simpler

UNIVLA_SIMPLER_BRIDGE_VIDEO_BS128_20K/config.json

@@ -0,0 +1,62 @@
+{
+  "_name_or_path": "/share/project/yuqi.wang/OmniSim/logs/ckpts/WORLD_MODEL_POSTTRAIN_VIDEO3",
+  "action_config": {
+    "_attn_implementation": "flash_attention_2",
+    "action_dim": 7,
+    "attention_dropout": 0.0,
+    "hidden_act": "silu",
+    "hidden_size": 4096,
+    "initializer_range": 0.02,
+    "intermediate_size": 2048,
+    "max_position_embeddings": 5120,
+    "num_attention_heads": 32,
+    "num_hidden_layers": 2,
+    "num_key_value_heads": 8,
+    "pad_token_id": 151643,
+    "pretraining_tp": 1,
+    "rms_norm_eps": 1e-05,
+    "rope_scaling": null,
+    "rope_theta": 10000.0,
+    "tie_word_embeddings": false,
+    "use_cache": true,
+    "vision_loss_weight": 5.0
+  },
+  "action_dim": 7,
+  "action_experts": false,
+  "architectures": [
+    "Emu3MoE"
+  ],
+  "attention_dropout": 0.1,
+  "boa_token_id": 151844,
+  "boi_token_id": 151852,
+  "bos_token_id": 151849,
+  "bov_token_id": 151854,
+  "eof_token_id": 151847,
+  "eoi_token_id": 151853,
+  "eol_token_id": 151846,
+  "eos_token_id": 151850,
+  "eov_token_id": 184621,
+  "hidden_act": "silu",
+  "hidden_size": 4096,
+  "image_area": 262144,
+  "img_token_id": 151851,
+  "initializer_range": 0.02,
+  "intermediate_size": 14336,
+  "max_position_embeddings": 2400,
+  "model_type": "Emu3",
+  "num_attention_heads": 32,
+  "num_hidden_layers": 32,
+  "num_key_value_heads": 8,
+  "pad_token_id": 151643,
+  "pretraining_tp": 1,
+  "qkv_bias": false,
+  "rms_norm_eps": 1e-05,
+  "rope_scaling": null,
+  "rope_theta": 1000000.0,
+  "tie_word_embeddings": false,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.44.0",
+  "use_cache": true,
+  "vision_loss_weight": 0.5,
+  "vocab_size": 184622
+}

UNIVLA_SIMPLER_BRIDGE_VIDEO_BS128_20K/generation_config.json

@@ -0,0 +1,7 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 151849,
+  "eos_token_id": 151850,
+  "pad_token_id": 151643,
+  "transformers_version": "4.44.0"
+}

UNIVLA_SIMPLER_BRIDGE_VIDEO_BS128_20K/model-00001-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:98e51788157918498bcad4d774bf8257a506f0bab94f99c411a8ebcb12e3bfd2
+size 4884766656

UNIVLA_SIMPLER_BRIDGE_VIDEO_BS128_20K/model-00002-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:207f2a0beaab76e2589e041502a046515524e5d3242d0411afba8b44b9067e59
+size 4999819320

UNIVLA_SIMPLER_BRIDGE_VIDEO_BS128_20K/model-00003-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cf39f161962aeb2633e3006e64319d46f5dc4984e697546907c9d88857cdd911
+size 4915916184

UNIVLA_SIMPLER_BRIDGE_VIDEO_BS128_20K/model-00004-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b2da3562acacf4a3908fffb5bfcc5892a198a5bec33531ace21e6f33ed9fee7c
+size 2183554760

UNIVLA_SIMPLER_BRIDGE_VIDEO_BS128_20K/model.safetensors.index.json

@@ -0,0 +1,298 @@
+{
+  "metadata": {
+    "total_size": 16984023040
+  },
+  "weight_map": {
+    "lm_head.weight": "model-00004-of-00004.safetensors",
+    "model.embed_tokens.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.10.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.10.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.10.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.10.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.10.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.10.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.10.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.10.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.10.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.18.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.18.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.18.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.18.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.18.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.18.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.18.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.18.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.18.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.19.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.19.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.19.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.19.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.19.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.19.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.19.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.19.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.19.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.2.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.2.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.2.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.2.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.2.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.2.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.2.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.2.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.2.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.20.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.20.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.20.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.20.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.20.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.20.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.20.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.20.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.20.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.21.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.21.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.21.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.21.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.21.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.21.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.21.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.21.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.21.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.22.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.22.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.22.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.22.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.22.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.22.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.22.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.22.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.22.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.28.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.28.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.28.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.28.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.28.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.28.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.28.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.28.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.28.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.29.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.29.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.29.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.29.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.29.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.29.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.29.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.29.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.29.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.3.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.3.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.3.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.3.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.3.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.3.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.3.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.3.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.3.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.30.input_layernorm.weight": "model-00004-of-00004.safetensors",
+    "model.layers.30.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.30.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.30.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.30.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
+    "model.layers.30.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.30.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.30.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.30.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.31.input_layernorm.weight": "model-00004-of-00004.safetensors",
+    "model.layers.31.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.31.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.31.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.31.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
+    "model.layers.31.self_attn.k_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.31.self_attn.o_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.31.self_attn.q_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.31.self_attn.v_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.4.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.7.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.7.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.7.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.8.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.8.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.8.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.8.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.8.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.8.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.8.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.8.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.8.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.9.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.9.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.9.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.9.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.9.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.9.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.9.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.9.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.9.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.norm.weight": "model-00004-of-00004.safetensors"
+  }
+}

UNIVLA_SIMPLER_BRIDGE_VIDEO_BS128_20K/special_tokens_map.json

@@ -0,0 +1,5 @@
+{
+  "bos_token": "<|extra_203|>",
+  "eos_token": "<|extra_204|>",
+  "pad_token": "<|endoftext|>"
+}

UNIVLA_SIMPLER_BRIDGE_VIDEO_BS128_20K/tokenizer_config.json

@@ -0,0 +1,17 @@
+{
+  "added_tokens_decoder": {},
+  "auto_map": {
+    "AutoTokenizer": [
+      "tokenization_emu3.Emu3Tokenizer",
+      null
+    ]
+  },
+  "bos_token": "<|extra_203|>",
+  "clean_up_tokenization_spaces": true,
+  "eos_token": "<|extra_204|>",
+  "model_max_length": 2400,
+  "pad_token": "<|endoftext|>",
+  "padding_side": "right",
+  "tokenizer_class": "Emu3Tokenizer",
+  "use_fast": false
+}

UNIVLA_SIMPLER_BRIDGE_VIDEO_BS128_20K/trainer_state.json

@@ -0,0 +1,2842 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 42.73504273504273,
+  "eval_steps": 500,
+  "global_step": 20000,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.10683760683760683,
+      "grad_norm": 17.50655133370389,
+      "learning_rate": 8.000000000000001e-06,
+      "loss": 12.5513,
+      "step": 50
+    },
+    {
+      "epoch": 0.21367521367521367,
+      "grad_norm": 3.9805514071551547,
+      "learning_rate": 1.6000000000000003e-05,
+      "loss": 6.0084,
+      "step": 100
+    },
+    {
+      "epoch": 0.32051282051282054,
+      "grad_norm": 4.788411878788504,
+      "learning_rate": 2.4e-05,
+      "loss": 5.3805,
+      "step": 150
+    },
+    {
+      "epoch": 0.42735042735042733,
+      "grad_norm": 5.333725120141322,
+      "learning_rate": 3.2000000000000005e-05,
+      "loss": 5.2249,
+      "step": 200
+    },
+    {
+      "epoch": 0.5341880341880342,
+      "grad_norm": 15.609303150641697,
+      "learning_rate": 4e-05,
+      "loss": 4.9602,
+      "step": 250
+    },
+    {
+      "epoch": 0.6410256410256411,
+      "grad_norm": 7.144767244257112,
+      "learning_rate": 4.8e-05,
+      "loss": 4.6148,
+      "step": 300
+    },
+    {
+      "epoch": 0.7478632478632479,
+      "grad_norm": 25.080573283918373,
+      "learning_rate": 5.6e-05,
+      "loss": 4.4841,
+      "step": 350
+    },
+    {
+      "epoch": 0.8547008547008547,
+      "grad_norm": 3.8773812391539053,
+      "learning_rate": 6.400000000000001e-05,
+      "loss": 4.3781,
+      "step": 400
+    },
+    {
+      "epoch": 0.9615384615384616,
+      "grad_norm": 3.1869326815909034,
+      "learning_rate": 7.2e-05,
+      "loss": 4.3205,
+      "step": 450
+    },
+    {
+      "epoch": 1.0683760683760684,
+      "grad_norm": 5.6076616736212594,
+      "learning_rate": 8e-05,
+      "loss": 4.2551,
+      "step": 500
+    },
+    {
+      "epoch": 1.1752136752136753,
+      "grad_norm": 3.645026039480543,
+      "learning_rate": 7.999878333936981e-05,
+      "loss": 4.2133,
+      "step": 550
+    },
+    {
+      "epoch": 1.282051282051282,
+      "grad_norm": 2.9794225693992953,
+      "learning_rate": 7.999513343642656e-05,
+      "loss": 4.151,
+      "step": 600
+    },
+    {
+      "epoch": 1.3888888888888888,
+      "grad_norm": 3.38064005034008,
+      "learning_rate": 7.998905052800726e-05,
+      "loss": 4.1177,
+      "step": 650
+    },
+    {
+      "epoch": 1.4957264957264957,
+      "grad_norm": 2.67282703236483,
+      "learning_rate": 7.998053500882307e-05,
+      "loss": 4.0967,
+      "step": 700
+    },
+    {
+      "epoch": 1.6025641025641026,
+      "grad_norm": 2.42080044442278,
+      "learning_rate": 7.996958743143387e-05,
+      "loss": 4.0583,
+      "step": 750
+    },
+    {
+      "epoch": 1.7094017094017095,
+      "grad_norm": 2.295466399827455,
+      "learning_rate": 7.995620850621226e-05,
+      "loss": 4.0284,
+      "step": 800
+    },
+    {
+      "epoch": 1.8162393162393162,
+      "grad_norm": 2.402563069574593,
+      "learning_rate": 7.994039910129751e-05,
+      "loss": 4.0166,
+      "step": 850
+    },
+    {
+      "epoch": 1.9230769230769231,
+      "grad_norm": 2.29603312108278,
+      "learning_rate": 7.992216024253926e-05,
+      "loss": 3.9985,
+      "step": 900
+    },
+    {
+      "epoch": 2.02991452991453,
+      "grad_norm": 3.248293648782872,
+      "learning_rate": 7.990149311343093e-05,
+      "loss": 3.9787,
+      "step": 950
+    },
+    {
+      "epoch": 2.1367521367521367,
+      "grad_norm": 3.134410726356944,
+      "learning_rate": 7.987839905503288e-05,
+      "loss": 3.9654,
+      "step": 1000
+    },
+    {
+      "epoch": 2.2435897435897436,
+      "grad_norm": 1.9375432532620036,
+      "learning_rate": 7.985287956588549e-05,
+      "loss": 3.9518,
+      "step": 1050
+    },
+    {
+      "epoch": 2.3504273504273505,
+      "grad_norm": 2.3186207778993406,
+      "learning_rate": 7.982493630191183e-05,
+      "loss": 3.9365,
+      "step": 1100
+    },
+    {
+      "epoch": 2.4572649572649574,
+      "grad_norm": 6.541119724742696,
+      "learning_rate": 7.979457107631025e-05,
+      "loss": 3.9315,
+      "step": 1150
+    },
+    {
+      "epoch": 2.564102564102564,
+      "grad_norm": 2.0067233088832968,
+      "learning_rate": 7.976178585943674e-05,
+      "loss": 3.9206,
+      "step": 1200
+    },
+    {
+      "epoch": 2.6709401709401708,
+      "grad_norm": 4.556924836426837,
+      "learning_rate": 7.972658277867703e-05,
+      "loss": 3.9179,
+      "step": 1250
+    },
+    {
+      "epoch": 2.7777777777777777,
+      "grad_norm": 2.0720526525590044,
+      "learning_rate": 7.968896411830859e-05,
+      "loss": 3.9087,
+      "step": 1300
+    },
+    {
+      "epoch": 2.8846153846153846,
+      "grad_norm": 2.1508177732338107,
+      "learning_rate": 7.964893231935236e-05,
+      "loss": 3.8873,
+      "step": 1350
+    },
+    {
+      "epoch": 2.9914529914529915,
+      "grad_norm": 2.3292099645567546,
+      "learning_rate": 7.960648997941442e-05,
+      "loss": 3.8856,
+      "step": 1400
+    },
+    {
+      "epoch": 3.0982905982905984,
+      "grad_norm": 2.0007587424147673,
+      "learning_rate": 7.95616398525174e-05,
+      "loss": 3.8742,
+      "step": 1450
+    },
+    {
+      "epoch": 3.2051282051282053,
+      "grad_norm": 1.846720032879114,
+      "learning_rate": 7.951438484892173e-05,
+      "loss": 3.87,
+      "step": 1500
+    },
+    {
+      "epoch": 3.3119658119658117,
+      "grad_norm": 1.9472394231718875,
+      "learning_rate": 7.946472803493691e-05,
+      "loss": 3.865,
+      "step": 1550
+    },
+    {
+      "epoch": 3.4188034188034186,
+      "grad_norm": 1.78425859382252,
+      "learning_rate": 7.941267263272242e-05,
+      "loss": 3.8535,
+      "step": 1600
+    },
+    {
+      "epoch": 3.5256410256410255,
+      "grad_norm": 4.031312927228092,
+      "learning_rate": 7.935822202007872e-05,
+      "loss": 3.8422,
+      "step": 1650
+    },
+    {
+      "epoch": 3.6324786324786325,
+      "grad_norm": 1.8872813142317872,
+      "learning_rate": 7.930137973022801e-05,
+      "loss": 3.8451,
+      "step": 1700
+    },
+    {
+      "epoch": 3.7393162393162394,
+      "grad_norm": 1.9797333625472995,
+      "learning_rate": 7.924214945158504e-05,
+      "loss": 3.8427,
+      "step": 1750
+    },
+    {
+      "epoch": 3.8461538461538463,
+      "grad_norm": 1.7965847278493243,
+      "learning_rate": 7.918053502751772e-05,
+      "loss": 3.8368,
+      "step": 1800
+    },
+    {
+      "epoch": 3.952991452991453,
+      "grad_norm": 2.011018560831316,
+      "learning_rate": 7.911654045609771e-05,
+      "loss": 3.8374,
+      "step": 1850
+    },
+    {
+      "epoch": 4.05982905982906,
+      "grad_norm": 1.6786645657635046,
+      "learning_rate": 7.905016988984106e-05,
+      "loss": 3.8245,
+      "step": 1900
+    },
+    {
+      "epoch": 4.166666666666667,
+      "grad_norm": 2.2311798939523557,
+      "learning_rate": 7.898142763543867e-05,
+      "loss": 3.8165,
+      "step": 1950
+    },
+    {
+      "epoch": 4.273504273504273,
+      "grad_norm": 1.7909392129453563,
+      "learning_rate": 7.891031815347695e-05,
+      "loss": 3.8081,
+      "step": 2000
+    },
+    {
+      "epoch": 4.380341880341881,
+      "grad_norm": 2.0523668535469204,
+      "learning_rate": 7.883684605814828e-05,
+      "loss": 3.8063,
+      "step": 2050
+    },
+    {
+      "epoch": 4.487179487179487,
+      "grad_norm": 1.874013704927665,
+      "learning_rate": 7.876101611695162e-05,
+      "loss": 3.7996,
+      "step": 2100
+    },
+    {
+      "epoch": 4.594017094017094,
+      "grad_norm": 1.8879502978248355,
+      "learning_rate": 7.868283325038319e-05,
+      "loss": 3.811,
+      "step": 2150
+    },
+    {
+      "epoch": 4.700854700854701,
+      "grad_norm": 1.7754414533299234,
+      "learning_rate": 7.860230253161715e-05,
+      "loss": 3.7997,
+      "step": 2200
+    },
+    {
+      "epoch": 4.8076923076923075,
+      "grad_norm": 1.8465961536252966,
+      "learning_rate": 7.851942918617646e-05,
+      "loss": 3.7968,
+      "step": 2250
+    },
+    {
+      "epoch": 4.914529914529915,
+      "grad_norm": 1.7798643913475303,
+      "learning_rate": 7.843421859159373e-05,
+      "loss": 3.7995,
+      "step": 2300
+    },
+    {
+      "epoch": 5.021367521367521,
+      "grad_norm": 1.6909531477579731,
+      "learning_rate": 7.834667627706234e-05,
+      "loss": 3.7938,
+      "step": 2350
+    },
+    {
+      "epoch": 5.128205128205128,
+      "grad_norm": 1.8289702979878133,
+      "learning_rate": 7.825680792307762e-05,
+      "loss": 3.7808,
+      "step": 2400
+    },
+    {
+      "epoch": 5.235042735042735,
+      "grad_norm": 1.6910478242983937,
+      "learning_rate": 7.816461936106827e-05,
+      "loss": 3.7769,
+      "step": 2450
+    },
+    {
+      "epoch": 5.3418803418803416,
+      "grad_norm": 1.8114942851638463,
+      "learning_rate": 7.807011657301797e-05,
+      "loss": 3.7716,
+      "step": 2500
+    },
+    {
+      "epoch": 5.448717948717949,
+      "grad_norm": 1.6319961332247308,
+      "learning_rate": 7.797330569107721e-05,
+      "loss": 3.7716,
+      "step": 2550
+    },
+    {
+      "epoch": 5.555555555555555,
+      "grad_norm": 1.7220662808660416,
+      "learning_rate": 7.787419299716536e-05,
+      "loss": 3.769,
+      "step": 2600
+    },
+    {
+      "epoch": 5.662393162393163,
+      "grad_norm": 1.8873247077840911,
+      "learning_rate": 7.777278492256316e-05,
+      "loss": 3.7706,
+      "step": 2650
+    },
+    {
+      "epoch": 5.769230769230769,
+      "grad_norm": 1.6976382567518187,
+      "learning_rate": 7.766908804749518e-05,
+      "loss": 3.7624,
+      "step": 2700
+    },
+    {
+      "epoch": 5.8760683760683765,
+      "grad_norm": 1.8676652496639343,
+      "learning_rate": 7.756310910070307e-05,
+      "loss": 3.77,
+      "step": 2750
+    },
+    {
+      "epoch": 5.982905982905983,
+      "grad_norm": 1.712178516145055,
+      "learning_rate": 7.74548549590088e-05,
+      "loss": 3.7639,
+      "step": 2800
+    },
+    {
+      "epoch": 6.089743589743589,
+      "grad_norm": 1.679182535606097,
+      "learning_rate": 7.73443326468685e-05,
+      "loss": 3.7478,
+      "step": 2850
+    },
+    {
+      "epoch": 6.196581196581197,
+      "grad_norm": 1.574526470389441,
+      "learning_rate": 7.723154933591662e-05,
+      "loss": 3.7542,
+      "step": 2900
+    },
+    {
+      "epoch": 6.303418803418803,
+      "grad_norm": 1.6785375872855937,
+      "learning_rate": 7.711651234450059e-05,
+      "loss": 3.7438,
+      "step": 2950
+    },
+    {
+      "epoch": 6.410256410256411,
+      "grad_norm": 1.7945514417084472,
+      "learning_rate": 7.699922913720591e-05,
+      "loss": 3.7441,
+      "step": 3000
+    },
+    {
+      "epoch": 6.517094017094017,
+      "grad_norm": 1.6372117640617172,
+      "learning_rate": 7.687970732437185e-05,
+      "loss": 3.7459,
+      "step": 3050
+    },
+    {
+      "epoch": 6.6239316239316235,
+      "grad_norm": 1.6732118912179677,
+      "learning_rate": 7.675795466159753e-05,
+      "loss": 3.7477,
+      "step": 3100
+    },
+    {
+      "epoch": 6.730769230769231,
+      "grad_norm": 1.6015289303122424,
+      "learning_rate": 7.66339790492388e-05,
+      "loss": 3.738,
+      "step": 3150
+    },
+    {
+      "epoch": 6.837606837606837,
+      "grad_norm": 1.679231377704378,
+      "learning_rate": 7.650778853189546e-05,
+      "loss": 3.7412,
+      "step": 3200
+    },
+    {
+      "epoch": 6.944444444444445,
+      "grad_norm": 1.6926843338267605,
+      "learning_rate": 7.637939129788935e-05,
+      "loss": 3.7447,
+      "step": 3250
+    },
+    {
+      "epoch": 7.051282051282051,
+      "grad_norm": 1.7053205005415435,
+      "learning_rate": 7.624879567873296e-05,
+      "loss": 3.7282,
+      "step": 3300
+    },
+    {
+      "epoch": 7.1581196581196584,
+      "grad_norm": 1.7400948311186981,
+      "learning_rate": 7.611601014858889e-05,
+      "loss": 3.7215,
+      "step": 3350
+    },
+    {
+      "epoch": 7.264957264957265,
+      "grad_norm": 2.528264958264508,
+      "learning_rate": 7.598104332371991e-05,
+      "loss": 3.7196,
+      "step": 3400
+    },
+    {
+      "epoch": 7.371794871794872,
+      "grad_norm": 1.5849521375824331,
+      "learning_rate": 7.584390396192986e-05,
+      "loss": 3.7221,
+      "step": 3450
+    },
+    {
+      "epoch": 7.478632478632479,
+      "grad_norm": 1.7924005277052726,
+      "learning_rate": 7.570460096199537e-05,
+      "loss": 3.7236,
+      "step": 3500
+    },
+    {
+      "epoch": 7.585470085470085,
+      "grad_norm": 1.7867674787154022,
+      "learning_rate": 7.556314336308853e-05,
+      "loss": 3.7173,
+      "step": 3550
+    },
+    {
+      "epoch": 7.6923076923076925,
+      "grad_norm": 1.6452085862488215,
+      "learning_rate": 7.541954034419019e-05,
+      "loss": 3.715,
+      "step": 3600
+    },
+    {
+      "epoch": 7.799145299145299,
+      "grad_norm": 2.100545190250131,
+      "learning_rate": 7.527380122349443e-05,
+      "loss": 3.7039,
+      "step": 3650
+    },
+    {
+      "epoch": 7.905982905982906,
+      "grad_norm": 1.6770874481466609,
+      "learning_rate": 7.512593545780399e-05,
+      "loss": 3.7224,
+      "step": 3700
+    },
+    {
+      "epoch": 8.012820512820513,
+      "grad_norm": 1.7001506486226698,
+      "learning_rate": 7.497595264191646e-05,
+      "loss": 3.7162,
+      "step": 3750
+    },
+    {
+      "epoch": 8.11965811965812,
+      "grad_norm": 1.6817162615240098,
+      "learning_rate": 7.482386250800185e-05,
+      "loss": 3.7046,
+      "step": 3800
+    },
+    {
+      "epoch": 8.226495726495726,
+      "grad_norm": 1.719195360223785,
+      "learning_rate": 7.4669674924971e-05,
+      "loss": 3.695,
+      "step": 3850
+    },
+    {
+      "epoch": 8.333333333333334,
+      "grad_norm": 1.6285706684402375,
+      "learning_rate": 7.451339989783522e-05,
+      "loss": 3.6942,
+      "step": 3900
+    },
+    {
+      "epoch": 8.44017094017094,
+      "grad_norm": 1.7752159712591726,
+      "learning_rate": 7.435504756705705e-05,
+      "loss": 3.692,
+      "step": 3950
+    },
+    {
+      "epoch": 8.547008547008547,
+      "grad_norm": 1.7150172534664192,
+      "learning_rate": 7.419462820789231e-05,
+      "loss": 3.6984,
+      "step": 4000
+    },
+    {
+      "epoch": 8.653846153846153,
+      "grad_norm": 1.5396911576453884,
+      "learning_rate": 7.403215222972335e-05,
+      "loss": 3.6963,
+      "step": 4050
+    },
+    {
+      "epoch": 8.760683760683762,
+      "grad_norm": 1.5283160224829966,
+      "learning_rate": 7.38676301753835e-05,
+      "loss": 3.6999,
+      "step": 4100
+    },
+    {
+      "epoch": 8.867521367521368,
+      "grad_norm": 1.5551933347468931,
+      "learning_rate": 7.370107272047316e-05,
+      "loss": 3.6933,
+      "step": 4150
+    },
+    {
+      "epoch": 8.974358974358974,
+      "grad_norm": 1.7511071381440466,
+      "learning_rate": 7.353249067266687e-05,
+      "loss": 3.6964,
+      "step": 4200
+    },
+    {
+      "epoch": 9.081196581196581,
+      "grad_norm": 1.4171376380434266,
+      "learning_rate": 7.336189497101211e-05,
+      "loss": 3.6785,
+      "step": 4250
+    },
+    {
+      "epoch": 9.188034188034187,
+      "grad_norm": 1.5981957007543055,
+      "learning_rate": 7.318929668521952e-05,
+      "loss": 3.6899,
+      "step": 4300
+    },
+    {
+      "epoch": 9.294871794871796,
+      "grad_norm": 3.0421520699356517,
+      "learning_rate": 7.301470701494449e-05,
+      "loss": 3.6823,
+      "step": 4350
+    },
+    {
+      "epoch": 9.401709401709402,
+      "grad_norm": 1.6095054351392535,
+      "learning_rate": 7.283813728906054e-05,
+      "loss": 3.6824,
+      "step": 4400
+    },
+    {
+      "epoch": 9.508547008547009,
+      "grad_norm": 1.7626031317872317,
+      "learning_rate": 7.265959896492414e-05,
+      "loss": 3.6732,
+      "step": 4450
+    },
+    {
+      "epoch": 9.615384615384615,
+      "grad_norm": 1.6601170434249126,
+      "learning_rate": 7.24791036276313e-05,
+      "loss": 3.6731,
+      "step": 4500
+    },
+    {
+      "epoch": 9.722222222222221,
+      "grad_norm": 1.6596574101192159,
+      "learning_rate": 7.22966629892658e-05,
+      "loss": 3.6719,
+      "step": 4550
+    },
+    {
+      "epoch": 9.82905982905983,
+      "grad_norm": 10.220820132678703,
+      "learning_rate": 7.211228888813919e-05,
+      "loss": 3.6815,
+      "step": 4600
+    },
+    {
+      "epoch": 9.935897435897436,
+      "grad_norm": 1.474086077471036,
+      "learning_rate": 7.19259932880227e-05,
+      "loss": 3.6809,
+      "step": 4650
+    },
+    {
+      "epoch": 10.042735042735043,
+      "grad_norm": 1.5361596630778747,
+      "learning_rate": 7.173778827737085e-05,
+      "loss": 3.6613,
+      "step": 4700
+    },
+    {
+      "epoch": 10.149572649572649,
+      "grad_norm": 1.5694651428853257,
+      "learning_rate": 7.154768606853705e-05,
+      "loss": 3.6593,
+      "step": 4750
+    },
+    {
+      "epoch": 10.256410256410255,
+      "grad_norm": 1.529138627518509,
+      "learning_rate": 7.135569899698122e-05,
+      "loss": 3.655,
+      "step": 4800
+    },
+    {
+      "epoch": 10.363247863247864,
+      "grad_norm": 1.7020816932150489,
+      "learning_rate": 7.116183952046932e-05,
+      "loss": 3.667,
+      "step": 4850
+    },
+    {
+      "epoch": 10.47008547008547,
+      "grad_norm": 1.6247700748638545,
+      "learning_rate": 7.096612021826493e-05,
+      "loss": 3.6613,
+      "step": 4900
+    },
+    {
+      "epoch": 10.576923076923077,
+      "grad_norm": 1.5047790372740069,
+      "learning_rate": 7.076855379031314e-05,
+      "loss": 3.6588,
+      "step": 4950
+    },
+    {
+      "epoch": 10.683760683760683,
+      "grad_norm": 2.084505311018084,
+      "learning_rate": 7.05691530564163e-05,
+      "loss": 3.6553,
+      "step": 5000
+    },
+    {
+      "epoch": 10.790598290598291,
+      "grad_norm": 1.546832536827408,
+      "learning_rate": 7.036793095540228e-05,
+      "loss": 3.6604,
+      "step": 5050
+    },
+    {
+      "epoch": 10.897435897435898,
+      "grad_norm": 1.604609052751935,
+      "learning_rate": 7.01649005442849e-05,
+      "loss": 3.6584,
+      "step": 5100
+    },
+    {
+      "epoch": 11.004273504273504,
+      "grad_norm": 1.485629705887829,
+      "learning_rate": 6.996007499741652e-05,
+      "loss": 3.6538,
+      "step": 5150
+    },
+    {
+      "epoch": 11.11111111111111,
+      "grad_norm": 1.8424475644024803,
+      "learning_rate": 6.975346760563337e-05,
+      "loss": 3.6424,
+      "step": 5200
+    },
+    {
+      "epoch": 11.217948717948717,
+      "grad_norm": 1.6286932716825095,
+      "learning_rate": 6.954509177539306e-05,
+      "loss": 3.6373,
+      "step": 5250
+    },
+    {
+      "epoch": 11.324786324786325,
+      "grad_norm": 1.5150525299886066,
+      "learning_rate": 6.933496102790455e-05,
+      "loss": 3.6433,
+      "step": 5300
+    },
+    {
+      "epoch": 11.431623931623932,
+      "grad_norm": 1.4628099488038697,
+      "learning_rate": 6.912308899825094e-05,
+      "loss": 3.6347,
+      "step": 5350
+    },
+    {
+      "epoch": 11.538461538461538,
+      "grad_norm": 1.5758824797735944,
+      "learning_rate": 6.890948943450462e-05,
+      "loss": 3.6372,
+      "step": 5400
+    },
+    {
+      "epoch": 11.645299145299145,
+      "grad_norm": 1.5210645926590014,
+      "learning_rate": 6.869417619683515e-05,
+      "loss": 3.6398,
+      "step": 5450
+    },
+    {
+      "epoch": 11.752136752136753,
+      "grad_norm": 1.4454661519624443,
+      "learning_rate": 6.847716325660999e-05,
+      "loss": 3.6382,
+      "step": 5500
+    },
+    {
+      "epoch": 11.85897435897436,
+      "grad_norm": 1.7686322014608322,
+      "learning_rate": 6.825846469548783e-05,
+      "loss": 3.645,
+      "step": 5550
+    },
+    {
+      "epoch": 11.965811965811966,
+      "grad_norm": 1.6215030750877166,
+      "learning_rate": 6.80380947045049e-05,
+      "loss": 3.6488,
+      "step": 5600
+    },
+    {
+      "epoch": 12.072649572649572,
+      "grad_norm": 1.5454385426401656,
+      "learning_rate": 6.781606758315411e-05,
+      "loss": 3.6292,
+      "step": 5650
+    },
+    {
+      "epoch": 12.179487179487179,
+      "grad_norm": 1.5219417956848793,
+      "learning_rate": 6.759239773845719e-05,
+      "loss": 3.6307,
+      "step": 5700
+    },
+    {
+      "epoch": 12.286324786324787,
+      "grad_norm": 1.5542005226736204,
+      "learning_rate": 6.736709968402982e-05,
+      "loss": 3.6293,
+      "step": 5750
+    },
+    {
+      "epoch": 12.393162393162394,
+      "grad_norm": 1.623986735247479,
+      "learning_rate": 6.714018803913991e-05,
+      "loss": 3.6219,
+      "step": 5800
+    },
+    {
+      "epoch": 12.5,
+      "grad_norm": 1.5195039486822781,
+      "learning_rate": 6.69116775277589e-05,
+      "loss": 3.6295,
+      "step": 5850
+    },
+    {
+      "epoch": 12.606837606837606,
+      "grad_norm": 1.594705930764426,
+      "learning_rate": 6.66815829776064e-05,
+      "loss": 3.6336,
+      "step": 5900
+    },
+    {
+      "epoch": 12.713675213675213,
+      "grad_norm": 1.480108373256865,
+      "learning_rate": 6.644991931918805e-05,
+      "loss": 3.6184,
+      "step": 5950
+    },
+    {
+      "epoch": 12.820512820512821,
+      "grad_norm": 1.5002000993141473,
+      "learning_rate": 6.621670158482664e-05,
+      "loss": 3.6275,
+      "step": 6000
+    },
+    {
+      "epoch": 12.927350427350428,
+      "grad_norm": 1.5712595498850146,
+      "learning_rate": 6.598194490768677e-05,
+      "loss": 3.6197,
+      "step": 6050
+    },
+    {
+      "epoch": 13.034188034188034,
+      "grad_norm": 1.5018502585873035,
+      "learning_rate": 6.574566452079276e-05,
+      "loss": 3.6143,
+      "step": 6100
+    },
+    {
+      "epoch": 13.14102564102564,
+      "grad_norm": 1.49259152360227,
+      "learning_rate": 6.550787575604034e-05,
+      "loss": 3.6135,
+      "step": 6150
+    },
+    {
+      "epoch": 13.247863247863247,
+      "grad_norm": 1.5699547507210274,
+      "learning_rate": 6.526859404320169e-05,
+      "loss": 3.6129,
+      "step": 6200
+    },
+    {
+      "epoch": 13.354700854700855,
+      "grad_norm": 1.5110194439447986,
+      "learning_rate": 6.502783490892421e-05,
+      "loss": 3.6102,
+      "step": 6250
+    },
+    {
+      "epoch": 13.461538461538462,
+      "grad_norm": 1.5863106575168338,
+      "learning_rate": 6.478561397572317e-05,
+      "loss": 3.6108,
+      "step": 6300
+    },
+    {
+      "epoch": 13.568376068376068,
+      "grad_norm": 1.4631757267400136,
+      "learning_rate": 6.454194696096775e-05,
+      "loss": 3.6127,
+      "step": 6350
+    },
+    {
+      "epoch": 13.675213675213675,
+      "grad_norm": 1.4900676589043844,
+      "learning_rate": 6.429684967586138e-05,
+      "loss": 3.6073,
+      "step": 6400
+    },
+    {
+      "epoch": 13.782051282051283,
+      "grad_norm": 2.079377678794041,
+      "learning_rate": 6.405033802441565e-05,
+      "loss": 3.6039,
+      "step": 6450
+    },
+    {
+      "epoch": 13.88888888888889,
+      "grad_norm": 1.5029050194176434,
+      "learning_rate": 6.380242800241835e-05,
+      "loss": 3.6211,
+      "step": 6500
+    },
+    {
+      "epoch": 13.995726495726496,
+      "grad_norm": 1.518341540130534,
+      "learning_rate": 6.355313569639555e-05,
+      "loss": 3.6128,
+      "step": 6550
+    },
+    {
+      "epoch": 14.102564102564102,
+      "grad_norm": 1.5345576626262458,
+      "learning_rate": 6.330247728256771e-05,
+      "loss": 3.5815,
+      "step": 6600
+    },
+    {
+      "epoch": 14.209401709401709,
+      "grad_norm": 1.4705033084244592,
+      "learning_rate": 6.305046902580013e-05,
+      "loss": 3.5873,
+      "step": 6650
+    },
+    {
+      "epoch": 14.316239316239317,
+      "grad_norm": 1.5238057009473032,
+      "learning_rate": 6.279712727854743e-05,
+      "loss": 3.5935,
+      "step": 6700
+    },
+    {
+      "epoch": 14.423076923076923,
+      "grad_norm": 1.5620920793205895,
+      "learning_rate": 6.254246847979254e-05,
+      "loss": 3.6014,
+      "step": 6750
+    },
+    {
+      "epoch": 14.52991452991453,
+      "grad_norm": 1.547710770903625,
+      "learning_rate": 6.228650915397998e-05,
+      "loss": 3.5993,
+      "step": 6800
+    },
+    {
+      "epoch": 14.636752136752136,
+      "grad_norm": 1.5496045315673772,
+      "learning_rate": 6.20292659099436e-05,
+      "loss": 3.5986,
+      "step": 6850
+    },
+    {
+      "epoch": 14.743589743589745,
+      "grad_norm": 1.5740526741210388,
+      "learning_rate": 6.177075543982883e-05,
+      "loss": 3.5988,
+      "step": 6900
+    },
+    {
+      "epoch": 14.850427350427351,
+      "grad_norm": 1.5398680616808194,
+      "learning_rate": 6.151099451800965e-05,
+      "loss": 3.5886,
+      "step": 6950
+    },
+    {
+      "epoch": 14.957264957264957,
+      "grad_norm": 1.6259731166677076,
+      "learning_rate": 6.125000000000001e-05,
+      "loss": 3.591,
+      "step": 7000
+    },
+    {
+      "epoch": 15.064102564102564,
+      "grad_norm": 1.4806043416840873,
+      "learning_rate": 6.0987788821360164e-05,
+      "loss": 3.5731,
+      "step": 7050
+    },
+    {
+      "epoch": 15.17094017094017,
+      "grad_norm": 1.5939703396343463,
+      "learning_rate": 6.0724377996597723e-05,
+      "loss": 3.5764,
+      "step": 7100
+    },
+    {
+      "epoch": 15.277777777777779,
+      "grad_norm": 1.4202060336422229,
+      "learning_rate": 6.045978461806366e-05,
+      "loss": 3.5845,
+      "step": 7150
+    },
+    {
+      "epoch": 15.384615384615385,
+      "grad_norm": 1.4775589098861128,
+      "learning_rate": 6.019402585484308e-05,
+      "loss": 3.571,
+      "step": 7200
+    },
+    {
+      "epoch": 15.491452991452991,
+      "grad_norm": 1.5630724811844643,
+      "learning_rate": 5.992711895164133e-05,
+      "loss": 3.5808,
+      "step": 7250
+    },
+    {
+      "epoch": 15.598290598290598,
+      "grad_norm": 1.5543909604764383,
+      "learning_rate": 5.965908122766483e-05,
+      "loss": 3.5753,
+      "step": 7300
+    },
+    {
+      "epoch": 15.705128205128204,
+      "grad_norm": 1.5465008832335148,
+      "learning_rate": 5.938993007549739e-05,
+      "loss": 3.5723,
+      "step": 7350
+    },
+    {
+      "epoch": 15.811965811965813,
+      "grad_norm": 1.4743809807346027,
+      "learning_rate": 5.9119682959971563e-05,
+      "loss": 3.578,
+      "step": 7400
+    },
+    {
+      "epoch": 15.918803418803419,
+      "grad_norm": 1.4493286420178282,
+      "learning_rate": 5.884835741703538e-05,
+      "loss": 3.5727,
+      "step": 7450
+    },
+    {
+      "epoch": 16.025641025641026,
+      "grad_norm": 1.436645336671343,
+      "learning_rate": 5.857597105261453e-05,
+      "loss": 3.571,
+      "step": 7500
+    },
+    {
+      "epoch": 16.132478632478634,
+      "grad_norm": 1.5383062683209867,
+      "learning_rate": 5.8302541541469855e-05,
+      "loss": 3.5543,
+      "step": 7550
+    },
+    {
+      "epoch": 16.23931623931624,
+      "grad_norm": 1.5681746624668993,
+      "learning_rate": 5.8028086626050457e-05,
+      "loss": 3.5595,
+      "step": 7600
+    },
+    {
+      "epoch": 16.346153846153847,
+      "grad_norm": 1.5410354284215502,
+      "learning_rate": 5.775262411534252e-05,
+      "loss": 3.5588,
+      "step": 7650
+    },
+    {
+      "epoch": 16.45299145299145,
+      "grad_norm": 1.5203964006378836,
+      "learning_rate": 5.747617188371358e-05,
+      "loss": 3.5569,
+      "step": 7700
+    },
+    {
+      "epoch": 16.55982905982906,
+      "grad_norm": 1.4817981236989008,
+      "learning_rate": 5.7198747869752824e-05,
+      "loss": 3.5629,
+      "step": 7750
+    },
+    {
+      "epoch": 16.666666666666668,
+      "grad_norm": 1.500308799747035,
+      "learning_rate": 5.692037007510691e-05,
+      "loss": 3.5596,
+      "step": 7800
+    },
+    {
+      "epoch": 16.773504273504273,
+      "grad_norm": 1.7695298352772697,
+      "learning_rate": 5.6641056563312026e-05,
+      "loss": 3.5527,
+      "step": 7850
+    },
+    {
+      "epoch": 16.88034188034188,
+      "grad_norm": 1.5340102094238413,
+      "learning_rate": 5.6360825458621665e-05,
+      "loss": 3.5622,
+      "step": 7900
+    },
+    {
+      "epoch": 16.98717948717949,
+      "grad_norm": 1.5412854247341883,
+      "learning_rate": 5.6079694944830634e-05,
+      "loss": 3.5621,
+      "step": 7950
+    },
+    {
+      "epoch": 17.094017094017094,
+      "grad_norm": 1.5584998998853048,
+      "learning_rate": 5.579768326409509e-05,
+      "loss": 3.54,
+      "step": 8000
+    },
+    {
+      "epoch": 17.200854700854702,
+      "grad_norm": 1.4673715169312684,
+      "learning_rate": 5.5514808715748866e-05,
+      "loss": 3.5478,
+      "step": 8050
+    },
+    {
+      "epoch": 17.307692307692307,
+      "grad_norm": 1.8376102051157188,
+      "learning_rate": 5.523108965511602e-05,
+      "loss": 3.5396,
+      "step": 8100
+    },
+    {
+      "epoch": 17.414529914529915,
+      "grad_norm": 1.4166786903918271,
+      "learning_rate": 5.494654449231982e-05,
+      "loss": 3.5426,
+      "step": 8150
+    },
+    {
+      "epoch": 17.521367521367523,
+      "grad_norm": 1.46334786902459,
+      "learning_rate": 5.466119169108811e-05,
+      "loss": 3.5411,
+      "step": 8200
+    },
+    {
+      "epoch": 17.628205128205128,
+      "grad_norm": 1.484842407521401,
+      "learning_rate": 5.437504976755523e-05,
+      "loss": 3.5372,
+      "step": 8250
+    },
+    {
+      "epoch": 17.735042735042736,
+      "grad_norm": 1.464526739786086,
+      "learning_rate": 5.408813728906053e-05,
+      "loss": 3.5437,
+      "step": 8300
+    },
+    {
+      "epoch": 17.84188034188034,
+      "grad_norm": 1.4706425749332082,
+      "learning_rate": 5.380047287294361e-05,
+      "loss": 3.5462,
+      "step": 8350
+    },
+    {
+      "epoch": 17.94871794871795,
+      "grad_norm": 1.5282873960612633,
+      "learning_rate": 5.351207518533616e-05,
+      "loss": 3.5453,
+      "step": 8400
+    },
+    {
+      "epoch": 18.055555555555557,
+      "grad_norm": 1.553954215098199,
+      "learning_rate": 5.322296293995085e-05,
+      "loss": 3.5261,
+      "step": 8450
+    },
+    {
+      "epoch": 18.162393162393162,
+      "grad_norm": 1.460026176296364,
+      "learning_rate": 5.293315489686698e-05,
+      "loss": 3.5165,
+      "step": 8500
+    },
+    {
+      "epoch": 18.26923076923077,
+      "grad_norm": 1.4489055613879473,
+      "learning_rate": 5.264266986131315e-05,
+      "loss": 3.5263,
+      "step": 8550
+    },
+    {
+      "epoch": 18.376068376068375,
+      "grad_norm": 1.5224175584643513,
+      "learning_rate": 5.235152668244701e-05,
+      "loss": 3.5297,
+      "step": 8600
+    },
+    {
+      "epoch": 18.482905982905983,
+      "grad_norm": 1.4655236052554852,
+      "learning_rate": 5.205974425213225e-05,
+      "loss": 3.524,
+      "step": 8650
+    },
+    {
+      "epoch": 18.58974358974359,
+      "grad_norm": 1.4634201252745132,
+      "learning_rate": 5.176734150371261e-05,
+      "loss": 3.5258,
+      "step": 8700
+    },
+    {
+      "epoch": 18.696581196581196,
+      "grad_norm": 1.4580158251307793,
+      "learning_rate": 5.147433741078342e-05,
+      "loss": 3.5203,
+      "step": 8750
+    },
+    {
+      "epoch": 18.803418803418804,
+      "grad_norm": 1.526325559628089,
+      "learning_rate": 5.118075098596038e-05,
+      "loss": 3.5241,
+      "step": 8800
+    },
+    {
+      "epoch": 18.91025641025641,
+      "grad_norm": 1.4073654836876215,
+      "learning_rate": 5.0886601279645845e-05,
+      "loss": 3.5104,
+      "step": 8850
+    },
+    {
+      "epoch": 19.017094017094017,
+      "grad_norm": 1.6261138766390175,
+      "learning_rate": 5.059190737879274e-05,
+      "loss": 3.513,
+      "step": 8900
+    },
+    {
+      "epoch": 19.123931623931625,
+      "grad_norm": 1.5122999705305489,
+      "learning_rate": 5.029668840566597e-05,
+      "loss": 3.5047,
+      "step": 8950
+    },
+    {
+      "epoch": 19.23076923076923,
+      "grad_norm": 1.4420542513870556,
+      "learning_rate": 5.000096351660167e-05,
+      "loss": 3.4968,
+      "step": 9000
+    },
+    {
+      "epoch": 19.337606837606838,
+      "grad_norm": 1.493772453873553,
+      "learning_rate": 4.970475190076407e-05,
+      "loss": 3.498,
+      "step": 9050
+    },
+    {
+      "epoch": 19.444444444444443,
+      "grad_norm": 1.545900673783324,
+      "learning_rate": 4.940807277890043e-05,
+      "loss": 3.4975,
+      "step": 9100
+    },
+    {
+      "epoch": 19.55128205128205,
+      "grad_norm": 1.474713414039892,
+      "learning_rate": 4.9110945402093846e-05,
+      "loss": 3.5058,
+      "step": 9150
+    },
+    {
+      "epoch": 19.65811965811966,
+      "grad_norm": 1.499848755523026,
+      "learning_rate": 4.881338905051394e-05,
+      "loss": 3.4989,
+      "step": 9200
+    },
+    {
+      "epoch": 19.764957264957264,
+      "grad_norm": 1.5022062130566565,
+      "learning_rate": 4.851542303216601e-05,
+      "loss": 3.5071,
+      "step": 9250
+    },
+    {
+      "epoch": 19.871794871794872,
+      "grad_norm": 1.4286746413547837,
+      "learning_rate": 4.8217066681637957e-05,
+      "loss": 3.5038,
+      "step": 9300
+    },
+    {
+      "epoch": 19.978632478632477,
+      "grad_norm": 1.443317748257993,
+      "learning_rate": 4.7918339358845805e-05,
+      "loss": 3.5052,
+      "step": 9350
+    },
+    {
+      "epoch": 20.085470085470085,
+      "grad_norm": 1.4168322503801087,
+      "learning_rate": 4.7619260447777446e-05,
+      "loss": 3.4788,
+      "step": 9400
+    },
+    {
+      "epoch": 20.192307692307693,
+      "grad_norm": 1.5714516374827268,
+      "learning_rate": 4.7319849355234804e-05,
+      "loss": 3.4783,
+      "step": 9450
+    },
+    {
+      "epoch": 20.299145299145298,
+      "grad_norm": 1.5603822470902127,
+      "learning_rate": 4.702012550957461e-05,
+      "loss": 3.4893,
+      "step": 9500
+    },
+    {
+      "epoch": 20.405982905982906,
+      "grad_norm": 1.4623232710258578,
+      "learning_rate": 4.672010835944771e-05,
+      "loss": 3.4892,
+      "step": 9550
+    },
+    {
+      "epoch": 20.51282051282051,
+      "grad_norm": 1.5754771160859642,
+      "learning_rate": 4.6419817372537015e-05,
+      "loss": 3.4803,
+      "step": 9600
+    },
+    {
+      "epoch": 20.61965811965812,
+      "grad_norm": 1.516101722857511,
+      "learning_rate": 4.6119272034294325e-05,
+      "loss": 3.4808,
+      "step": 9650
+    },
+    {
+      "epoch": 20.726495726495727,
+      "grad_norm": 1.591646788840805,
+      "learning_rate": 4.5818491846675944e-05,
+      "loss": 3.486,
+      "step": 9700
+    },
+    {
+      "epoch": 20.833333333333332,
+      "grad_norm": 1.5392104334340817,
+      "learning_rate": 4.551749632687723e-05,
+      "loss": 3.487,
+      "step": 9750
+    },
+    {
+      "epoch": 20.94017094017094,
+      "grad_norm": 1.510393848466065,
+      "learning_rate": 4.52163050060661e-05,
+      "loss": 3.49,
+      "step": 9800
+    },
+    {
+      "epoch": 21.04700854700855,
+      "grad_norm": 1.5683979110448965,
+      "learning_rate": 4.491493742811573e-05,
+      "loss": 3.4729,
+      "step": 9850
+    },
+    {
+      "epoch": 21.153846153846153,
+      "grad_norm": 1.49609375,
+      "learning_rate": 4.461341314833641e-05,
+      "loss": 3.4667,
+      "step": 9900
+    },
+    {
+      "epoch": 21.26068376068376,
+      "grad_norm": 1.591729323001061,
+      "learning_rate": 4.431175173220652e-05,
+      "loss": 3.454,
+      "step": 9950
+    },
+    {
+      "epoch": 21.367521367521366,
+      "grad_norm": 1.4498777305089914,
+      "learning_rate": 4.400997275410307e-05,
+      "loss": 3.4678,
+      "step": 10000
+    },
+    {
+      "epoch": 21.474358974358974,
+      "grad_norm": 1.5371081341322819,
+      "learning_rate": 4.3708095796031497e-05,
+      "loss": 3.4591,
+      "step": 10050
+    },
+    {
+      "epoch": 21.581196581196583,
+      "grad_norm": 1.5191382363247403,
+      "learning_rate": 4.340614044635496e-05,
+      "loss": 3.4687,
+      "step": 10100
+    },
+    {
+      "epoch": 21.688034188034187,
+      "grad_norm": 1.6543824084800223,
+      "learning_rate": 4.3104126298523424e-05,
+      "loss": 3.4513,
+      "step": 10150
+    },
+    {
+      "epoch": 21.794871794871796,
+      "grad_norm": 1.5503952353080468,
+      "learning_rate": 4.280207294980212e-05,
+      "loss": 3.4601,
+      "step": 10200
+    },
+    {
+      "epoch": 21.9017094017094,
+      "grad_norm": 1.6721050068584753,
+      "learning_rate": 4.25e-05,
+      "loss": 3.4676,
+      "step": 10250
+    },
+    {
+      "epoch": 22.00854700854701,
+      "grad_norm": 1.534759314862195,
+      "learning_rate": 4.219792705019789e-05,
+      "loss": 3.4573,
+      "step": 10300
+    },
+    {
+      "epoch": 22.115384615384617,
+      "grad_norm": 1.6067637026570358,
+      "learning_rate": 4.189587370147658e-05,
+      "loss": 3.4376,
+      "step": 10350
+    },
+    {
+      "epoch": 22.22222222222222,
+      "grad_norm": 1.5281042366723503,
+      "learning_rate": 4.1593859553645044e-05,
+      "loss": 3.4359,
+      "step": 10400
+    },
+    {
+      "epoch": 22.32905982905983,
+      "grad_norm": 1.5541495921379098,
+      "learning_rate": 4.129190420396853e-05,
+      "loss": 3.4464,
+      "step": 10450
+    },
+    {
+      "epoch": 22.435897435897434,
+      "grad_norm": 1.5269639891879354,
+      "learning_rate": 4.099002724589694e-05,
+      "loss": 3.4484,
+      "step": 10500
+    },
+    {
+      "epoch": 22.542735042735043,
+      "grad_norm": 1.5232774699351355,
+      "learning_rate": 4.0688248267793484e-05,
+      "loss": 3.4422,
+      "step": 10550
+    },
+    {
+      "epoch": 22.64957264957265,
+      "grad_norm": 1.5252723794548009,
+      "learning_rate": 4.0386586851663594e-05,
+      "loss": 3.4402,
+      "step": 10600
+    },
+    {
+      "epoch": 22.756410256410255,
+      "grad_norm": 1.5220017147380378,
+      "learning_rate": 4.008506257188428e-05,
+      "loss": 3.4424,
+      "step": 10650
+    },
+    {
+      "epoch": 22.863247863247864,
+      "grad_norm": 1.446123380269863,
+      "learning_rate": 3.978369499393392e-05,
+      "loss": 3.4518,
+      "step": 10700
+    },
+    {
+      "epoch": 22.97008547008547,
+      "grad_norm": 1.4734617099815863,
+      "learning_rate": 3.948250367312279e-05,
+      "loss": 3.4402,
+      "step": 10750
+    },
+    {
+      "epoch": 23.076923076923077,
+      "grad_norm": 1.500255086189984,
+      "learning_rate": 3.918150815332407e-05,
+      "loss": 3.4211,
+      "step": 10800
+    },
+    {
+      "epoch": 23.183760683760685,
+      "grad_norm": 1.5683304153371995,
+      "learning_rate": 3.888072796570568e-05,
+      "loss": 3.42,
+      "step": 10850
+    },
+    {
+      "epoch": 23.29059829059829,
+      "grad_norm": 1.597209690800403,
+      "learning_rate": 3.8580182627463004e-05,
+      "loss": 3.4182,
+      "step": 10900
+    },
+    {
+      "epoch": 23.397435897435898,
+      "grad_norm": 1.548782020835125,
+      "learning_rate": 3.8279891640552296e-05,
+      "loss": 3.4128,
+      "step": 10950
+    },
+    {
+      "epoch": 23.504273504273506,
+      "grad_norm": 1.5510359778298841,
+      "learning_rate": 3.7979874490425396e-05,
+      "loss": 3.4189,
+      "step": 11000
+    },
+    {
+      "epoch": 23.61111111111111,
+      "grad_norm": 1.4877303097447068,
+      "learning_rate": 3.768015064476521e-05,
+      "loss": 3.413,
+      "step": 11050
+    },
+    {
+      "epoch": 23.71794871794872,
+      "grad_norm": 1.4998179961094804,
+      "learning_rate": 3.7380739552222574e-05,
+      "loss": 3.4195,
+      "step": 11100
+    },
+    {
+      "epoch": 23.824786324786324,
+      "grad_norm": 1.6276035725910374,
+      "learning_rate": 3.7081660641154215e-05,
+      "loss": 3.4181,
+      "step": 11150
+    },
+    {
+      "epoch": 23.931623931623932,
+      "grad_norm": 1.607925058174758,
+      "learning_rate": 3.678293331836205e-05,
+      "loss": 3.4212,
+      "step": 11200
+    },
+    {
+      "epoch": 24.03846153846154,
+      "grad_norm": 1.7136000031007326,
+      "learning_rate": 3.648457696783399e-05,
+      "loss": 3.4093,
+      "step": 11250
+    },
+    {
+      "epoch": 24.145299145299145,
+      "grad_norm": 1.6337873843059596,
+      "learning_rate": 3.618661094948606e-05,
+      "loss": 3.3832,
+      "step": 11300
+    },
+    {
+      "epoch": 24.252136752136753,
+      "grad_norm": 1.6296652802427443,
+      "learning_rate": 3.588905459790617e-05,
+      "loss": 3.3787,
+      "step": 11350
+    },
+    {
+      "epoch": 24.358974358974358,
+      "grad_norm": 1.4873240839694675,
+      "learning_rate": 3.5591927221099575e-05,
+      "loss": 3.3954,
+      "step": 11400
+    },
+    {
+      "epoch": 24.465811965811966,
+      "grad_norm": 1.5356828014656962,
+      "learning_rate": 3.529524809923594e-05,
+      "loss": 3.3886,
+      "step": 11450
+    },
+    {
+      "epoch": 24.572649572649574,
+      "grad_norm": 1.6430980390249932,
+      "learning_rate": 3.499903648339834e-05,
+      "loss": 3.3932,
+      "step": 11500
+    },
+    {
+      "epoch": 24.67948717948718,
+      "grad_norm": 1.5788207833678471,
+      "learning_rate": 3.470331159433404e-05,
+      "loss": 3.3857,
+      "step": 11550
+    },
+    {
+      "epoch": 24.786324786324787,
+      "grad_norm": 1.7653838811655183,
+      "learning_rate": 3.440809262120728e-05,
+      "loss": 3.4,
+      "step": 11600
+    },
+    {
+      "epoch": 24.89316239316239,
+      "grad_norm": 1.7491130624850513,
+      "learning_rate": 3.411339872035418e-05,
+      "loss": 3.3933,
+      "step": 11650
+    },
+    {
+      "epoch": 25.0,
+      "grad_norm": 1.6409601505004305,
+      "learning_rate": 3.381924901403964e-05,
+      "loss": 3.3935,
+      "step": 11700
+    },
+    {
+      "epoch": 25.10683760683761,
+      "grad_norm": 1.606820532856031,
+      "learning_rate": 3.35256625892166e-05,
+      "loss": 3.3522,
+      "step": 11750
+    },
+    {
+      "epoch": 25.213675213675213,
+      "grad_norm": 1.6564514469500473,
+      "learning_rate": 3.323265849628739e-05,
+      "loss": 3.3691,
+      "step": 11800
+    },
+    {
+      "epoch": 25.32051282051282,
+      "grad_norm": 1.6048468431738794,
+      "learning_rate": 3.294025574786775e-05,
+      "loss": 3.3633,
+      "step": 11850
+    },
+    {
+      "epoch": 25.427350427350426,
+      "grad_norm": 1.7161740506697982,
+      "learning_rate": 3.2648473317552994e-05,
+      "loss": 3.3663,
+      "step": 11900
+    },
+    {
+      "epoch": 25.534188034188034,
+      "grad_norm": 1.5546586690558504,
+      "learning_rate": 3.2357330138686864e-05,
+      "loss": 3.3628,
+      "step": 11950
+    },
+    {
+      "epoch": 25.641025641025642,
+      "grad_norm": 1.582958486589817,
+      "learning_rate": 3.206684510313303e-05,
+      "loss": 3.3603,
+      "step": 12000
+    },
+    {
+      "epoch": 25.747863247863247,
+      "grad_norm": 1.542709599231902,
+      "learning_rate": 3.1777037060049166e-05,
+      "loss": 3.3655,
+      "step": 12050
+    },
+    {
+      "epoch": 25.854700854700855,
+      "grad_norm": 1.6117766154964244,
+      "learning_rate": 3.1487924814663854e-05,
+      "loss": 3.3727,
+      "step": 12100
+    },
+    {
+      "epoch": 25.96153846153846,
+      "grad_norm": 1.7027238845939607,
+      "learning_rate": 3.119952712705641e-05,
+      "loss": 3.3723,
+      "step": 12150
+    },
+    {
+      "epoch": 26.068376068376068,
+      "grad_norm": 1.7056235088986273,
+      "learning_rate": 3.091186271093947e-05,
+      "loss": 3.3468,
+      "step": 12200
+    },
+    {
+      "epoch": 26.175213675213676,
+      "grad_norm": 1.6578623283297151,
+      "learning_rate": 3.062495023244479e-05,
+      "loss": 3.3408,
+      "step": 12250
+    },
+    {
+      "epoch": 26.28205128205128,
+      "grad_norm": 1.6672004877959212,
+      "learning_rate": 3.0338808308911903e-05,
+      "loss": 3.3351,
+      "step": 12300
+    },
+    {
+      "epoch": 26.38888888888889,
+      "grad_norm": 1.725424136664808,
+      "learning_rate": 3.0053455507680195e-05,
+      "loss": 3.3385,
+      "step": 12350
+    },
+    {
+      "epoch": 26.495726495726494,
+      "grad_norm": 1.6013693344114848,
+      "learning_rate": 2.9768910344883982e-05,
+      "loss": 3.3414,
+      "step": 12400
+    },
+    {
+      "epoch": 26.602564102564102,
+      "grad_norm": 1.6507989682981181,
+      "learning_rate": 2.9485191284251144e-05,
+      "loss": 3.3418,
+      "step": 12450
+    },
+    {
+      "epoch": 26.70940170940171,
+      "grad_norm": 1.7581686379678405,
+      "learning_rate": 2.9202316735904915e-05,
+      "loss": 3.347,
+      "step": 12500
+    },
+    {
+      "epoch": 26.816239316239315,
+      "grad_norm": 1.6879890580930017,
+      "learning_rate": 2.892030505516938e-05,
+      "loss": 3.329,
+      "step": 12550
+    },
+    {
+      "epoch": 26.923076923076923,
+      "grad_norm": 1.6792107903652251,
+      "learning_rate": 2.863917454137834e-05,
+      "loss": 3.3429,
+      "step": 12600
+    },
+    {
+      "epoch": 27.02991452991453,
+      "grad_norm": 1.7124059637383322,
+      "learning_rate": 2.8358943436687987e-05,
+      "loss": 3.3266,
+      "step": 12650
+    },
+    {
+      "epoch": 27.136752136752136,
+      "grad_norm": 1.7334326168505239,
+      "learning_rate": 2.80796299248931e-05,
+      "loss": 3.3149,
+      "step": 12700
+    },
+    {
+      "epoch": 27.243589743589745,
+      "grad_norm": 1.7176568456381722,
+      "learning_rate": 2.7801252130247195e-05,
+      "loss": 3.3212,
+      "step": 12750
+    },
+    {
+      "epoch": 27.35042735042735,
+      "grad_norm": 1.7570421353380783,
+      "learning_rate": 2.7523828116286425e-05,
+      "loss": 3.3191,
+      "step": 12800
+    },
+    {
+      "epoch": 27.457264957264957,
+      "grad_norm": 1.7448672226066309,
+      "learning_rate": 2.7247375884657498e-05,
+      "loss": 3.3158,
+      "step": 12850
+    },
+    {
+      "epoch": 27.564102564102566,
+      "grad_norm": 1.6265664253590588,
+      "learning_rate": 2.697191337394956e-05,
+      "loss": 3.3184,
+      "step": 12900
+    },
+    {
+      "epoch": 27.67094017094017,
+      "grad_norm": 1.7674312298681476,
+      "learning_rate": 2.6697458458530164e-05,
+      "loss": 3.3068,
+      "step": 12950
+    },
+    {
+      "epoch": 27.77777777777778,
+      "grad_norm": 1.7281653097738896,
+      "learning_rate": 2.6424028947385463e-05,
+      "loss": 3.305,
+      "step": 13000
+    },
+    {
+      "epoch": 27.884615384615383,
+      "grad_norm": 1.6757655998868577,
+      "learning_rate": 2.615164258296462e-05,
+      "loss": 3.3151,
+      "step": 13050
+    },
+    {
+      "epoch": 27.99145299145299,
+      "grad_norm": 1.6927031375120576,
+      "learning_rate": 2.5880317040028456e-05,
+      "loss": 3.3091,
+      "step": 13100
+    },
+    {
+      "epoch": 28.0982905982906,
+      "grad_norm": 1.7454895749333972,
+      "learning_rate": 2.5610069924502617e-05,
+      "loss": 3.279,
+      "step": 13150
+    },
+    {
+      "epoch": 28.205128205128204,
+      "grad_norm": 1.8316623991528065,
+      "learning_rate": 2.534091877233518e-05,
+      "loss": 3.2811,
+      "step": 13200
+    },
+    {
+      "epoch": 28.311965811965813,
+      "grad_norm": 1.725868258132857,
+      "learning_rate": 2.5072881048358683e-05,
+      "loss": 3.2789,
+      "step": 13250
+    },
+    {
+      "epoch": 28.418803418803417,
+      "grad_norm": 1.6963201244367443,
+      "learning_rate": 2.4805974145156925e-05,
+      "loss": 3.2826,
+      "step": 13300
+    },
+    {
+      "epoch": 28.525641025641026,
+      "grad_norm": 1.742348291478239,
+      "learning_rate": 2.4540215381936355e-05,
+      "loss": 3.2826,
+      "step": 13350
+    },
+    {
+      "epoch": 28.632478632478634,
+      "grad_norm": 1.8214431329353475,
+      "learning_rate": 2.4275622003402272e-05,
+      "loss": 3.2902,
+      "step": 13400
+    },
+    {
+      "epoch": 28.73931623931624,
+      "grad_norm": 1.7553417558831383,
+      "learning_rate": 2.4012211178639852e-05,
+      "loss": 3.2872,
+      "step": 13450
+    },
+    {
+      "epoch": 28.846153846153847,
+      "grad_norm": 1.8105617551387587,
+      "learning_rate": 2.375000000000001e-05,
+      "loss": 3.2916,
+      "step": 13500
+    },
+    {
+      "epoch": 28.95299145299145,
+      "grad_norm": 1.7855238921981962,
+      "learning_rate": 2.348900548199037e-05,
+      "loss": 3.2871,
+      "step": 13550
+    },
+    {
+      "epoch": 29.05982905982906,
+      "grad_norm": 1.7871660734260615,
+      "learning_rate": 2.322924456017118e-05,
+      "loss": 3.2676,
+      "step": 13600
+    },
+    {
+      "epoch": 29.166666666666668,
+      "grad_norm": 1.8967670265399914,
+      "learning_rate": 2.2970734090056414e-05,
+      "loss": 3.2382,
+      "step": 13650
+    },
+    {
+      "epoch": 29.273504273504273,
+      "grad_norm": 1.7868169826763705,
+      "learning_rate": 2.271349084602002e-05,
+      "loss": 3.2517,
+      "step": 13700
+    },
+    {
+      "epoch": 29.38034188034188,
+      "grad_norm": 1.9317444868871811,
+      "learning_rate": 2.245753152020746e-05,
+      "loss": 3.2592,
+      "step": 13750
+    },
+    {
+      "epoch": 29.487179487179485,
+      "grad_norm": 1.8694439903902382,
+      "learning_rate": 2.2202872721452575e-05,
+      "loss": 3.2402,
+      "step": 13800
+    },
+    {
+      "epoch": 29.594017094017094,
+      "grad_norm": 1.907871807154451,
+      "learning_rate": 2.1949530974199887e-05,
+      "loss": 3.2545,
+      "step": 13850
+    },
+    {
+      "epoch": 29.700854700854702,
+      "grad_norm": 1.809374515091958,
+      "learning_rate": 2.1697522717432292e-05,
+      "loss": 3.247,
+      "step": 13900
+    },
+    {
+      "epoch": 29.807692307692307,
+      "grad_norm": 1.805345291545111,
+      "learning_rate": 2.144686430360447e-05,
+      "loss": 3.2506,
+      "step": 13950
+    },
+    {
+      "epoch": 29.914529914529915,
+      "grad_norm": 1.797933648104017,
+      "learning_rate": 2.1197571997581665e-05,
+      "loss": 3.2524,
+      "step": 14000
+    },
+    {
+      "epoch": 30.021367521367523,
+      "grad_norm": 1.9388853473260568,
+      "learning_rate": 2.0949661975584367e-05,
+      "loss": 3.2459,
+      "step": 14050
+    },
+    {
+      "epoch": 30.128205128205128,
+      "grad_norm": 1.7582868148049866,
+      "learning_rate": 2.070315032413864e-05,
+      "loss": 3.2272,
+      "step": 14100
+    },
+    {
+      "epoch": 30.235042735042736,
+      "grad_norm": 1.8745030698305942,
+      "learning_rate": 2.0458053039032265e-05,
+      "loss": 3.2212,
+      "step": 14150
+    },
+    {
+      "epoch": 30.34188034188034,
+      "grad_norm": 1.800409969690226,
+      "learning_rate": 2.0214386024276854e-05,
+      "loss": 3.224,
+      "step": 14200
+    },
+    {
+      "epoch": 30.44871794871795,
+      "grad_norm": 1.8714060990555732,
+      "learning_rate": 1.9972165091075787e-05,
+      "loss": 3.2262,
+      "step": 14250
+    },
+    {
+      "epoch": 30.555555555555557,
+      "grad_norm": 1.8482145147540658,
+      "learning_rate": 1.9731405956798333e-05,
+      "loss": 3.2264,
+      "step": 14300
+    },
+    {
+      "epoch": 30.662393162393162,
+      "grad_norm": 1.9172055965648316,
+      "learning_rate": 1.949212424395967e-05,
+      "loss": 3.2248,
+      "step": 14350
+    },
+    {
+      "epoch": 30.76923076923077,
+      "grad_norm": 1.9310880204205334,
+      "learning_rate": 1.925433547920724e-05,
+      "loss": 3.2169,
+      "step": 14400
+    },
+    {
+      "epoch": 30.876068376068375,
+      "grad_norm": 1.9329611620092266,
+      "learning_rate": 1.901805509231324e-05,
+      "loss": 3.2267,
+      "step": 14450
+    },
+    {
+      "epoch": 30.982905982905983,
+      "grad_norm": 1.8547174800001984,
+      "learning_rate": 1.878329841517336e-05,
+      "loss": 3.217,
+      "step": 14500
+    },
+    {
+      "epoch": 31.08974358974359,
+      "grad_norm": 1.7364694818856983,
+      "learning_rate": 1.855008068081196e-05,
+      "loss": 3.2016,
+      "step": 14550
+    },
+    {
+      "epoch": 31.196581196581196,
+      "grad_norm": 1.8819759932435103,
+      "learning_rate": 1.8318417022393614e-05,
+      "loss": 3.1824,
+      "step": 14600
+    },
+    {
+      "epoch": 31.303418803418804,
+      "grad_norm": 1.9235117076004822,
+      "learning_rate": 1.808832247224111e-05,
+      "loss": 3.1979,
+      "step": 14650
+    },
+    {
+      "epoch": 31.41025641025641,
+      "grad_norm": 1.8096199172951792,
+      "learning_rate": 1.7859811960860106e-05,
+      "loss": 3.1959,
+      "step": 14700
+    },
+    {
+      "epoch": 31.517094017094017,
+      "grad_norm": 1.9824080443252146,
+      "learning_rate": 1.763290031597019e-05,
+      "loss": 3.1929,
+      "step": 14750
+    },
+    {
+      "epoch": 31.623931623931625,
+      "grad_norm": 1.8470203684544455,
+      "learning_rate": 1.740760226154283e-05,
+      "loss": 3.1979,
+      "step": 14800
+    },
+    {
+      "epoch": 31.73076923076923,
+      "grad_norm": 1.8520854742540558,
+      "learning_rate": 1.71839324168459e-05,
+      "loss": 3.1934,
+      "step": 14850
+    },
+    {
+      "epoch": 31.837606837606838,
+      "grad_norm": 1.8470971065908115,
+      "learning_rate": 1.6961905295495106e-05,
+      "loss": 3.1851,
+      "step": 14900
+    },
+    {
+      "epoch": 31.944444444444443,
+      "grad_norm": 1.8606032474106087,
+      "learning_rate": 1.6741535304512178e-05,
+      "loss": 3.1972,
+      "step": 14950
+    },
+    {
+      "epoch": 32.05128205128205,
+      "grad_norm": 1.9467679748998603,
+      "learning_rate": 1.6522836743390025e-05,
+      "loss": 3.1685,
+      "step": 15000
+    },
+    {
+      "epoch": 32.15811965811966,
+      "grad_norm": 1.9755201644621432,
+      "learning_rate": 1.630582380316487e-05,
+      "loss": 3.1585,
+      "step": 15050
+    },
+    {
+      "epoch": 32.26495726495727,
+      "grad_norm": 1.9717752736743752,
+      "learning_rate": 1.6090510565495406e-05,
+      "loss": 3.1647,
+      "step": 15100
+    },
+    {
+      "epoch": 32.37179487179487,
+      "grad_norm": 1.9484597603291864,
+      "learning_rate": 1.5876911001749066e-05,
+      "loss": 3.1572,
+      "step": 15150
+    },
+    {
+      "epoch": 32.47863247863248,
+      "grad_norm": 2.0190422011977076,
+      "learning_rate": 1.5665038972095462e-05,
+      "loss": 3.1677,
+      "step": 15200
+    },
+    {
+      "epoch": 32.585470085470085,
+      "grad_norm": 2.0018694008788507,
+      "learning_rate": 1.545490822460696e-05,
+      "loss": 3.1596,
+      "step": 15250
+    },
+    {
+      "epoch": 32.69230769230769,
+      "grad_norm": 1.9317622594847947,
+      "learning_rate": 1.5246532394366637e-05,
+      "loss": 3.1512,
+      "step": 15300
+    },
+    {
+      "epoch": 32.7991452991453,
+      "grad_norm": 1.939556015374055,
+      "learning_rate": 1.5039925002583505e-05,
+      "loss": 3.1508,
+      "step": 15350
+    },
+    {
+      "epoch": 32.9059829059829,
+      "grad_norm": 1.9314825072339465,
+      "learning_rate": 1.4835099455715124e-05,
+      "loss": 3.1559,
+      "step": 15400
+    },
+    {
+      "epoch": 33.01282051282051,
+      "grad_norm": 2.002199155994756,
+      "learning_rate": 1.4632069044597711e-05,
+      "loss": 3.1583,
+      "step": 15450
+    },
+    {
+      "epoch": 33.11965811965812,
+      "grad_norm": 1.94120643247509,
+      "learning_rate": 1.4430846943583707e-05,
+      "loss": 3.137,
+      "step": 15500
+    },
+    {
+      "epoch": 33.22649572649573,
+      "grad_norm": 2.0323075109268594,
+      "learning_rate": 1.4231446209686875e-05,
+      "loss": 3.144,
+      "step": 15550
+    },
+    {
+      "epoch": 33.333333333333336,
+      "grad_norm": 1.8398275516388178,
+      "learning_rate": 1.4033879781735071e-05,
+      "loss": 3.1402,
+      "step": 15600
+    },
+    {
+      "epoch": 33.44017094017094,
+      "grad_norm": 1.9227151042513004,
+      "learning_rate": 1.3838160479530688e-05,
+      "loss": 3.1145,
+      "step": 15650
+    },
+    {
+      "epoch": 33.547008547008545,
+      "grad_norm": 2.0943205326548644,
+      "learning_rate": 1.3644301003018776e-05,
+      "loss": 3.1343,
+      "step": 15700
+    },
+    {
+      "epoch": 33.65384615384615,
+      "grad_norm": 2.056329687365839,
+      "learning_rate": 1.3452313931462957e-05,
+      "loss": 3.1516,
+      "step": 15750
+    },
+    {
+      "epoch": 33.76068376068376,
+      "grad_norm": 2.0407832693322616,
+      "learning_rate": 1.3262211722629166e-05,
+      "loss": 3.1306,
+      "step": 15800
+    },
+    {
+      "epoch": 33.86752136752137,
+      "grad_norm": 1.9753732835530098,
+      "learning_rate": 1.3074006711977312e-05,
+      "loss": 3.1255,
+      "step": 15850
+    },
+    {
+      "epoch": 33.97435897435897,
+      "grad_norm": 2.0533348823929596,
+      "learning_rate": 1.2887711111860816e-05,
+      "loss": 3.1291,
+      "step": 15900
+    },
+    {
+      "epoch": 34.08119658119658,
+      "grad_norm": 1.9851072867584276,
+      "learning_rate": 1.270333701073422e-05,
+      "loss": 3.1151,
+      "step": 15950
+    },
+    {
+      "epoch": 34.18803418803419,
+      "grad_norm": 1.9583359440995798,
+      "learning_rate": 1.2520896372368716e-05,
+      "loss": 3.1057,
+      "step": 16000
+    },
+    {
+      "epoch": 34.294871794871796,
+      "grad_norm": 2.020343078712297,
+      "learning_rate": 1.2340401035075869e-05,
+      "loss": 3.117,
+      "step": 16050
+    },
+    {
+      "epoch": 34.401709401709404,
+      "grad_norm": 1.9569342126602145,
+      "learning_rate": 1.2161862710939476e-05,
+      "loss": 3.1123,
+      "step": 16100
+    },
+    {
+      "epoch": 34.50854700854701,
+      "grad_norm": 1.9598035839496117,
+      "learning_rate": 1.1985292985055514e-05,
+      "loss": 3.1129,
+      "step": 16150
+    },
+    {
+      "epoch": 34.61538461538461,
+      "grad_norm": 1.9758674092962976,
+      "learning_rate": 1.1810703314780488e-05,
+      "loss": 3.117,
+      "step": 16200
+    },
+    {
+      "epoch": 34.72222222222222,
+      "grad_norm": 2.020822964146809,
+      "learning_rate": 1.1638105028987887e-05,
+      "loss": 3.1107,
+      "step": 16250
+    },
+    {
+      "epoch": 34.82905982905983,
+      "grad_norm": 2.034351973395989,
+      "learning_rate": 1.1467509327333134e-05,
+      "loss": 3.1121,
+      "step": 16300
+    },
+    {
+      "epoch": 34.93589743589744,
+      "grad_norm": 2.0787426023644113,
+      "learning_rate": 1.1298927279526844e-05,
+      "loss": 3.0885,
+      "step": 16350
+    },
+    {
+      "epoch": 35.042735042735046,
+      "grad_norm": 2.0398209721415927,
+      "learning_rate": 1.1132369824616499e-05,
+      "loss": 3.1038,
+      "step": 16400
+    },
+    {
+      "epoch": 35.14957264957265,
+      "grad_norm": 2.0772037615310985,
+      "learning_rate": 1.0967847770276674e-05,
+      "loss": 3.0692,
+      "step": 16450
+    },
+    {
+      "epoch": 35.256410256410255,
+      "grad_norm": 2.0532519761394976,
+      "learning_rate": 1.0805371792107705e-05,
+      "loss": 3.0918,
+      "step": 16500
+    },
+    {
+      "epoch": 35.363247863247864,
+      "grad_norm": 2.092273932018464,
+      "learning_rate": 1.0644952432942967e-05,
+      "loss": 3.0851,
+      "step": 16550
+    },
+    {
+      "epoch": 35.47008547008547,
+      "grad_norm": 2.0272103855610695,
+      "learning_rate": 1.0486600102164802e-05,
+      "loss": 3.095,
+      "step": 16600
+    },
+    {
+      "epoch": 35.57692307692308,
+      "grad_norm": 1.9982121344245152,
+      "learning_rate": 1.0330325075029005e-05,
+      "loss": 3.075,
+      "step": 16650
+    },
+    {
+      "epoch": 35.68376068376068,
+      "grad_norm": 2.06961519986216,
+      "learning_rate": 1.0176137491998153e-05,
+      "loss": 3.0707,
+      "step": 16700
+    },
+    {
+      "epoch": 35.79059829059829,
+      "grad_norm": 2.0995806456973414,
+      "learning_rate": 1.002404735808355e-05,
+      "loss": 3.0728,
+      "step": 16750
+    },
+    {
+      "epoch": 35.8974358974359,
+      "grad_norm": 2.0785784692696536,
+      "learning_rate": 9.874064542196033e-06,
+      "loss": 3.101,
+      "step": 16800
+    },
+    {
+      "epoch": 36.004273504273506,
+      "grad_norm": 2.046538463619717,
+      "learning_rate": 9.72619877650558e-06,
+      "loss": 3.0851,
+      "step": 16850
+    },
+    {
+      "epoch": 36.111111111111114,
+      "grad_norm": 2.023732403212745,
+      "learning_rate": 9.580459655809827e-06,
+      "loss": 3.0623,
+      "step": 16900
+    },
+    {
+      "epoch": 36.217948717948715,
+      "grad_norm": 2.0773001733336076,
+      "learning_rate": 9.436856636911479e-06,
+      "loss": 3.0529,
+      "step": 16950
+    },
+    {
+      "epoch": 36.324786324786324,
+      "grad_norm": 2.0516402125983695,
+      "learning_rate": 9.295399038004633e-06,
+      "loss": 3.0461,
+      "step": 17000
+    },
+    {
+      "epoch": 36.43162393162393,
+      "grad_norm": 2.1174524570789073,
+      "learning_rate": 9.156096038070163e-06,
+      "loss": 3.0454,
+      "step": 17050
+    },
+    {
+      "epoch": 36.53846153846154,
+      "grad_norm": 2.0514328856826998,
+      "learning_rate": 9.018956676280101e-06,
+      "loss": 3.0703,
+      "step": 17100
+    },
+    {
+      "epoch": 36.64529914529915,
+      "grad_norm": 2.165466183870704,
+      "learning_rate": 8.88398985141111e-06,
+      "loss": 3.0554,
+      "step": 17150
+    },
+    {
+      "epoch": 36.75213675213675,
+      "grad_norm": 2.0912934880999225,
+      "learning_rate": 8.75120432126705e-06,
+      "loss": 3.0511,
+      "step": 17200
+    },
+    {
+      "epoch": 36.85897435897436,
+      "grad_norm": 2.0425737921879845,
+      "learning_rate": 8.620608702110672e-06,
+      "loss": 3.0763,
+      "step": 17250
+    },
+    {
+      "epoch": 36.965811965811966,
+      "grad_norm": 2.1232600100726087,
+      "learning_rate": 8.492211468104547e-06,
+      "loss": 3.062,
+      "step": 17300
+    },
+    {
+      "epoch": 37.072649572649574,
+      "grad_norm": 2.1110434256405495,
+      "learning_rate": 8.366020950761204e-06,
+      "loss": 3.0434,
+      "step": 17350
+    },
+    {
+      "epoch": 37.17948717948718,
+      "grad_norm": 2.1450862756909035,
+      "learning_rate": 8.242045338402464e-06,
+      "loss": 3.0422,
+      "step": 17400
+    },
+    {
+      "epoch": 37.28632478632478,
+      "grad_norm": 2.0605806610085566,
+      "learning_rate": 8.12029267562816e-06,
+      "loss": 3.0413,
+      "step": 17450
+    },
+    {
+      "epoch": 37.39316239316239,
+      "grad_norm": 2.0451305614475324,
+      "learning_rate": 8.000770862794092e-06,
+      "loss": 3.047,
+      "step": 17500
+    },
+    {
+      "epoch": 37.5,
+      "grad_norm": 2.2090559323000467,
+      "learning_rate": 7.883487655499422e-06,
+      "loss": 3.0383,
+      "step": 17550
+    },
+    {
+      "epoch": 37.60683760683761,
+      "grad_norm": 2.11643422369848,
+      "learning_rate": 7.768450664083389e-06,
+      "loss": 3.0443,
+      "step": 17600
+    },
+    {
+      "epoch": 37.71367521367522,
+      "grad_norm": 2.116595083312843,
+      "learning_rate": 7.655667353131507e-06,
+      "loss": 3.0281,
+      "step": 17650
+    },
+    {
+      "epoch": 37.82051282051282,
+      "grad_norm": 2.0966523963349597,
+      "learning_rate": 7.545145040991213e-06,
+      "loss": 3.0375,
+      "step": 17700
+    },
+    {
+      "epoch": 37.927350427350426,
+      "grad_norm": 2.1393354109439686,
+      "learning_rate": 7.4368908992969494e-06,
+      "loss": 3.0451,
+      "step": 17750
+    },
+    {
+      "epoch": 38.034188034188034,
+      "grad_norm": 2.171999152497746,
+      "learning_rate": 7.330911952504839e-06,
+      "loss": 3.0244,
+      "step": 17800
+    },
+    {
+      "epoch": 38.14102564102564,
+      "grad_norm": 2.201534265102945,
+      "learning_rate": 7.227215077436858e-06,
+      "loss": 3.0086,
+      "step": 17850
+    },
+    {
+      "epoch": 38.24786324786325,
+      "grad_norm": 2.1852434781244177,
+      "learning_rate": 7.125807002834633e-06,
+      "loss": 3.0096,
+      "step": 17900
+    },
+    {
+      "epoch": 38.35470085470085,
+      "grad_norm": 2.1435749304995184,
+      "learning_rate": 7.026694308922806e-06,
+      "loss": 3.0155,
+      "step": 17950
+    },
+    {
+      "epoch": 38.46153846153846,
+      "grad_norm": 2.013436955865304,
+      "learning_rate": 6.929883426982045e-06,
+      "loss": 3.0241,
+      "step": 18000
+    },
+    {
+      "epoch": 38.56837606837607,
+      "grad_norm": 2.3006132676567166,
+      "learning_rate": 6.835380638931743e-06,
+      "loss": 3.0139,
+      "step": 18050
+    },
+    {
+      "epoch": 38.675213675213676,
+      "grad_norm": 2.053756909257222,
+      "learning_rate": 6.7431920769223915e-06,
+      "loss": 3.0185,
+      "step": 18100
+    },
+    {
+      "epoch": 38.782051282051285,
+      "grad_norm": 2.212010530974445,
+      "learning_rate": 6.653323722937661e-06,
+      "loss": 3.0187,
+      "step": 18150
+    },
+    {
+      "epoch": 38.888888888888886,
+      "grad_norm": 2.1214954424397146,
+      "learning_rate": 6.565781408406267e-06,
+      "loss": 3.0172,
+      "step": 18200
+    },
+    {
+      "epoch": 38.995726495726494,
+      "grad_norm": 2.240490101433792,
+      "learning_rate": 6.4805708138235415e-06,
+      "loss": 3.0359,
+      "step": 18250
+    },
+    {
+      "epoch": 39.1025641025641,
+      "grad_norm": 2.054957615027752,
+      "learning_rate": 6.397697468382851e-06,
+      "loss": 3.0037,
+      "step": 18300
+    },
+    {
+      "epoch": 39.20940170940171,
+      "grad_norm": 2.1534118771277666,
+      "learning_rate": 6.317166749616825e-06,
+      "loss": 2.9826,
+      "step": 18350
+    },
+    {
+      "epoch": 39.31623931623932,
+      "grad_norm": 2.1670160623191896,
+      "learning_rate": 6.238983883048393e-06,
+      "loss": 2.9874,
+      "step": 18400
+    },
+    {
+      "epoch": 39.42307692307692,
+      "grad_norm": 2.0968230739935096,
+      "learning_rate": 6.163153941851729e-06,
+      "loss": 2.997,
+      "step": 18450
+    },
+    {
+      "epoch": 39.52991452991453,
+      "grad_norm": 2.2848578364181877,
+      "learning_rate": 6.08968184652305e-06,
+      "loss": 2.998,
+      "step": 18500
+    },
+    {
+      "epoch": 39.636752136752136,
+      "grad_norm": 2.1921025494234345,
+      "learning_rate": 6.018572364561333e-06,
+      "loss": 2.9946,
+      "step": 18550
+    },
+    {
+      "epoch": 39.743589743589745,
+      "grad_norm": 2.1811327227787447,
+      "learning_rate": 5.949830110158959e-06,
+      "loss": 3.0138,
+      "step": 18600
+    },
+    {
+      "epoch": 39.85042735042735,
+      "grad_norm": 2.1892063570097218,
+      "learning_rate": 5.883459543902297e-06,
+      "loss": 2.9962,
+      "step": 18650
+    },
+    {
+      "epoch": 39.957264957264954,
+      "grad_norm": 2.1993511153341863,
+      "learning_rate": 5.819464972482287e-06,
+      "loss": 2.9952,
+      "step": 18700
+    },
+    {
+      "epoch": 40.06410256410256,
+      "grad_norm": 2.2680369199677757,
+      "learning_rate": 5.7578505484149624e-06,
+      "loss": 3.0053,
+      "step": 18750
+    },
+    {
+      "epoch": 40.17094017094017,
+      "grad_norm": 2.178325268419898,
+      "learning_rate": 5.698620269771997e-06,
+      "loss": 2.985,
+      "step": 18800
+    },
+    {
+      "epoch": 40.27777777777778,
+      "grad_norm": 2.1967082626425634,
+      "learning_rate": 5.641777979921297e-06,
+      "loss": 2.9902,
+      "step": 18850
+    },
+    {
+      "epoch": 40.38461538461539,
+      "grad_norm": 2.3133523761980537,
+      "learning_rate": 5.587327367277585e-06,
+      "loss": 2.9592,
+      "step": 18900
+    },
+    {
+      "epoch": 40.49145299145299,
+      "grad_norm": 2.171592419580568,
+      "learning_rate": 5.535271965063095e-06,
+      "loss": 2.9763,
+      "step": 18950
+    },
+    {
+      "epoch": 40.598290598290596,
+      "grad_norm": 2.205495978905872,
+      "learning_rate": 5.4856151510782684e-06,
+      "loss": 2.9977,
+      "step": 19000
+    },
+    {
+      "epoch": 40.705128205128204,
+      "grad_norm": 2.219457808947546,
+      "learning_rate": 5.438360147482613e-06,
+      "loss": 2.9937,
+      "step": 19050
+    },
+    {
+      "epoch": 40.81196581196581,
+      "grad_norm": 2.143182048689108,
+      "learning_rate": 5.393510020585584e-06,
+      "loss": 2.9858,
+      "step": 19100
+    },
+    {
+      "epoch": 40.91880341880342,
+      "grad_norm": 2.2602763212966908,
+      "learning_rate": 5.351067680647651e-06,
+      "loss": 2.984,
+      "step": 19150
+    },
+    {
+      "epoch": 41.02564102564103,
+      "grad_norm": 2.237613918025757,
+      "learning_rate": 5.311035881691424e-06,
+      "loss": 2.9765,
+      "step": 19200
+    },
+    {
+      "epoch": 41.13247863247863,
+      "grad_norm": 2.253413260507676,
+      "learning_rate": 5.273417221322976e-06,
+      "loss": 2.9712,
+      "step": 19250
+    },
+    {
+      "epoch": 41.23931623931624,
+      "grad_norm": 2.2190398980533974,
+      "learning_rate": 5.2382141405632635e-06,
+      "loss": 2.9578,
+      "step": 19300
+    },
+    {
+      "epoch": 41.34615384615385,
+      "grad_norm": 2.1465746762525546,
+      "learning_rate": 5.205428923689753e-06,
+      "loss": 2.9654,
+      "step": 19350
+    },
+    {
+      "epoch": 41.452991452991455,
+      "grad_norm": 2.3382052127852138,
+      "learning_rate": 5.175063698088176e-06,
+      "loss": 2.9657,
+      "step": 19400
+    },
+    {
+      "epoch": 41.55982905982906,
+      "grad_norm": 2.307601276460699,
+      "learning_rate": 5.147120434114517e-06,
+      "loss": 2.9467,
+      "step": 19450
+    },
+    {
+      "epoch": 41.666666666666664,
+      "grad_norm": 2.1265313857448955,
+      "learning_rate": 5.121600944967128e-06,
+      "loss": 2.9668,
+      "step": 19500
+    },
+    {
+      "epoch": 41.77350427350427,
+      "grad_norm": 2.2877212083930907,
+      "learning_rate": 5.098506886569086e-06,
+      "loss": 2.9834,
+      "step": 19550
+    },
+    {
+      "epoch": 41.88034188034188,
+      "grad_norm": 2.2687584616763368,
+      "learning_rate": 5.077839757460745e-06,
+      "loss": 2.9736,
+      "step": 19600
+    },
+    {
+      "epoch": 41.98717948717949,
+      "grad_norm": 2.218212008732717,
+      "learning_rate": 5.059600898702501e-06,
+      "loss": 2.9504,
+      "step": 19650
+    },
+    {
+      "epoch": 42.0940170940171,
+      "grad_norm": 2.2536964781674453,
+      "learning_rate": 5.043791493787751e-06,
+      "loss": 2.9602,
+      "step": 19700
+    },
+    {
+      "epoch": 42.2008547008547,
+      "grad_norm": 2.2352386086099743,
+      "learning_rate": 5.030412568566141e-06,
+      "loss": 2.9694,
+      "step": 19750
+    },
+    {
+      "epoch": 42.30769230769231,
+      "grad_norm": 2.1999717493844324,
+      "learning_rate": 5.019464991176942e-06,
+      "loss": 2.9492,
+      "step": 19800
+    },
+    {
+      "epoch": 42.414529914529915,
+      "grad_norm": 2.2665941993498455,
+      "learning_rate": 5.010949471992765e-06,
+      "loss": 2.9823,
+      "step": 19850
+    },
+    {
+      "epoch": 42.52136752136752,
+      "grad_norm": 2.2472033604271853,
+      "learning_rate": 5.00486656357345e-06,
+      "loss": 2.9619,
+      "step": 19900
+    },
+    {
+      "epoch": 42.62820512820513,
+      "grad_norm": 2.165219324821127,
+      "learning_rate": 5.001216660630201e-06,
+      "loss": 2.9508,
+      "step": 19950
+    },
+    {
+      "epoch": 42.73504273504273,
+      "grad_norm": 2.1433868412005084,
+      "learning_rate": 5e-06,
+      "loss": 2.9738,
+      "step": 20000
+    },
+    {
+      "epoch": 42.73504273504273,
+      "step": 20000,
+      "total_flos": 9814671360000000.0,
+      "train_loss": 3.479709552001953,
+      "train_runtime": 156320.8197,
+      "train_samples_per_second": 16.377,
+      "train_steps_per_second": 0.128
+    }
+  ],
+  "logging_steps": 50,
+  "max_steps": 20000,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 43,
+  "save_steps": 4000,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 9814671360000000.0,
+  "train_batch_size": 4,
+  "trial_name": null,
+  "trial_params": null
+}

UNIVLA_SIMPLER_BRIDGE_VIDEO_BS128_20K/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:419f36618e094e884dab139501850c14c9d35bbff7741e4de0a5b42f07ba8c9f
+size 7352