034,
+      "num_input_tokens_seen": 924192,
+      "step": 135
+    },
+    {
+      "epoch": 3.4983922829581995,
+      "grad_norm": 1.3292278051376343,
+      "learning_rate": 1.0305368692688175e-06,
+      "loss": 0.0156,
+      "num_input_tokens_seen": 930784,
+      "step": 136
+    },
+    {
+      "epoch": 3.5241157556270095,
+      "grad_norm": 0.2977862060070038,
+      "learning_rate": 9.95462442119879e-07,
+      "loss": 0.0013,
+      "num_input_tokens_seen": 937856,
+      "step": 137
+    },
+    {
+      "epoch": 3.54983922829582,
+      "grad_norm": 0.3527968227863312,
+      "learning_rate": 9.608463116858544e-07,
+      "loss": 0.0007,
+      "num_input_tokens_seen": 944640,
+      "step": 138
+    },
+    {
+      "epoch": 3.57556270096463,
+      "grad_norm": 0.16851164400577545,
+      "learning_rate": 9.266990223754069e-07,
+      "loss": 0.0005,
+      "num_input_tokens_seen": 951504,
+      "step": 139
+    },
+    {
+      "epoch": 3.6012861736334405,
+      "grad_norm": 0.4394027292728424,
+      "learning_rate": 8.930309757836517e-07,
+      "loss": 0.0034,
+      "num_input_tokens_seen": 958240,
+      "step": 140
+    },
+    {
+      "epoch": 3.627009646302251,
+      "grad_norm": 0.0209315475076437,
+      "learning_rate": 8.598524275237321e-07,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 964912,
+      "step": 141
+    },
+    {
+      "epoch": 3.652733118971061,
+      "grad_norm": 0.35701486468315125,
+      "learning_rate": 8.271734841028553e-07,
+      "loss": 0.001,
+      "num_input_tokens_seen": 971872,
+      "step": 142
+    },
+    {
+      "epoch": 3.6784565916398715,
+      "grad_norm": 2.0450756549835205,
+      "learning_rate": 7.950040998437541e-07,
+      "loss": 0.0123,
+      "num_input_tokens_seen": 978640,
+      "step": 143
+    },
+    {
+      "epoch": 3.7041800643086815,
+      "grad_norm": 0.057205893099308014,
+      "learning_rate": 7.633540738525066e-07,
+      "loss": 0.0002,
+      "num_input_tokens_seen": 985328,
+      "step": 144
+    },
+    {
+      "epoch": 3.729903536977492,
+      "grad_norm": 1.7251015901565552,
+      "learning_rate": 7.322330470336314e-07,
+      "loss": 0.011,
+      "num_input_tokens_seen": 992224,
+      "step": 145
+    },
+    {
+      "epoch": 3.755627009646302,
+      "grad_norm": 0.417669415473938,
+      "learning_rate": 7.016504991533727e-07,
+      "loss": 0.0008,
+      "num_input_tokens_seen": 998688,
+      "step": 146
+    },
+    {
+      "epoch": 3.7813504823151125,
+      "grad_norm": 0.06796720623970032,
+      "learning_rate": 6.716157459520739e-07,
+      "loss": 0.0003,
+      "num_input_tokens_seen": 1006032,
+      "step": 147
+    },
+    {
+      "epoch": 3.807073954983923,
+      "grad_norm": 0.5570574998855591,
+      "learning_rate": 6.421379363065142e-07,
+      "loss": 0.0018,
+      "num_input_tokens_seen": 1012944,
+      "step": 148
+    },
+    {
+      "epoch": 3.832797427652733,
+      "grad_norm": 0.557479977607727,
+      "learning_rate": 6.1322604944307e-07,
+      "loss": 0.0016,
+      "num_input_tokens_seen": 1019856,
+      "step": 149
+    },
+    {
+      "epoch": 3.8585209003215435,
+      "grad_norm": 0.48776909708976746,
+      "learning_rate": 5.848888922025553e-07,
+      "loss": 0.0021,
+      "num_input_tokens_seen": 1026656,
+      "step": 150
+    },
+    {
+      "epoch": 3.884244372990354,
+      "grad_norm": 0.025394350290298462,
+      "learning_rate": 5.571350963575728e-07,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 1033696,
+      "step": 151
+    },
+    {
+      "epoch": 3.909967845659164,
+      "grad_norm": 0.03195232152938843,
+      "learning_rate": 5.299731159831953e-07,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 1040240,
+      "step": 152
+    },
+    {
+      "epoch": 3.935691318327974,
+      "grad_norm": 0.0676019936800003,
+      "learning_rate": 5.034112248817685e-07,
+      "loss": 0.0003,
+      "num_input_tokens_seen": 1046992,
+      "step": 153
+    },
+    {
+      "epoch": 3.9614147909967845,
+      "grad_norm": 0.04871657118201256,
+      "learning_rate": 4.774575140626317e-07,
+      "loss": 0.0002,
+      "num_input_tokens_seen": 1053936,
+      "step": 154
+    },
+    {
+      "epoch": 3.987138263665595,
+      "grad_norm": 0.08272106945514679,
+      "learning_rate": 4.5211988927752026e-07,
+      "loss": 0.0002,
+      "num_input_tokens_seen": 1060576,
+      "step": 155
+    },
+    {
+      "epoch": 4.012861736334405,
+      "grad_norm": 0.009170151315629482,
+      "learning_rate": 4.27406068612396e-07,
+      "loss": 0.0,
+      "num_input_tokens_seen": 1067648,
+      "step": 156
+    },
+    {
+      "epoch": 4.038585209003215,
+      "grad_norm": 0.10544967651367188,
+      "learning_rate": 4.033235801364402e-07,
+      "loss": 0.0003,
+      "num_input_tokens_seen": 1074512,
+      "step": 157
+    },
+    {
+      "epoch": 4.064308681672026,
+      "grad_norm": 0.04709336906671524,
+      "learning_rate": 3.798797596089351e-07,
+      "loss": 0.0002,
+      "num_input_tokens_seen": 1081296,
+      "step": 158
+    },
+    {
+      "epoch": 4.090032154340836,
+      "grad_norm": 0.042879387736320496,
+      "learning_rate": 3.5708174824471947e-07,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 1087888,
+      "step": 159
+    },
+    {
+      "epoch": 4.115755627009646,
+      "grad_norm": 0.03649509325623512,
+      "learning_rate": 3.3493649053890325e-07,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 1094960,
+      "step": 160
+    },
+    {
+      "epoch": 4.141479099678457,
+      "grad_norm": 0.01273419987410307,
+      "learning_rate": 3.134507321515107e-07,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 1101776,
+      "step": 161
+    },
+    {
+      "epoch": 4.167202572347267,
+      "grad_norm": 0.006498055998235941,
+      "learning_rate": 2.9263101785268253e-07,
+      "loss": 0.0,
+      "num_input_tokens_seen": 1108736,
+      "step": 162
+    },
+    {
+      "epoch": 4.192926045016077,
+      "grad_norm": 0.027757082134485245,
+      "learning_rate": 2.7248368952908055e-07,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 1115744,
+      "step": 163
+    },
+    {
+      "epoch": 4.218649517684887,
+      "grad_norm": 0.005632288288325071,
+      "learning_rate": 2.53014884252083e-07,
+      "loss": 0.0,
+      "num_input_tokens_seen": 1122592,
+      "step": 164
+    },
+    {
+      "epoch": 4.244372990353698,
+      "grad_norm": 0.014449907466769218,
+      "learning_rate": 2.3423053240837518e-07,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 1129136,
+      "step": 165
+    },
+    {
+      "epoch": 4.270096463022508,
+      "grad_norm": 0.013245908543467522,
+      "learning_rate": 2.1613635589349756e-07,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 1136080,
+      "step": 166
+    },
+    {
+      "epoch": 4.295819935691318,
+      "grad_norm": 0.12783974409103394,
+      "learning_rate": 1.9873786636889908e-07,
+      "loss": 0.0002,
+      "num_input_tokens_seen": 1142976,
+      "step": 167
+    },
+    {
+      "epoch": 4.321543408360129,
+      "grad_norm": 0.08086368441581726,
+      "learning_rate": 1.8204036358303173e-07,
+      "loss": 0.0002,
+      "num_input_tokens_seen": 1149712,
+      "step": 168
+    },
+    {
+      "epoch": 4.347266881028939,
+      "grad_norm": 0.018286822363734245,
+      "learning_rate": 1.6604893375699594e-07,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 1156336,
+      "step": 169
+    },
+    {
+      "epoch": 4.372990353697749,
+      "grad_norm": 0.009972570464015007,
+      "learning_rate": 1.507684480352292e-07,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 1163120,
+      "step": 170
+    },
+    {
+      "epoch": 4.39871382636656,
+      "grad_norm": 0.9909194707870483,
+      "learning_rate": 1.362035610017079e-07,
+      "loss": 0.0115,
+      "num_input_tokens_seen": 1170032,
+      "step": 171
+    },
+    {
+      "epoch": 4.42443729903537,
+      "grad_norm": 0.21059785783290863,
+      "learning_rate": 1.223587092621162e-07,
+      "loss": 0.0005,
+      "num_input_tokens_seen": 1176832,
+      "step": 172
+    },
+    {
+      "epoch": 4.45016077170418,
+      "grad_norm": 0.08872511237859726,
+      "learning_rate": 1.0923811009241142e-07,
+      "loss": 0.0003,
+      "num_input_tokens_seen": 1183856,
+      "step": 173
+    },
+    {
+      "epoch": 4.47588424437299,
+      "grad_norm": 0.7515408396720886,
+      "learning_rate": 9.684576015420277e-08,
+      "loss": 0.005,
+      "num_input_tokens_seen": 1190544,
+      "step": 174
+    },
+    {
+      "epoch": 4.501607717041801,
+      "grad_norm": 0.0772569328546524,
+      "learning_rate": 8.518543427732951e-08,
+      "loss": 0.0003,
+      "num_input_tokens_seen": 1197168,
+      "step": 175
+    },
+    {
+      "epoch": 4.527331189710611,
+      "grad_norm": 0.19207893311977386,
+      "learning_rate": 7.426068431000883e-08,
+      "loss": 0.0008,
+      "num_input_tokens_seen": 1203776,
+      "step": 176
+    },
+    {
+      "epoch": 4.553054662379421,
+      "grad_norm": 0.007837573066353798,
+      "learning_rate": 6.407483803691216e-08,
+      "loss": 0.0,
+      "num_input_tokens_seen": 1210816,
+      "step": 177
+    },
+    {
+      "epoch": 4.578778135048232,
+      "grad_norm": 0.009602434933185577,
+      "learning_rate": 5.463099816548578e-08,
+      "loss": 0.0,
+      "num_input_tokens_seen": 1217696,
+      "step": 178
+    },
+    {
+      "epoch": 4.604501607717042,
+      "grad_norm": 1.3553417921066284,
+      "learning_rate": 4.593204138084006e-08,
+      "loss": 0.0042,
+      "num_input_tokens_seen": 1224704,
+      "step": 179
+    },
+    {
+      "epoch": 4.630225080385852,
+      "grad_norm": 0.17981038987636566,
+      "learning_rate": 3.798061746947995e-08,
+      "loss": 0.0004,
+      "num_input_tokens_seen": 1231664,
+      "step": 180
+    },
+    {
+      "epoch": 4.655948553054662,
+      "grad_norm": 0.04047521948814392,
+      "learning_rate": 3.077914851215585e-08,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 1238240,
+      "step": 181
+    },
+    {
+      "epoch": 4.681672025723473,
+      "grad_norm": 0.02045159973204136,
+      "learning_rate": 2.4329828146074096e-08,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 1244832,
+      "step": 182
+    },
+    {
+      "epoch": 4.707395498392283,
+      "grad_norm": 0.10965674370527267,
+      "learning_rate": 1.8634620896695044e-08,
+      "loss": 0.0004,
+      "num_input_tokens_seen": 1251296,
+      "step": 183
+    },
+    {
+      "epoch": 4.733118971061093,
+      "grad_norm": 0.008195169270038605,
+      "learning_rate": 1.3695261579316776e-08,
+      "loss": 0.0,
+      "num_input_tokens_seen": 1257968,
+      "step": 184
+    },
+    {
+      "epoch": 4.758842443729904,
+      "grad_norm": 0.25205764174461365,
+      "learning_rate": 9.513254770636138e-09,
+      "loss": 0.0009,
+      "num_input_tokens_seen": 1264928,
+      "step": 185
+    },
+    {
+      "epoch": 4.784565916398714,
+      "grad_norm": 0.031606338918209076,
+      "learning_rate": 6.089874350439507e-09,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 1271792,
+      "step": 186
+    },
+    {
+      "epoch": 4.810289389067524,
+      "grad_norm": 0.19231323897838593,
+      "learning_rate": 3.4261631135654174e-09,
+      "loss": 0.0004,
+      "num_input_tokens_seen": 1279024,
+      "step": 187
+    },
+    {
+      "epoch": 4.836012861736334,
+      "grad_norm": 0.06766192615032196,
+      "learning_rate": 1.5229324522605949e-09,
+      "loss": 0.0002,
+      "num_input_tokens_seen": 1285792,
+      "step": 188
+    },
+    {
+      "epoch": 4.861736334405145,
+      "grad_norm": 0.005693785380572081,
+      "learning_rate": 3.8076210902182607e-10,
+      "loss": 0.0,
+      "num_input_tokens_seen": 1292576,
+      "step": 189
+    },
+    {
+      "epoch": 4.887459807073955,
+      "grad_norm": 0.02735370770096779,
+      "learning_rate": 0.0,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 1299392,
+      "step": 190
+    },
+    {
+      "epoch": 4.887459807073955,
+      "num_input_tokens_seen": 1299392,
+      "step": 190,
+      "total_flos": 5.151317702790349e+16,
+      "train_loss": 0.3433768034317166,
+      "train_runtime": 2162.0959,
+      "train_samples_per_second": 11.489,
+      "train_steps_per_second": 0.088
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 190,
+  "num_input_tokens_seen": 1299392,
+  "num_train_epochs": 5,
+  "save_steps": 1000,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 5.151317702790349e+16,
+  "train_batch_size": 2,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:59c20394a81d6a411e14385c1f4bccd2cbf8486e7c193698844b9070fbad87d6
+size 6584