Model save

README.md

@@ -0,0 +1,72 @@
+---
+license: llama2
+base_model: meta-llama/Llama-2-7b-hf
+tags:
+- generated_from_trainer
+model-index:
+- name: llama2-spin-zephyr-data
+  results: []
+---
+
+<!-- This model card has been generated automatically according to the information the Trainer had access to. You
+should probably proofread and complete it, then remove this comment. -->
+
+# llama2-spin-zephyr-data
+
+This model is a fine-tuned version of [meta-llama/Llama-2-7b-hf](https://huggingface.co/meta-llama/Llama-2-7b-hf) on the None dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.0032
+- Rewards/real: -1.5322
+- Rewards/generated: -20.8812
+- Rewards/accuracies: 0.9992
+- Rewards/margins: 19.3491
+- Logps/generated: -369.1998
+- Logps/real: -194.6499
+- Logits/generated: 0.0527
+- Logits/real: -0.5587
+
+## Model description
+
+More information needed
+
+## Intended uses & limitations
+
+More information needed
+
+## Training and evaluation data
+
+More information needed
+
+## Training procedure
+
+### Training hyperparameters
+
+The following hyperparameters were used during training:
+- learning_rate: 5e-07
+- train_batch_size: 8
+- eval_batch_size: 8
+- seed: 42
+- distributed_type: multi-GPU
+- num_devices: 4
+- total_train_batch_size: 32
+- total_eval_batch_size: 32
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: linear
+- lr_scheduler_warmup_ratio: 0.1
+- num_epochs: 1
+
+### Training results
+
+| Training Loss | Epoch | Step | Validation Loss | Rewards/real | Rewards/generated | Rewards/accuracies | Rewards/margins | Logps/generated | Logps/real | Logits/generated | Logits/real |
+|:-------------:|:-----:|:----:|:---------------:|:------------:|:-----------------:|:------------------:|:---------------:|:---------------:|:----------:|:----------------:|:-----------:|
+| 0.0098        | 0.32  | 500  | 0.0108          | -0.7647      | -14.3850          | 0.9976             | 13.6202         | -304.2368       | -186.9758  | 0.0151           | -0.6190     |
+| 0.0019        | 0.64  | 1000 | 0.0044          | -1.7759      | -21.6593          | 0.9984             | 19.8833         | -376.9800       | -197.0876  | 0.0480           | -0.5656     |
+| 0.0009        | 0.96  | 1500 | 0.0032          | -1.5322      | -20.8812          | 0.9992             | 19.3491         | -369.1998       | -194.6499  | 0.0527           | -0.5587     |
+
+
+### Framework versions
+
+- Transformers 4.37.0
+- Pytorch 2.1.2+cu121
+- Datasets 2.14.6
+- Tokenizers 0.15.2

all_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 1.0,
+    "train_loss": 0.03590757104014634,
+    "train_runtime": 9394.7927,
+    "train_samples": 50000,
+    "train_samples_per_second": 5.322,
+    "train_steps_per_second": 0.166
+}

generation_config.json

@@ -0,0 +1,10 @@
+{
+  "bos_token_id": 1,
+  "do_sample": true,
+  "eos_token_id": 2,
+  "max_length": 4096,
+  "pad_token_id": 0,
+  "temperature": 0.6,
+  "top_p": 0.9,
+  "transformers_version": "4.37.0"
+}

model-00001-of-00003.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:12890ea53e4f97feb9e132091ae43deaf87fa0d9954d150ed9d0a44201177e5b
+size 4938985352

model-00002-of-00003.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9767f2f79f804ebd0054bec49717333ce50760e7b0a7e4178d7722af4d63c638
+size 4947390880

model-00003-of-00003.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6a1c4199bc7a928c6f3dfc034fa7dc4c590f03f110176eb06c442b6ff5008f48
+size 3590488816

model.safetensors.index.json

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

runs/Jul23_00-51-14_gilbreth-j001.rcac.purdue.edu/events.out.tfevents.1721710511.gilbreth-j001.rcac.purdue.edu.238330.0

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:efa529fa5313cb58d58e10aee070c55a05bab34edfadd7e94d0ba9701a38d097
-size 101985
+oid sha256:cfc3895aa7196efef945130f1ea61b5d13e2f929b511bf6bd998e08835df7847
+size 106125

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 1.0,
+    "train_loss": 0.03590757104014634,
+    "train_runtime": 9394.7927,
+    "train_samples": 50000,
+    "train_samples_per_second": 5.322,
+    "train_steps_per_second": 0.166
+}

trainer_state.json

@@ -0,0 +1,2276 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 1.0,
+  "eval_steps": 500,
+  "global_step": 1563,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.1847133757961784e-09,
+      "logits/generated": -0.09026163071393967,
+      "logits/real": -0.800382137298584,
+      "logps/generated": -180.3804931640625,
+      "logps/real": -164.2542724609375,
+      "loss": 0.6931,
+      "rewards/accuracies": 0.0,
+      "rewards/generated": 0.0,
+      "rewards/margins": 0.0,
+      "rewards/real": 0.0,
+      "step": 1
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.184713375796178e-08,
+      "logits/generated": -0.08320371806621552,
+      "logits/real": -0.7675037980079651,
+      "logps/generated": -161.61961364746094,
+      "logps/real": -178.428466796875,
+      "loss": 0.6945,
+      "rewards/accuracies": 0.4444444477558136,
+      "rewards/generated": 0.007918823510408401,
+      "rewards/margins": -0.007639557123184204,
+      "rewards/real": 0.00027926763868890703,
+      "step": 10
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 6.369426751592356e-08,
+      "logits/generated": -0.1008232831954956,
+      "logits/real": -0.9005411863327026,
+      "logps/generated": -165.965576171875,
+      "logps/real": -174.21055603027344,
+      "loss": 0.6805,
+      "rewards/accuracies": 0.612500011920929,
+      "rewards/generated": -0.011105736717581749,
+      "rewards/margins": 0.017230339348316193,
+      "rewards/real": 0.006124601699411869,
+      "step": 20
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 9.554140127388536e-08,
+      "logits/generated": -0.08152450621128082,
+      "logits/real": -0.7754586935043335,
+      "logps/generated": -165.53176879882812,
+      "logps/real": -185.11846923828125,
+      "loss": 0.6398,
+      "rewards/accuracies": 0.8374999761581421,
+      "rewards/generated": -0.0970459133386612,
+      "rewards/margins": 0.10573717206716537,
+      "rewards/real": 0.008691254071891308,
+      "step": 30
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 1.2738853503184713e-07,
+      "logits/generated": 0.019896607846021652,
+      "logits/real": -0.8798272013664246,
+      "logps/generated": -153.7320556640625,
+      "logps/real": -180.03219604492188,
+      "loss": 0.5699,
+      "rewards/accuracies": 0.9624999761581421,
+      "rewards/generated": -0.25305813550949097,
+      "rewards/margins": 0.2757338285446167,
+      "rewards/real": 0.022675666958093643,
+      "step": 40
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 1.592356687898089e-07,
+      "logits/generated": -0.05750712752342224,
+      "logits/real": -0.8368139266967773,
+      "logps/generated": -161.1537322998047,
+      "logps/real": -172.9757537841797,
+      "loss": 0.4647,
+      "rewards/accuracies": 0.949999988079071,
+      "rewards/generated": -0.4674338400363922,
+      "rewards/margins": 0.5483574271202087,
+      "rewards/real": 0.0809236392378807,
+      "step": 50
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 1.9108280254777072e-07,
+      "logits/generated": -0.011790583841502666,
+      "logits/real": -0.7276524305343628,
+      "logps/generated": -175.3116455078125,
+      "logps/real": -182.06173706054688,
+      "loss": 0.3557,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -0.9212247729301453,
+      "rewards/margins": 1.0005762577056885,
+      "rewards/real": 0.07935139536857605,
+      "step": 60
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 2.2292993630573247e-07,
+      "logits/generated": 0.037871506065130234,
+      "logits/real": -0.757712721824646,
+      "logps/generated": -168.0603790283203,
+      "logps/real": -185.0572509765625,
+      "loss": 0.2403,
+      "rewards/accuracies": 0.9624999761581421,
+      "rewards/generated": -1.2728922367095947,
+      "rewards/margins": 1.4110310077667236,
+      "rewards/real": 0.1381385624408722,
+      "step": 70
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 2.5477707006369425e-07,
+      "logits/generated": 0.008935372345149517,
+      "logits/real": -0.8579050302505493,
+      "logps/generated": -176.29393005371094,
+      "logps/real": -171.60939025878906,
+      "loss": 0.1815,
+      "rewards/accuracies": 0.9624999761581421,
+      "rewards/generated": -1.7076524496078491,
+      "rewards/margins": 1.8567367792129517,
+      "rewards/real": 0.1490844190120697,
+      "step": 80
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 2.86624203821656e-07,
+      "logits/generated": -0.0024472028017044067,
+      "logits/real": -0.8354307413101196,
+      "logps/generated": -194.20306396484375,
+      "logps/real": -181.1865997314453,
+      "loss": 0.1284,
+      "rewards/accuracies": 0.987500011920929,
+      "rewards/generated": -2.3786566257476807,
+      "rewards/margins": 2.5542044639587402,
+      "rewards/real": 0.17554807662963867,
+      "step": 90
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 3.184713375796178e-07,
+      "logits/generated": -0.007845225743949413,
+      "logits/real": -0.8014926910400391,
+      "logps/generated": -193.57733154296875,
+      "logps/real": -172.54800415039062,
+      "loss": 0.117,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -2.8054349422454834,
+      "rewards/margins": 2.9194352626800537,
+      "rewards/real": 0.11400020122528076,
+      "step": 100
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 3.5031847133757957e-07,
+      "logits/generated": 0.023908555507659912,
+      "logits/real": -0.7339428663253784,
+      "logps/generated": -195.67071533203125,
+      "logps/real": -188.1604461669922,
+      "loss": 0.0891,
+      "rewards/accuracies": 0.9750000238418579,
+      "rewards/generated": -3.3142905235290527,
+      "rewards/margins": 3.3457164764404297,
+      "rewards/real": 0.03142569214105606,
+      "step": 110
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 3.8216560509554143e-07,
+      "logits/generated": 0.04019797593355179,
+      "logits/real": -0.6734101176261902,
+      "logps/generated": -209.1021728515625,
+      "logps/real": -176.04254150390625,
+      "loss": 0.0794,
+      "rewards/accuracies": 0.987500011920929,
+      "rewards/generated": -4.002293109893799,
+      "rewards/margins": 4.102808475494385,
+      "rewards/real": 0.1005152240395546,
+      "step": 120
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 4.140127388535032e-07,
+      "logits/generated": 0.07181330770254135,
+      "logits/real": -0.6598816514015198,
+      "logps/generated": -198.8852081298828,
+      "logps/real": -184.90480041503906,
+      "loss": 0.0761,
+      "rewards/accuracies": 0.9750000238418579,
+      "rewards/generated": -4.574056625366211,
+      "rewards/margins": 4.542202949523926,
+      "rewards/real": -0.031853675842285156,
+      "step": 130
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 4.4585987261146494e-07,
+      "logits/generated": 0.09077299386262894,
+      "logits/real": -0.7375579476356506,
+      "logps/generated": -217.79299926757812,
+      "logps/real": -169.71145629882812,
+      "loss": 0.048,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -5.316611289978027,
+      "rewards/margins": 5.399137020111084,
+      "rewards/real": 0.08252569288015366,
+      "step": 140
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 4.777070063694267e-07,
+      "logits/generated": 0.019507689401507378,
+      "logits/real": -0.529100775718689,
+      "logps/generated": -227.7717742919922,
+      "logps/real": -181.4638214111328,
+      "loss": 0.0624,
+      "rewards/accuracies": 0.9750000238418579,
+      "rewards/generated": -5.985389709472656,
+      "rewards/margins": 5.939720630645752,
+      "rewards/real": -0.04566919058561325,
+      "step": 150
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 4.989331436699858e-07,
+      "logits/generated": 0.07675327360630035,
+      "logits/real": -0.6792179346084595,
+      "logps/generated": -230.45321655273438,
+      "logps/real": -177.54190063476562,
+      "loss": 0.0403,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -6.596798896789551,
+      "rewards/margins": 6.574460506439209,
+      "rewards/real": -0.02233867719769478,
+      "step": 160
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 4.953769559032717e-07,
+      "logits/generated": 0.10339117050170898,
+      "logits/real": -0.6523188352584839,
+      "logps/generated": -238.82418823242188,
+      "logps/real": -172.30154418945312,
+      "loss": 0.044,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -7.214089870452881,
+      "rewards/margins": 7.0066819190979,
+      "rewards/real": -0.2074071168899536,
+      "step": 170
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 4.918207681365576e-07,
+      "logits/generated": -0.02095809206366539,
+      "logits/real": -0.5748814344406128,
+      "logps/generated": -243.3128204345703,
+      "logps/real": -193.81964111328125,
+      "loss": 0.0408,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -7.754981994628906,
+      "rewards/margins": 7.589665412902832,
+      "rewards/real": -0.16531690955162048,
+      "step": 180
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 4.882645803698435e-07,
+      "logits/generated": 0.08188272267580032,
+      "logits/real": -0.6443125009536743,
+      "logps/generated": -239.46932983398438,
+      "logps/real": -188.2861328125,
+      "loss": 0.0393,
+      "rewards/accuracies": 0.9750000238418579,
+      "rewards/generated": -7.951255798339844,
+      "rewards/margins": 7.745802402496338,
+      "rewards/real": -0.2054535448551178,
+      "step": 190
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 4.847083926031294e-07,
+      "logits/generated": 0.0788329690694809,
+      "logits/real": -0.6880910396575928,
+      "logps/generated": -249.65542602539062,
+      "logps/real": -176.37173461914062,
+      "loss": 0.045,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -8.179863929748535,
+      "rewards/margins": 7.825772762298584,
+      "rewards/real": -0.35409015417099,
+      "step": 200
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 4.811522048364154e-07,
+      "logits/generated": 0.08456435799598694,
+      "logits/real": -0.7073934674263,
+      "logps/generated": -243.7119903564453,
+      "logps/real": -164.15887451171875,
+      "loss": 0.0312,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -8.219891548156738,
+      "rewards/margins": 8.312192916870117,
+      "rewards/real": 0.09230276197195053,
+      "step": 210
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 4.775960170697012e-07,
+      "logits/generated": 0.05936474725604057,
+      "logits/real": -0.6907894015312195,
+      "logps/generated": -248.00094604492188,
+      "logps/real": -162.20657348632812,
+      "loss": 0.0321,
+      "rewards/accuracies": 0.9624999761581421,
+      "rewards/generated": -8.413338661193848,
+      "rewards/margins": 8.484495162963867,
+      "rewards/real": 0.07115854322910309,
+      "step": 220
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 4.7403982930298717e-07,
+      "logits/generated": 0.10569562762975693,
+      "logits/real": -0.7800209522247314,
+      "logps/generated": -250.5619659423828,
+      "logps/real": -172.96253967285156,
+      "loss": 0.0278,
+      "rewards/accuracies": 0.987500011920929,
+      "rewards/generated": -9.249897003173828,
+      "rewards/margins": 9.181459426879883,
+      "rewards/real": -0.06843843311071396,
+      "step": 230
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 4.7048364153627306e-07,
+      "logits/generated": 0.053275883197784424,
+      "logits/real": -0.5331145524978638,
+      "logps/generated": -257.272705078125,
+      "logps/real": -204.5064239501953,
+      "loss": 0.0336,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -9.000371932983398,
+      "rewards/margins": 8.605985641479492,
+      "rewards/real": -0.39438483119010925,
+      "step": 240
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 4.66927453769559e-07,
+      "logits/generated": 0.03677482530474663,
+      "logits/real": -0.6608942151069641,
+      "logps/generated": -256.04937744140625,
+      "logps/real": -171.3863067626953,
+      "loss": 0.027,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -9.227904319763184,
+      "rewards/margins": 8.95788860321045,
+      "rewards/real": -0.2700158953666687,
+      "step": 250
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 4.633712660028449e-07,
+      "logits/generated": 0.047906339168548584,
+      "logits/real": -0.762579083442688,
+      "logps/generated": -265.7643127441406,
+      "logps/real": -178.68174743652344,
+      "loss": 0.0227,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -9.940900802612305,
+      "rewards/margins": 9.152986526489258,
+      "rewards/real": -0.7879153490066528,
+      "step": 260
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 4.5981507823613085e-07,
+      "logits/generated": 0.0478428415954113,
+      "logits/real": -0.6810993552207947,
+      "logps/generated": -258.02484130859375,
+      "logps/real": -183.0740509033203,
+      "loss": 0.0224,
+      "rewards/accuracies": 0.987500011920929,
+      "rewards/generated": -10.063767433166504,
+      "rewards/margins": 9.119011878967285,
+      "rewards/real": -0.9447552561759949,
+      "step": 270
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 4.562588904694168e-07,
+      "logits/generated": 0.07144404947757721,
+      "logits/real": -0.6452735662460327,
+      "logps/generated": -260.16082763671875,
+      "logps/real": -188.64410400390625,
+      "loss": 0.0228,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -10.373169898986816,
+      "rewards/margins": 9.765576362609863,
+      "rewards/real": -0.6075931787490845,
+      "step": 280
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 4.5270270270270264e-07,
+      "logits/generated": 0.08266101777553558,
+      "logits/real": -0.716413140296936,
+      "logps/generated": -276.4801330566406,
+      "logps/real": -185.527099609375,
+      "loss": 0.0148,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -11.44568920135498,
+      "rewards/margins": 10.838994979858398,
+      "rewards/real": -0.6066935062408447,
+      "step": 290
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 4.491465149359886e-07,
+      "logits/generated": 0.0009769715834408998,
+      "logits/real": -0.7033424973487854,
+      "logps/generated": -272.52093505859375,
+      "logps/real": -179.6094512939453,
+      "loss": 0.0218,
+      "rewards/accuracies": 0.987500011920929,
+      "rewards/generated": -11.34181022644043,
+      "rewards/margins": 10.347066879272461,
+      "rewards/real": -0.9947425127029419,
+      "step": 300
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 4.4559032716927454e-07,
+      "logits/generated": 0.07658599317073822,
+      "logits/real": -0.5730828046798706,
+      "logps/generated": -276.45294189453125,
+      "logps/real": -200.97845458984375,
+      "loss": 0.0238,
+      "rewards/accuracies": 0.9750000238418579,
+      "rewards/generated": -11.649636268615723,
+      "rewards/margins": 10.637258529663086,
+      "rewards/real": -1.0123790502548218,
+      "step": 310
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 4.420341394025605e-07,
+      "logits/generated": 0.019550871104002,
+      "logits/real": -0.5840874910354614,
+      "logps/generated": -280.53192138671875,
+      "logps/real": -189.03172302246094,
+      "loss": 0.0142,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -11.912101745605469,
+      "rewards/margins": 11.156845092773438,
+      "rewards/real": -0.7552580237388611,
+      "step": 320
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 4.384779516358463e-07,
+      "logits/generated": 0.016371339559555054,
+      "logits/real": -0.5726695656776428,
+      "logps/generated": -266.34918212890625,
+      "logps/real": -202.59814453125,
+      "loss": 0.0145,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -11.33076286315918,
+      "rewards/margins": 9.75381851196289,
+      "rewards/real": -1.576944351196289,
+      "step": 330
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 4.3492176386913227e-07,
+      "logits/generated": 0.0978037491440773,
+      "logits/real": -0.6194095611572266,
+      "logps/generated": -282.8346862792969,
+      "logps/real": -204.97592163085938,
+      "loss": 0.0191,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -12.424463272094727,
+      "rewards/margins": 11.069517135620117,
+      "rewards/real": -1.3549461364746094,
+      "step": 340
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 4.313655761024182e-07,
+      "logits/generated": 0.10168097913265228,
+      "logits/real": -0.625900149345398,
+      "logps/generated": -289.7073974609375,
+      "logps/real": -203.98751831054688,
+      "loss": 0.0158,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -13.3988676071167,
+      "rewards/margins": 11.112601280212402,
+      "rewards/real": -2.2862656116485596,
+      "step": 350
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 4.278093883357041e-07,
+      "logits/generated": 0.05643658712506294,
+      "logits/real": -0.613519549369812,
+      "logps/generated": -285.1255187988281,
+      "logps/real": -188.86361694335938,
+      "loss": 0.0182,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -12.26307487487793,
+      "rewards/margins": 11.37584400177002,
+      "rewards/real": -0.8872316479682922,
+      "step": 360
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 4.2425320056899e-07,
+      "logits/generated": 0.017720462754368782,
+      "logits/real": -0.5340021848678589,
+      "logps/generated": -296.48681640625,
+      "logps/real": -216.097900390625,
+      "loss": 0.0136,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -13.157635688781738,
+      "rewards/margins": 11.516908645629883,
+      "rewards/real": -1.640728235244751,
+      "step": 370
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 4.2069701280227595e-07,
+      "logits/generated": -0.021676432341337204,
+      "logits/real": -0.7345054745674133,
+      "logps/generated": -295.46429443359375,
+      "logps/real": -198.0674591064453,
+      "loss": 0.0108,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -12.902421951293945,
+      "rewards/margins": 11.71843147277832,
+      "rewards/real": -1.1839900016784668,
+      "step": 380
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 4.1714082503556185e-07,
+      "logits/generated": 0.07925084233283997,
+      "logits/real": -0.5841912031173706,
+      "logps/generated": -298.8299255371094,
+      "logps/real": -189.8162384033203,
+      "loss": 0.0155,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -13.550666809082031,
+      "rewards/margins": 12.233985900878906,
+      "rewards/real": -1.3166826963424683,
+      "step": 390
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 4.135846372688478e-07,
+      "logits/generated": 0.025105977430939674,
+      "logits/real": -0.6952486634254456,
+      "logps/generated": -279.43560791015625,
+      "logps/real": -197.62535095214844,
+      "loss": 0.0192,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -12.729209899902344,
+      "rewards/margins": 11.284834861755371,
+      "rewards/real": -1.444373369216919,
+      "step": 400
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 4.100284495021337e-07,
+      "logits/generated": 0.05239884927868843,
+      "logits/real": -0.6132751703262329,
+      "logps/generated": -305.1339416503906,
+      "logps/real": -191.66676330566406,
+      "loss": 0.0079,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -14.304582595825195,
+      "rewards/margins": 12.9876127243042,
+      "rewards/real": -1.3169682025909424,
+      "step": 410
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 4.064722617354196e-07,
+      "logits/generated": 0.034467507153749466,
+      "logits/real": -0.7083422541618347,
+      "logps/generated": -304.05780029296875,
+      "logps/real": -200.8745880126953,
+      "loss": 0.0128,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -14.695897102355957,
+      "rewards/margins": 12.881891250610352,
+      "rewards/real": -1.8140056133270264,
+      "step": 420
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 4.0291607396870553e-07,
+      "logits/generated": 0.03394109755754471,
+      "logits/real": -0.6564615964889526,
+      "logps/generated": -291.5844421386719,
+      "logps/real": -188.79075622558594,
+      "loss": 0.0055,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -13.64738941192627,
+      "rewards/margins": 12.21278190612793,
+      "rewards/real": -1.4346075057983398,
+      "step": 430
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 3.993598862019915e-07,
+      "logits/generated": 0.07299565523862839,
+      "logits/real": -0.6604090332984924,
+      "logps/generated": -309.156982421875,
+      "logps/real": -203.24072265625,
+      "loss": 0.0294,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -15.03246784210205,
+      "rewards/margins": 12.468481063842773,
+      "rewards/real": -2.563986301422119,
+      "step": 440
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 3.9580369843527737e-07,
+      "logits/generated": 0.0038179433904588223,
+      "logits/real": -0.6290857195854187,
+      "logps/generated": -326.712890625,
+      "logps/real": -202.99484252929688,
+      "loss": 0.0064,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -15.489263534545898,
+      "rewards/margins": 12.454556465148926,
+      "rewards/real": -3.0347084999084473,
+      "step": 450
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 3.9224751066856327e-07,
+      "logits/generated": 0.031186867505311966,
+      "logits/real": -0.4273042678833008,
+      "logps/generated": -323.9992980957031,
+      "logps/real": -225.23263549804688,
+      "loss": 0.0108,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -15.867002487182617,
+      "rewards/margins": 13.411462783813477,
+      "rewards/real": -2.455543041229248,
+      "step": 460
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 3.886913229018492e-07,
+      "logits/generated": 0.011080889031291008,
+      "logits/real": -0.6169866919517517,
+      "logps/generated": -313.5555114746094,
+      "logps/real": -195.68914794921875,
+      "loss": 0.0104,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -15.070566177368164,
+      "rewards/margins": 13.439569473266602,
+      "rewards/real": -1.630995750427246,
+      "step": 470
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 3.851351351351351e-07,
+      "logits/generated": -0.029481088742613792,
+      "logits/real": -0.6428096890449524,
+      "logps/generated": -306.30755615234375,
+      "logps/real": -198.16749572753906,
+      "loss": 0.0107,
+      "rewards/accuracies": 0.987500011920929,
+      "rewards/generated": -14.478108406066895,
+      "rewards/margins": 12.663415908813477,
+      "rewards/real": -1.814692497253418,
+      "step": 480
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 3.8157894736842105e-07,
+      "logits/generated": 0.015122579410672188,
+      "logits/real": -0.6262849569320679,
+      "logps/generated": -301.0884704589844,
+      "logps/real": -190.44667053222656,
+      "loss": 0.0176,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -13.957537651062012,
+      "rewards/margins": 11.782486915588379,
+      "rewards/real": -2.1750526428222656,
+      "step": 490
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 3.7802275960170695e-07,
+      "logits/generated": -0.011642997153103352,
+      "logits/real": -0.5468995571136475,
+      "logps/generated": -322.72308349609375,
+      "logps/real": -188.96719360351562,
+      "loss": 0.0098,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -15.367193222045898,
+      "rewards/margins": 14.392413139343262,
+      "rewards/real": -0.9747812151908875,
+      "step": 500
+    },
+    {
+      "epoch": 0.32,
+      "eval_logits/generated": 0.015148750506341457,
+      "eval_logits/real": -0.618992805480957,
+      "eval_logps/generated": -304.2368469238281,
+      "eval_logps/real": -186.9757843017578,
+      "eval_loss": 0.010758413933217525,
+      "eval_rewards/accuracies": 0.9976114630699158,
+      "eval_rewards/generated": -14.384950637817383,
+      "eval_rewards/margins": 13.620210647583008,
+      "eval_rewards/real": -0.7647396922111511,
+      "eval_runtime": 424.798,
+      "eval_samples_per_second": 11.77,
+      "eval_steps_per_second": 0.37,
+      "step": 500
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 3.7446657183499284e-07,
+      "logits/generated": 0.04711627587676048,
+      "logits/real": -0.6923630833625793,
+      "logps/generated": -311.3076171875,
+      "logps/real": -192.34048461914062,
+      "loss": 0.0038,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -15.26708698272705,
+      "rewards/margins": 13.872920036315918,
+      "rewards/real": -1.394165277481079,
+      "step": 510
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 3.709103840682788e-07,
+      "logits/generated": 0.03908165544271469,
+      "logits/real": -0.631860613822937,
+      "logps/generated": -310.2327880859375,
+      "logps/real": -186.8871612548828,
+      "loss": 0.0051,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -15.100445747375488,
+      "rewards/margins": 13.6294527053833,
+      "rewards/real": -1.470994234085083,
+      "step": 520
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 3.6735419630156474e-07,
+      "logits/generated": 0.04315485060214996,
+      "logits/real": -0.5955843329429626,
+      "logps/generated": -324.2378845214844,
+      "logps/real": -187.12576293945312,
+      "loss": 0.0065,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -16.153406143188477,
+      "rewards/margins": 14.313261032104492,
+      "rewards/real": -1.8401434421539307,
+      "step": 530
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 3.637980085348506e-07,
+      "logits/generated": -0.03953739255666733,
+      "logits/real": -0.6422590017318726,
+      "logps/generated": -323.29638671875,
+      "logps/real": -203.27786254882812,
+      "loss": 0.0136,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -16.259048461914062,
+      "rewards/margins": 13.929720878601074,
+      "rewards/real": -2.3293280601501465,
+      "step": 540
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 3.602418207681365e-07,
+      "logits/generated": -0.04372464120388031,
+      "logits/real": -0.6528729796409607,
+      "logps/generated": -336.2181701660156,
+      "logps/real": -208.175048828125,
+      "loss": 0.0094,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -16.822994232177734,
+      "rewards/margins": 14.49010944366455,
+      "rewards/real": -2.3328843116760254,
+      "step": 550
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 3.5668563300142247e-07,
+      "logits/generated": -0.020463664084672928,
+      "logits/real": -0.5609344244003296,
+      "logps/generated": -317.5655822753906,
+      "logps/real": -197.87765502929688,
+      "loss": 0.0085,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -15.073542594909668,
+      "rewards/margins": 12.639741897583008,
+      "rewards/real": -2.433799982070923,
+      "step": 560
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 3.5312944523470837e-07,
+      "logits/generated": 0.016961723566055298,
+      "logits/real": -0.7112401723861694,
+      "logps/generated": -322.7782897949219,
+      "logps/real": -191.416015625,
+      "loss": 0.0074,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -15.618112564086914,
+      "rewards/margins": 14.197979927062988,
+      "rewards/real": -1.4201303720474243,
+      "step": 570
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 3.495732574679943e-07,
+      "logits/generated": 0.03125763684511185,
+      "logits/real": -0.6455451250076294,
+      "logps/generated": -309.74517822265625,
+      "logps/real": -192.3853302001953,
+      "loss": 0.012,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -15.057307243347168,
+      "rewards/margins": 13.615495681762695,
+      "rewards/real": -1.4418113231658936,
+      "step": 580
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 3.460170697012802e-07,
+      "logits/generated": 0.030627410858869553,
+      "logits/real": -0.6639117002487183,
+      "logps/generated": -330.22894287109375,
+      "logps/real": -195.18409729003906,
+      "loss": 0.0092,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -16.0391902923584,
+      "rewards/margins": 14.437360763549805,
+      "rewards/real": -1.6018317937850952,
+      "step": 590
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 3.424608819345661e-07,
+      "logits/generated": 0.06100524589419365,
+      "logits/real": -0.6973519325256348,
+      "logps/generated": -319.52166748046875,
+      "logps/real": -179.44314575195312,
+      "loss": 0.0041,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -15.46537971496582,
+      "rewards/margins": 14.726943969726562,
+      "rewards/real": -0.7384368777275085,
+      "step": 600
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 3.3890469416785205e-07,
+      "logits/generated": 0.06144358962774277,
+      "logits/real": -0.6208174228668213,
+      "logps/generated": -311.03912353515625,
+      "logps/real": -192.3145751953125,
+      "loss": 0.0037,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -15.110832214355469,
+      "rewards/margins": 13.566309928894043,
+      "rewards/real": -1.544521689414978,
+      "step": 610
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 3.35348506401138e-07,
+      "logits/generated": -0.018809977918863297,
+      "logits/real": -0.6647375822067261,
+      "logps/generated": -331.81158447265625,
+      "logps/real": -182.822021484375,
+      "loss": 0.0105,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -16.408363342285156,
+      "rewards/margins": 15.543344497680664,
+      "rewards/real": -0.8650201559066772,
+      "step": 620
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 3.3179231863442384e-07,
+      "logits/generated": -0.014424433931708336,
+      "logits/real": -0.5232574939727783,
+      "logps/generated": -321.45355224609375,
+      "logps/real": -204.66917419433594,
+      "loss": 0.0055,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -16.32326889038086,
+      "rewards/margins": 14.821706771850586,
+      "rewards/real": -1.501560091972351,
+      "step": 630
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 3.282361308677098e-07,
+      "logits/generated": -0.04307156428694725,
+      "logits/real": -0.6383107900619507,
+      "logps/generated": -332.80963134765625,
+      "logps/real": -191.7883758544922,
+      "loss": 0.0052,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -15.911900520324707,
+      "rewards/margins": 14.832717895507812,
+      "rewards/real": -1.0791819095611572,
+      "step": 640
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 3.2467994310099573e-07,
+      "logits/generated": 0.020891521126031876,
+      "logits/real": -0.6430577039718628,
+      "logps/generated": -350.6064147949219,
+      "logps/real": -215.30593872070312,
+      "loss": 0.0097,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -18.383892059326172,
+      "rewards/margins": 16.04793357849121,
+      "rewards/real": -2.3359580039978027,
+      "step": 650
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 3.211237553342817e-07,
+      "logits/generated": -0.024233415722846985,
+      "logits/real": -0.6739251017570496,
+      "logps/generated": -333.5208435058594,
+      "logps/real": -194.91732788085938,
+      "loss": 0.0092,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -16.731473922729492,
+      "rewards/margins": 14.82574462890625,
+      "rewards/real": -1.9057306051254272,
+      "step": 660
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 3.175675675675675e-07,
+      "logits/generated": 0.023840907961130142,
+      "logits/real": -0.6880885362625122,
+      "logps/generated": -318.2210998535156,
+      "logps/real": -184.1695556640625,
+      "loss": 0.0089,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -16.254070281982422,
+      "rewards/margins": 14.820034980773926,
+      "rewards/real": -1.4340364933013916,
+      "step": 670
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 3.1401137980085347e-07,
+      "logits/generated": 0.040734268724918365,
+      "logits/real": -0.6580570340156555,
+      "logps/generated": -342.2274475097656,
+      "logps/real": -205.65689086914062,
+      "loss": 0.011,
+      "rewards/accuracies": 0.987500011920929,
+      "rewards/generated": -17.80494499206543,
+      "rewards/margins": 15.115242004394531,
+      "rewards/real": -2.689703941345215,
+      "step": 680
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 3.104551920341394e-07,
+      "logits/generated": -0.019141068682074547,
+      "logits/real": -0.6394578814506531,
+      "logps/generated": -324.29833984375,
+      "logps/real": -201.87010192871094,
+      "loss": 0.0021,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -16.063854217529297,
+      "rewards/margins": 14.041638374328613,
+      "rewards/real": -2.0222160816192627,
+      "step": 690
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 3.068990042674253e-07,
+      "logits/generated": -0.02943194843828678,
+      "logits/real": -0.6647931337356567,
+      "logps/generated": -309.67327880859375,
+      "logps/real": -181.3977508544922,
+      "loss": 0.0109,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -15.244012832641602,
+      "rewards/margins": 13.764185905456543,
+      "rewards/real": -1.4798262119293213,
+      "step": 700
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 3.033428165007112e-07,
+      "logits/generated": -0.0134804155677557,
+      "logits/real": -0.688398003578186,
+      "logps/generated": -333.3070373535156,
+      "logps/real": -196.06004333496094,
+      "loss": 0.0058,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -16.678394317626953,
+      "rewards/margins": 14.339719772338867,
+      "rewards/real": -2.3386740684509277,
+      "step": 710
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 2.9978662873399715e-07,
+      "logits/generated": -0.01525292731821537,
+      "logits/real": -0.5911905169487,
+      "logps/generated": -330.7987060546875,
+      "logps/real": -198.35397338867188,
+      "loss": 0.0048,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -16.70512580871582,
+      "rewards/margins": 14.928197860717773,
+      "rewards/real": -1.776925802230835,
+      "step": 720
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 2.9623044096728305e-07,
+      "logits/generated": 0.047141142189502716,
+      "logits/real": -0.5441254377365112,
+      "logps/generated": -320.4557800292969,
+      "logps/real": -211.35848999023438,
+      "loss": 0.0026,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -16.464683532714844,
+      "rewards/margins": 14.595657348632812,
+      "rewards/real": -1.8690249919891357,
+      "step": 730
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 2.92674253200569e-07,
+      "logits/generated": 0.018107902258634567,
+      "logits/real": -0.6069762110710144,
+      "logps/generated": -335.17401123046875,
+      "logps/real": -191.33583068847656,
+      "loss": 0.001,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -17.258358001708984,
+      "rewards/margins": 15.983779907226562,
+      "rewards/real": -1.2745764255523682,
+      "step": 740
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 2.8911806543385494e-07,
+      "logits/generated": 0.05721588060259819,
+      "logits/real": -0.6501365900039673,
+      "logps/generated": -334.0870666503906,
+      "logps/real": -201.9585418701172,
+      "loss": 0.0045,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -16.955425262451172,
+      "rewards/margins": 16.156780242919922,
+      "rewards/real": -0.7986453175544739,
+      "step": 750
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 2.855618776671408e-07,
+      "logits/generated": -0.00029001757502555847,
+      "logits/real": -0.4777015745639801,
+      "logps/generated": -349.4190368652344,
+      "logps/real": -214.5007781982422,
+      "loss": 0.0033,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -18.192649841308594,
+      "rewards/margins": 16.606849670410156,
+      "rewards/real": -1.5857971906661987,
+      "step": 760
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 2.8200568990042673e-07,
+      "logits/generated": 0.006303996779024601,
+      "logits/real": -0.6401196718215942,
+      "logps/generated": -335.5325622558594,
+      "logps/real": -194.22637939453125,
+      "loss": 0.0045,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -17.122852325439453,
+      "rewards/margins": 15.594779968261719,
+      "rewards/real": -1.5280735492706299,
+      "step": 770
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 2.784495021337127e-07,
+      "logits/generated": -0.015740731731057167,
+      "logits/real": -0.6430305242538452,
+      "logps/generated": -333.03778076171875,
+      "logps/real": -185.57615661621094,
+      "loss": 0.006,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -17.422563552856445,
+      "rewards/margins": 15.94238567352295,
+      "rewards/real": -1.4801769256591797,
+      "step": 780
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 2.7489331436699857e-07,
+      "logits/generated": -0.06366153061389923,
+      "logits/real": -0.5953705310821533,
+      "logps/generated": -346.43719482421875,
+      "logps/real": -207.64895629882812,
+      "loss": 0.0044,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -17.68976402282715,
+      "rewards/margins": 15.903286933898926,
+      "rewards/real": -1.7864751815795898,
+      "step": 790
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 2.7133712660028446e-07,
+      "logits/generated": 0.05433814972639084,
+      "logits/real": -0.659256637096405,
+      "logps/generated": -347.67437744140625,
+      "logps/real": -188.8865509033203,
+      "loss": 0.0055,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -18.18773651123047,
+      "rewards/margins": 16.599082946777344,
+      "rewards/real": -1.5886526107788086,
+      "step": 800
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 2.677809388335704e-07,
+      "logits/generated": -0.006709927227348089,
+      "logits/real": -0.6219618916511536,
+      "logps/generated": -339.08404541015625,
+      "logps/real": -192.36767578125,
+      "loss": 0.0044,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -17.780960083007812,
+      "rewards/margins": 15.594240188598633,
+      "rewards/real": -2.186721086502075,
+      "step": 810
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 2.642247510668563e-07,
+      "logits/generated": 0.024327615275979042,
+      "logits/real": -0.6300166845321655,
+      "logps/generated": -359.64288330078125,
+      "logps/real": -189.8053741455078,
+      "loss": 0.0027,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -19.835926055908203,
+      "rewards/margins": 18.566604614257812,
+      "rewards/real": -1.2693183422088623,
+      "step": 820
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 2.6066856330014225e-07,
+      "logits/generated": 0.060337960720062256,
+      "logits/real": -0.6221122145652771,
+      "logps/generated": -346.8661193847656,
+      "logps/real": -193.4240264892578,
+      "loss": 0.0046,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -18.56460952758789,
+      "rewards/margins": 17.463274002075195,
+      "rewards/real": -1.1013351678848267,
+      "step": 830
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 2.5711237553342815e-07,
+      "logits/generated": -0.023245109245181084,
+      "logits/real": -0.5685423612594604,
+      "logps/generated": -363.9253234863281,
+      "logps/real": -206.6332244873047,
+      "loss": 0.0058,
+      "rewards/accuracies": 0.987500011920929,
+      "rewards/generated": -18.62310791015625,
+      "rewards/margins": 17.27718162536621,
+      "rewards/real": -1.3459270000457764,
+      "step": 840
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 2.5355618776671404e-07,
+      "logits/generated": 0.05930706113576889,
+      "logits/real": -0.6789587736129761,
+      "logps/generated": -344.4968566894531,
+      "logps/real": -199.61949157714844,
+      "loss": 0.0038,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -19.401254653930664,
+      "rewards/margins": 17.86998748779297,
+      "rewards/real": -1.5312663316726685,
+      "step": 850
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 2.5e-07,
+      "logits/generated": -0.02314385026693344,
+      "logits/real": -0.5737181305885315,
+      "logps/generated": -353.2101135253906,
+      "logps/real": -192.0513458251953,
+      "loss": 0.0023,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -19.10390853881836,
+      "rewards/margins": 17.52272605895996,
+      "rewards/real": -1.581182599067688,
+      "step": 860
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 2.4644381223328594e-07,
+      "logits/generated": 0.058255720883607864,
+      "logits/real": -0.5887473821640015,
+      "logps/generated": -358.34857177734375,
+      "logps/real": -195.61085510253906,
+      "loss": 0.003,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -19.416332244873047,
+      "rewards/margins": 18.28666877746582,
+      "rewards/real": -1.129664421081543,
+      "step": 870
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 2.4288762446657183e-07,
+      "logits/generated": 0.08289220184087753,
+      "logits/real": -0.5633417367935181,
+      "logps/generated": -371.24578857421875,
+      "logps/real": -209.87020874023438,
+      "loss": 0.0021,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -21.660175323486328,
+      "rewards/margins": 19.35919189453125,
+      "rewards/real": -2.3009822368621826,
+      "step": 880
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 2.393314366998578e-07,
+      "logits/generated": 0.05838945508003235,
+      "logits/real": -0.5614827871322632,
+      "logps/generated": -385.99591064453125,
+      "logps/real": -205.49008178710938,
+      "loss": 0.0049,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -22.40130043029785,
+      "rewards/margins": 19.79316520690918,
+      "rewards/real": -2.6081345081329346,
+      "step": 890
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 2.3577524893314365e-07,
+      "logits/generated": 0.0925709456205368,
+      "logits/real": -0.6390553712844849,
+      "logps/generated": -385.9515075683594,
+      "logps/real": -199.3788299560547,
+      "loss": 0.0059,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -22.699363708496094,
+      "rewards/margins": 20.273067474365234,
+      "rewards/real": -2.4262948036193848,
+      "step": 900
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 2.322190611664296e-07,
+      "logits/generated": -0.001485310262069106,
+      "logits/real": -0.4676692485809326,
+      "logps/generated": -368.5473937988281,
+      "logps/real": -213.4370880126953,
+      "loss": 0.0043,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -21.21763801574707,
+      "rewards/margins": 19.124008178710938,
+      "rewards/real": -2.0936279296875,
+      "step": 910
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 2.2866287339971549e-07,
+      "logits/generated": -0.01602059043943882,
+      "logits/real": -0.6552165150642395,
+      "logps/generated": -399.9144592285156,
+      "logps/real": -208.2056121826172,
+      "loss": 0.0007,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -23.339561462402344,
+      "rewards/margins": 20.945171356201172,
+      "rewards/real": -2.3943886756896973,
+      "step": 920
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 2.251066856330014e-07,
+      "logits/generated": 0.05472123622894287,
+      "logits/real": -0.5271707773208618,
+      "logps/generated": -377.61981201171875,
+      "logps/real": -199.87130737304688,
+      "loss": 0.002,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -22.324268341064453,
+      "rewards/margins": 19.352123260498047,
+      "rewards/real": -2.972146511077881,
+      "step": 930
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 2.2155049786628733e-07,
+      "logits/generated": -0.025077398866415024,
+      "logits/real": -0.5260539054870605,
+      "logps/generated": -394.1059265136719,
+      "logps/real": -210.8734130859375,
+      "loss": 0.0032,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -23.064285278320312,
+      "rewards/margins": 20.856571197509766,
+      "rewards/real": -2.2077155113220215,
+      "step": 940
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 2.1799431009957325e-07,
+      "logits/generated": 0.008644811809062958,
+      "logits/real": -0.5587931275367737,
+      "logps/generated": -402.0521545410156,
+      "logps/real": -202.02291870117188,
+      "loss": 0.0037,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -23.23007583618164,
+      "rewards/margins": 19.87685203552246,
+      "rewards/real": -3.3532238006591797,
+      "step": 950
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 2.1443812233285914e-07,
+      "logits/generated": -0.009537003934383392,
+      "logits/real": -0.4732537269592285,
+      "logps/generated": -380.66107177734375,
+      "logps/real": -210.02407836914062,
+      "loss": 0.0035,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -22.10177993774414,
+      "rewards/margins": 19.873910903930664,
+      "rewards/real": -2.2278692722320557,
+      "step": 960
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 2.108819345661451e-07,
+      "logits/generated": 0.04497765749692917,
+      "logits/real": -0.5100663900375366,
+      "logps/generated": -383.2145080566406,
+      "logps/real": -200.04031372070312,
+      "loss": 0.0011,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -21.68323516845703,
+      "rewards/margins": 19.536602020263672,
+      "rewards/real": -2.146634340286255,
+      "step": 970
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 2.0732574679943098e-07,
+      "logits/generated": 0.051282238215208054,
+      "logits/real": -0.7591557502746582,
+      "logps/generated": -358.7184143066406,
+      "logps/real": -178.8624267578125,
+      "loss": 0.0028,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -19.63443374633789,
+      "rewards/margins": 18.222341537475586,
+      "rewards/real": -1.4120899438858032,
+      "step": 980
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 2.0376955903271693e-07,
+      "logits/generated": 0.05088866874575615,
+      "logits/real": -0.4944595694541931,
+      "logps/generated": -370.34417724609375,
+      "logps/real": -208.4033966064453,
+      "loss": 0.005,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -21.147663116455078,
+      "rewards/margins": 19.88173484802246,
+      "rewards/real": -1.2659282684326172,
+      "step": 990
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 2.0021337126600283e-07,
+      "logits/generated": 0.0677376538515091,
+      "logits/real": -0.5833539366722107,
+      "logps/generated": -366.4508361816406,
+      "logps/real": -182.74696350097656,
+      "loss": 0.0019,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -20.16263198852539,
+      "rewards/margins": 18.523616790771484,
+      "rewards/real": -1.6390106678009033,
+      "step": 1000
+    },
+    {
+      "epoch": 0.64,
+      "eval_logits/generated": 0.04800041764974594,
+      "eval_logits/real": -0.5656154155731201,
+      "eval_logps/generated": -376.97998046875,
+      "eval_logps/real": -197.0876007080078,
+      "eval_loss": 0.004381492733955383,
+      "eval_rewards/accuracies": 0.9984076619148254,
+      "eval_rewards/generated": -21.65926742553711,
+      "eval_rewards/margins": 19.883346557617188,
+      "eval_rewards/real": -1.7759193181991577,
+      "eval_runtime": 321.6683,
+      "eval_samples_per_second": 15.544,
+      "eval_steps_per_second": 0.488,
+      "step": 1000
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 1.9665718349928875e-07,
+      "logits/generated": 0.10137276351451874,
+      "logits/real": -0.5880488753318787,
+      "logps/generated": -336.6202392578125,
+      "logps/real": -179.52365112304688,
+      "loss": 0.0111,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -18.403759002685547,
+      "rewards/margins": 17.433141708374023,
+      "rewards/real": -0.9706158638000488,
+      "step": 1010
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 1.931009957325747e-07,
+      "logits/generated": 0.07191314548254013,
+      "logits/real": -0.7045632600784302,
+      "logps/generated": -371.539306640625,
+      "logps/real": -174.90870666503906,
+      "loss": 0.0035,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -20.966075897216797,
+      "rewards/margins": 19.68502426147461,
+      "rewards/real": -1.2810522317886353,
+      "step": 1020
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 1.895448079658606e-07,
+      "logits/generated": 0.021189400926232338,
+      "logits/real": -0.5944398641586304,
+      "logps/generated": -345.79901123046875,
+      "logps/real": -197.9308319091797,
+      "loss": 0.0123,
+      "rewards/accuracies": 0.9750000238418579,
+      "rewards/generated": -18.253501892089844,
+      "rewards/margins": 16.816665649414062,
+      "rewards/real": -1.436837077140808,
+      "step": 1030
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 1.859886201991465e-07,
+      "logits/generated": 0.028367796912789345,
+      "logits/real": -0.5693169832229614,
+      "logps/generated": -369.0507507324219,
+      "logps/real": -203.1673126220703,
+      "loss": 0.001,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -20.233728408813477,
+      "rewards/margins": 18.871437072753906,
+      "rewards/real": -1.3622897863388062,
+      "step": 1040
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 1.8243243243243243e-07,
+      "logits/generated": 0.11130674183368683,
+      "logits/real": -0.6199926137924194,
+      "logps/generated": -396.66668701171875,
+      "logps/real": -199.72732543945312,
+      "loss": 0.007,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -22.905460357666016,
+      "rewards/margins": 21.137920379638672,
+      "rewards/real": -1.767538070678711,
+      "step": 1050
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 1.7887624466571835e-07,
+      "logits/generated": 0.051538754254579544,
+      "logits/real": -0.5638888478279114,
+      "logps/generated": -382.5877380371094,
+      "logps/real": -193.47872924804688,
+      "loss": 0.0021,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -22.123483657836914,
+      "rewards/margins": 20.473840713500977,
+      "rewards/real": -1.6496423482894897,
+      "step": 1060
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 1.7532005689900424e-07,
+      "logits/generated": 0.056547343730926514,
+      "logits/real": -0.5736340284347534,
+      "logps/generated": -399.03253173828125,
+      "logps/real": -205.837158203125,
+      "loss": 0.0024,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -23.35310173034668,
+      "rewards/margins": 20.6170597076416,
+      "rewards/real": -2.7360422611236572,
+      "step": 1070
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 1.717638691322902e-07,
+      "logits/generated": 0.12326414883136749,
+      "logits/real": -0.5854828357696533,
+      "logps/generated": -368.6696472167969,
+      "logps/real": -194.57887268066406,
+      "loss": 0.0035,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -21.72028923034668,
+      "rewards/margins": 19.422595977783203,
+      "rewards/real": -2.297696352005005,
+      "step": 1080
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 1.6820768136557609e-07,
+      "logits/generated": 0.03324466198682785,
+      "logits/real": -0.5589950680732727,
+      "logps/generated": -379.6975402832031,
+      "logps/real": -189.21792602539062,
+      "loss": 0.0013,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -20.778305053710938,
+      "rewards/margins": 19.09261131286621,
+      "rewards/real": -1.6856931447982788,
+      "step": 1090
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 1.64651493598862e-07,
+      "logits/generated": 0.02137361653149128,
+      "logits/real": -0.4791427552700043,
+      "logps/generated": -372.0145263671875,
+      "logps/real": -201.42239379882812,
+      "loss": 0.0017,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -20.45681381225586,
+      "rewards/margins": 19.019607543945312,
+      "rewards/real": -1.4372069835662842,
+      "step": 1100
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 1.6109530583214793e-07,
+      "logits/generated": 0.06219879537820816,
+      "logits/real": -0.6077834367752075,
+      "logps/generated": -361.52496337890625,
+      "logps/real": -187.40945434570312,
+      "loss": 0.0017,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -20.687042236328125,
+      "rewards/margins": 19.435745239257812,
+      "rewards/real": -1.2512991428375244,
+      "step": 1110
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 1.5753911806543385e-07,
+      "logits/generated": 0.02082439325749874,
+      "logits/real": -0.6671017408370972,
+      "logps/generated": -399.9078369140625,
+      "logps/real": -197.09945678710938,
+      "loss": 0.0014,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -23.697898864746094,
+      "rewards/margins": 21.734453201293945,
+      "rewards/real": -1.9634456634521484,
+      "step": 1120
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 1.5398293029871974e-07,
+      "logits/generated": 0.07103635370731354,
+      "logits/real": -0.5315567851066589,
+      "logps/generated": -390.80413818359375,
+      "logps/real": -209.8661651611328,
+      "loss": 0.0032,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -23.044010162353516,
+      "rewards/margins": 20.588970184326172,
+      "rewards/real": -2.4550397396087646,
+      "step": 1130
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 1.504267425320057e-07,
+      "logits/generated": 0.05514199659228325,
+      "logits/real": -0.5959967374801636,
+      "logps/generated": -372.26470947265625,
+      "logps/real": -187.91561889648438,
+      "loss": 0.0063,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -21.262130737304688,
+      "rewards/margins": 19.89595603942871,
+      "rewards/real": -1.366172194480896,
+      "step": 1140
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1.4687055476529158e-07,
+      "logits/generated": 0.03651849180459976,
+      "logits/real": -0.5296968817710876,
+      "logps/generated": -375.4384765625,
+      "logps/real": -198.07846069335938,
+      "loss": 0.0009,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -20.608938217163086,
+      "rewards/margins": 19.235652923583984,
+      "rewards/real": -1.373286247253418,
+      "step": 1150
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1.4331436699857753e-07,
+      "logits/generated": -0.031901903450489044,
+      "logits/real": -0.42245230078697205,
+      "logps/generated": -385.30987548828125,
+      "logps/real": -198.95516967773438,
+      "loss": 0.0011,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -21.40415382385254,
+      "rewards/margins": 19.745981216430664,
+      "rewards/real": -1.6581722497940063,
+      "step": 1160
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 1.3975817923186345e-07,
+      "logits/generated": 0.07458638399839401,
+      "logits/real": -0.4577251970767975,
+      "logps/generated": -376.2591552734375,
+      "logps/real": -211.70458984375,
+      "loss": 0.0037,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -21.238889694213867,
+      "rewards/margins": 19.84615135192871,
+      "rewards/real": -1.3927379846572876,
+      "step": 1170
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 1.3620199146514935e-07,
+      "logits/generated": 0.06895387917757034,
+      "logits/real": -0.4532155990600586,
+      "logps/generated": -361.93939208984375,
+      "logps/real": -179.66732788085938,
+      "loss": 0.0018,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -20.205707550048828,
+      "rewards/margins": 18.50905990600586,
+      "rewards/real": -1.6966466903686523,
+      "step": 1180
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 1.326458036984353e-07,
+      "logits/generated": 0.026944806799292564,
+      "logits/real": -0.5042958855628967,
+      "logps/generated": -378.512939453125,
+      "logps/real": -193.09017944335938,
+      "loss": 0.0017,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -21.360944747924805,
+      "rewards/margins": 19.692447662353516,
+      "rewards/real": -1.6684958934783936,
+      "step": 1190
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 1.290896159317212e-07,
+      "logits/generated": 0.05960095673799515,
+      "logits/real": -0.5585105419158936,
+      "logps/generated": -394.94696044921875,
+      "logps/real": -191.21163940429688,
+      "loss": 0.0009,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -23.144710540771484,
+      "rewards/margins": 21.34160041809082,
+      "rewards/real": -1.8031113147735596,
+      "step": 1200
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 1.255334281650071e-07,
+      "logits/generated": 0.04382283240556717,
+      "logits/real": -0.4478573203086853,
+      "logps/generated": -375.7854309082031,
+      "logps/real": -205.15869140625,
+      "loss": 0.0006,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -21.472270965576172,
+      "rewards/margins": 19.42158317565918,
+      "rewards/real": -2.050690174102783,
+      "step": 1210
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 1.2197724039829303e-07,
+      "logits/generated": 0.07887273281812668,
+      "logits/real": -0.6188726425170898,
+      "logps/generated": -378.6775817871094,
+      "logps/real": -190.72250366210938,
+      "loss": 0.0053,
+      "rewards/accuracies": 0.987500011920929,
+      "rewards/generated": -21.472759246826172,
+      "rewards/margins": 19.55636215209961,
+      "rewards/real": -1.916398286819458,
+      "step": 1220
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 1.1842105263157894e-07,
+      "logits/generated": 0.0605277419090271,
+      "logits/real": -0.6036852598190308,
+      "logps/generated": -369.0652160644531,
+      "logps/real": -183.35617065429688,
+      "loss": 0.001,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -20.453426361083984,
+      "rewards/margins": 18.384937286376953,
+      "rewards/real": -2.0684916973114014,
+      "step": 1230
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 1.1486486486486487e-07,
+      "logits/generated": 0.06586415320634842,
+      "logits/real": -0.5749965906143188,
+      "logps/generated": -372.5237121582031,
+      "logps/real": -195.25076293945312,
+      "loss": 0.0043,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -20.654064178466797,
+      "rewards/margins": 19.203500747680664,
+      "rewards/real": -1.450567364692688,
+      "step": 1240
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 1.1130867709815078e-07,
+      "logits/generated": 0.08540566265583038,
+      "logits/real": -0.6170912981033325,
+      "logps/generated": -374.17303466796875,
+      "logps/real": -194.3673095703125,
+      "loss": 0.0005,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -21.099042892456055,
+      "rewards/margins": 19.65339469909668,
+      "rewards/real": -1.4456470012664795,
+      "step": 1250
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 1.077524893314367e-07,
+      "logits/generated": 0.06357467174530029,
+      "logits/real": -0.687986433506012,
+      "logps/generated": -392.5172424316406,
+      "logps/real": -191.45401000976562,
+      "loss": 0.0031,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -22.80368423461914,
+      "rewards/margins": 20.335935592651367,
+      "rewards/real": -2.467747688293457,
+      "step": 1260
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 1.0419630156472262e-07,
+      "logits/generated": 0.016886264085769653,
+      "logits/real": -0.4550386965274811,
+      "logps/generated": -390.584228515625,
+      "logps/real": -218.6328887939453,
+      "loss": 0.0019,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -23.16832733154297,
+      "rewards/margins": 20.595144271850586,
+      "rewards/real": -2.5731775760650635,
+      "step": 1270
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 1.0064011379800854e-07,
+      "logits/generated": -0.022652573883533478,
+      "logits/real": -0.4286680817604065,
+      "logps/generated": -379.67706298828125,
+      "logps/real": -206.49093627929688,
+      "loss": 0.0043,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -21.629016876220703,
+      "rewards/margins": 19.87604522705078,
+      "rewards/real": -1.7529706954956055,
+      "step": 1280
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 9.708392603129445e-08,
+      "logits/generated": 0.056550562381744385,
+      "logits/real": -0.4019811749458313,
+      "logps/generated": -370.08251953125,
+      "logps/real": -197.77896118164062,
+      "loss": 0.0026,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -21.446491241455078,
+      "rewards/margins": 19.191186904907227,
+      "rewards/real": -2.255300521850586,
+      "step": 1290
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 9.352773826458037e-08,
+      "logits/generated": 0.008305387571454048,
+      "logits/real": -0.5809749364852905,
+      "logps/generated": -400.37347412109375,
+      "logps/real": -197.43136596679688,
+      "loss": 0.0018,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -23.287565231323242,
+      "rewards/margins": 21.73343276977539,
+      "rewards/real": -1.5541306734085083,
+      "step": 1300
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 8.997155049786629e-08,
+      "logits/generated": 0.046849604696035385,
+      "logits/real": -0.6843993663787842,
+      "logps/generated": -390.62030029296875,
+      "logps/real": -181.0294647216797,
+      "loss": 0.0007,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -22.432353973388672,
+      "rewards/margins": 21.127017974853516,
+      "rewards/real": -1.3053334951400757,
+      "step": 1310
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 8.64153627311522e-08,
+      "logits/generated": 0.052729617804288864,
+      "logits/real": -0.5115618109703064,
+      "logps/generated": -375.7951354980469,
+      "logps/real": -196.12313842773438,
+      "loss": 0.0002,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -21.45965003967285,
+      "rewards/margins": 19.496532440185547,
+      "rewards/real": -1.9631179571151733,
+      "step": 1320
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 8.285917496443812e-08,
+      "logits/generated": 0.035291388630867004,
+      "logits/real": -0.5863553881645203,
+      "logps/generated": -395.09197998046875,
+      "logps/real": -202.11160278320312,
+      "loss": 0.0013,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -22.766231536865234,
+      "rewards/margins": 20.85399627685547,
+      "rewards/real": -1.912235975265503,
+      "step": 1330
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 7.930298719772404e-08,
+      "logits/generated": 0.07267922163009644,
+      "logits/real": -0.49688243865966797,
+      "logps/generated": -373.9420166015625,
+      "logps/real": -205.9117889404297,
+      "loss": 0.0013,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -21.421688079833984,
+      "rewards/margins": 19.645835876464844,
+      "rewards/real": -1.7758514881134033,
+      "step": 1340
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 7.574679943100994e-08,
+      "logits/generated": 0.07065759599208832,
+      "logits/real": -0.5725045204162598,
+      "logps/generated": -378.9009704589844,
+      "logps/real": -181.6143035888672,
+      "loss": 0.0013,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -22.113445281982422,
+      "rewards/margins": 20.17637062072754,
+      "rewards/real": -1.937076210975647,
+      "step": 1350
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 7.219061166429587e-08,
+      "logits/generated": 0.05919628217816353,
+      "logits/real": -0.5659859776496887,
+      "logps/generated": -379.78924560546875,
+      "logps/real": -190.56930541992188,
+      "loss": 0.0016,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -22.432254791259766,
+      "rewards/margins": 20.832565307617188,
+      "rewards/real": -1.5996865034103394,
+      "step": 1360
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 6.863442389758179e-08,
+      "logits/generated": 0.044001154601573944,
+      "logits/real": -0.5443329215049744,
+      "logps/generated": -378.08587646484375,
+      "logps/real": -204.49148559570312,
+      "loss": 0.0139,
+      "rewards/accuracies": 0.987500011920929,
+      "rewards/generated": -21.367244720458984,
+      "rewards/margins": 19.974727630615234,
+      "rewards/real": -1.3925195932388306,
+      "step": 1370
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 6.507823613086771e-08,
+      "logits/generated": 0.046902261674404144,
+      "logits/real": -0.5738197565078735,
+      "logps/generated": -380.58148193359375,
+      "logps/real": -207.26724243164062,
+      "loss": 0.0022,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -21.340457916259766,
+      "rewards/margins": 19.454242706298828,
+      "rewards/real": -1.886214017868042,
+      "step": 1380
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 6.152204836415363e-08,
+      "logits/generated": 0.0785381942987442,
+      "logits/real": -0.569561243057251,
+      "logps/generated": -367.63519287109375,
+      "logps/real": -203.30282592773438,
+      "loss": 0.0004,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -21.134796142578125,
+      "rewards/margins": 18.931886672973633,
+      "rewards/real": -2.2029080390930176,
+      "step": 1390
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 5.796586059743954e-08,
+      "logits/generated": -0.007580602075904608,
+      "logits/real": -0.5636885762214661,
+      "logps/generated": -381.52642822265625,
+      "logps/real": -198.91770935058594,
+      "loss": 0.0009,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -21.588598251342773,
+      "rewards/margins": 19.854793548583984,
+      "rewards/real": -1.7338052988052368,
+      "step": 1400
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 5.4409672830725456e-08,
+      "logits/generated": 0.0493854358792305,
+      "logits/real": -0.5976940989494324,
+      "logps/generated": -399.3151550292969,
+      "logps/real": -194.88201904296875,
+      "loss": 0.0044,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -22.92904281616211,
+      "rewards/margins": 21.451683044433594,
+      "rewards/real": -1.477359652519226,
+      "step": 1410
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 5.0853485064011376e-08,
+      "logits/generated": 0.08968095481395721,
+      "logits/real": -0.6684913039207458,
+      "logps/generated": -380.35552978515625,
+      "logps/real": -183.67074584960938,
+      "loss": 0.0016,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -21.802627563476562,
+      "rewards/margins": 20.21030616760254,
+      "rewards/real": -1.5923227071762085,
+      "step": 1420
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 4.72972972972973e-08,
+      "logits/generated": 0.08673722296953201,
+      "logits/real": -0.6326942443847656,
+      "logps/generated": -379.32373046875,
+      "logps/real": -193.1202392578125,
+      "loss": 0.0033,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -22.219440460205078,
+      "rewards/margins": 20.84255599975586,
+      "rewards/real": -1.3768887519836426,
+      "step": 1430
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 4.374110953058322e-08,
+      "logits/generated": 0.031193479895591736,
+      "logits/real": -0.5564336776733398,
+      "logps/generated": -359.17181396484375,
+      "logps/real": -193.7860107421875,
+      "loss": 0.0023,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -20.49233055114746,
+      "rewards/margins": 18.665729522705078,
+      "rewards/real": -1.8266017436981201,
+      "step": 1440
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 4.018492176386913e-08,
+      "logits/generated": 0.09109187871217728,
+      "logits/real": -0.5564282536506653,
+      "logps/generated": -374.38702392578125,
+      "logps/real": -194.70132446289062,
+      "loss": 0.0015,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -20.670825958251953,
+      "rewards/margins": 19.197032928466797,
+      "rewards/real": -1.4737932682037354,
+      "step": 1450
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 3.6628733997155046e-08,
+      "logits/generated": 0.08201099932193756,
+      "logits/real": -0.6359506845474243,
+      "logps/generated": -386.39190673828125,
+      "logps/real": -176.69708251953125,
+      "loss": 0.0068,
+      "rewards/accuracies": 0.987500011920929,
+      "rewards/generated": -22.43024444580078,
+      "rewards/margins": 20.56548309326172,
+      "rewards/real": -1.8647606372833252,
+      "step": 1460
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 3.3072546230440967e-08,
+      "logits/generated": 0.08995040506124496,
+      "logits/real": -0.5263134241104126,
+      "logps/generated": -378.1429138183594,
+      "logps/real": -191.90821838378906,
+      "loss": 0.003,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -21.608402252197266,
+      "rewards/margins": 19.824888229370117,
+      "rewards/real": -1.7835137844085693,
+      "step": 1470
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 2.9516358463726884e-08,
+      "logits/generated": 0.030845308676362038,
+      "logits/real": -0.5490936040878296,
+      "logps/generated": -369.69525146484375,
+      "logps/real": -200.53517150878906,
+      "loss": 0.0026,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -20.388917922973633,
+      "rewards/margins": 19.097196578979492,
+      "rewards/real": -1.2917201519012451,
+      "step": 1480
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 2.59601706970128e-08,
+      "logits/generated": 0.0783102884888649,
+      "logits/real": -0.5921775698661804,
+      "logps/generated": -361.94158935546875,
+      "logps/real": -206.4454803466797,
+      "loss": 0.0069,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -20.625545501708984,
+      "rewards/margins": 19.025144577026367,
+      "rewards/real": -1.6004012823104858,
+      "step": 1490
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 2.240398293029872e-08,
+      "logits/generated": 0.06931595504283905,
+      "logits/real": -0.5738247632980347,
+      "logps/generated": -371.99847412109375,
+      "logps/real": -194.58804321289062,
+      "loss": 0.0009,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -21.12423324584961,
+      "rewards/margins": 19.16390037536621,
+      "rewards/real": -1.9603347778320312,
+      "step": 1500
+    },
+    {
+      "epoch": 0.96,
+      "eval_logits/generated": 0.05270044878125191,
+      "eval_logits/real": -0.5586805939674377,
+      "eval_logps/generated": -369.19976806640625,
+      "eval_logps/real": -194.6499481201172,
+      "eval_loss": 0.0032336623407900333,
+      "eval_rewards/accuracies": 0.9992038011550903,
+      "eval_rewards/generated": -20.88124656677246,
+      "eval_rewards/margins": 19.349092483520508,
+      "eval_rewards/real": -1.5321544408798218,
+      "eval_runtime": 319.8593,
+      "eval_samples_per_second": 15.632,
+      "eval_steps_per_second": 0.491,
+      "step": 1500
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 1.8847795163584636e-08,
+      "logits/generated": 0.010430006310343742,
+      "logits/real": -0.5986403226852417,
+      "logps/generated": -369.27630615234375,
+      "logps/real": -194.7587127685547,
+      "loss": 0.0014,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -20.733909606933594,
+      "rewards/margins": 19.032808303833008,
+      "rewards/real": -1.7011024951934814,
+      "step": 1510
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 1.5291607396870554e-08,
+      "logits/generated": 0.0546521432697773,
+      "logits/real": -0.5425094366073608,
+      "logps/generated": -366.5072021484375,
+      "logps/real": -192.2476348876953,
+      "loss": 0.0017,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -21.005374908447266,
+      "rewards/margins": 19.410686492919922,
+      "rewards/real": -1.594689965248108,
+      "step": 1520
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 1.1735419630156473e-08,
+      "logits/generated": 0.07052431255578995,
+      "logits/real": -0.591436505317688,
+      "logps/generated": -371.2902526855469,
+      "logps/real": -192.4247589111328,
+      "loss": 0.0023,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -21.630722045898438,
+      "rewards/margins": 20.07946014404297,
+      "rewards/real": -1.5512605905532837,
+      "step": 1530
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 8.179231863442388e-09,
+      "logits/generated": 0.043054401874542236,
+      "logits/real": -0.547429084777832,
+      "logps/generated": -377.97113037109375,
+      "logps/real": -199.3291473388672,
+      "loss": 0.0036,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -21.497386932373047,
+      "rewards/margins": 19.968860626220703,
+      "rewards/real": -1.5285276174545288,
+      "step": 1540
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 4.623044096728307e-09,
+      "logits/generated": 0.04375555366277695,
+      "logits/real": -0.5209413766860962,
+      "logps/generated": -369.3624267578125,
+      "logps/real": -201.68338012695312,
+      "loss": 0.0019,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -21.231218338012695,
+      "rewards/margins": 19.51310157775879,
+      "rewards/real": -1.718117356300354,
+      "step": 1550
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.0668563300142248e-09,
+      "logits/generated": 0.011072209104895592,
+      "logits/real": -0.5893142223358154,
+      "logps/generated": -360.9320983886719,
+      "logps/real": -193.64450073242188,
+      "loss": 0.004,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -19.670141220092773,
+      "rewards/margins": 17.939659118652344,
+      "rewards/real": -1.7304834127426147,
+      "step": 1560
+    },
+    {
+      "epoch": 1.0,
+      "step": 1563,
+      "total_flos": 0.0,
+      "train_loss": 0.03590757104014634,
+      "train_runtime": 9394.7927,
+      "train_samples_per_second": 5.322,
+      "train_steps_per_second": 0.166
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 1563,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 1,
+  "save_steps": 100,
+  "total_flos": 0.0,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}