diff --git a/.gitattributes b/.gitattributes
index a6344aac8c09253b3b630fb776ae94478aa0275b..611ff5fafd7958070c77a047227ef1239a7bbe1b 100644
--- a/.gitattributes
+++ b/.gitattributes
@@ -33,3 +33,7 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+checkpoint-600/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+checkpoint-700/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+checkpoint-800/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+checkpoint-900/tokenizer.json filter=lfs diff=lfs merge=lfs -text
diff --git a/checkpoint-600/README.md b/checkpoint-600/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..85f3785e3148e4fdfa69be9cd1528c3af1891702
--- /dev/null
+++ b/checkpoint-600/README.md
@@ -0,0 +1,202 @@
+---
+base_model: mistralai/Mistral-Nemo-Instruct-2407
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.14.0
\ No newline at end of file
diff --git a/checkpoint-600/adapter_config.json b/checkpoint-600/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..9c49bfb22ee23ec25de631c27cc6cc7f9fc6fdc8
--- /dev/null
+++ b/checkpoint-600/adapter_config.json
@@ -0,0 +1,37 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "mistralai/Mistral-Nemo-Instruct-2407",
+  "bias": "none",
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 16,
+  "lora_bias": false,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 16,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "o_proj",
+    "q_proj",
+    "gate_proj",
+    "up_proj",
+    "down_proj",
+    "v_proj",
+    "k_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/checkpoint-600/adapter_model.safetensors b/checkpoint-600/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..45b5bde442f6f1996795d3f37955f50ec0b97ccd
--- /dev/null
+++ b/checkpoint-600/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8eb8d7e6b0eb1c1fb4d7eeec1d1244d694dabaf5e58fc6ebda4460ce419e3bf5
+size 114107976
diff --git a/checkpoint-600/global_step600/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt b/checkpoint-600/global_step600/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..fc5b46a0e47858791a0012606c60204e8ca6344b
--- /dev/null
+++ b/checkpoint-600/global_step600/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d5dee0661cd3f9972856545c2785d3b317fc18a10f64136501d7d2b69d8e9584
+size 49133072
diff --git a/checkpoint-600/global_step600/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt b/checkpoint-600/global_step600/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..7684f2a4ec5c6e0b286e58a1ef74cdd0ff37ffaa
--- /dev/null
+++ b/checkpoint-600/global_step600/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ff3078ec43802417c82c0e78c3a49427cd50d42ee6a63f3fbf74a4d179fbca4c
+size 49133072
diff --git a/checkpoint-600/global_step600/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt b/checkpoint-600/global_step600/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..d8d036b62e2c783f80253bcef8580d02e92e48ca
--- /dev/null
+++ b/checkpoint-600/global_step600/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f2dc4829363ffbfb43d423df58ce7d45775222074e5bb98566aec79bc64519e8
+size 49133072
diff --git a/checkpoint-600/global_step600/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt b/checkpoint-600/global_step600/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..c6efbb769ff4febe58da4e7876326e38d66f6d5e
--- /dev/null
+++ b/checkpoint-600/global_step600/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2c403396424a7828c6aeac58eb04af62bb14bca2d060447c1b301d0809e027fc
+size 49133072
diff --git a/checkpoint-600/global_step600/bf16_zero_pp_rank_4_mp_rank_00_optim_states.pt b/checkpoint-600/global_step600/bf16_zero_pp_rank_4_mp_rank_00_optim_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..8008b2cccfe02544a27e525077b908010dea26a6
--- /dev/null
+++ b/checkpoint-600/global_step600/bf16_zero_pp_rank_4_mp_rank_00_optim_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bea09f73f91110df3d3393e65cd854158e3cc056303289b70cfd7e340fd24c8d
+size 49133072
diff --git a/checkpoint-600/global_step600/bf16_zero_pp_rank_5_mp_rank_00_optim_states.pt b/checkpoint-600/global_step600/bf16_zero_pp_rank_5_mp_rank_00_optim_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..f67be620537b4c24bab09ef59a763c5d1c8fb50d
--- /dev/null
+++ b/checkpoint-600/global_step600/bf16_zero_pp_rank_5_mp_rank_00_optim_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b8eb7d524428ec744acb78f13d624f1604bcc0efa5a4a6370fb939e1dd397ee6
+size 49133072
diff --git a/checkpoint-600/global_step600/bf16_zero_pp_rank_6_mp_rank_00_optim_states.pt b/checkpoint-600/global_step600/bf16_zero_pp_rank_6_mp_rank_00_optim_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..77d06360f3894397c0d912fa369ce6f3a30a2a12
--- /dev/null
+++ b/checkpoint-600/global_step600/bf16_zero_pp_rank_6_mp_rank_00_optim_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9a06f004921231b7b3aed7218102d5d87fa4407b0960d9029555f2e9cb17843c
+size 49133072
diff --git a/checkpoint-600/global_step600/zero_pp_rank_0_mp_rank_00_model_states.pt b/checkpoint-600/global_step600/zero_pp_rank_0_mp_rank_00_model_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..819896cdd0d0197e295b45c56fb4723aa644294b
--- /dev/null
+++ b/checkpoint-600/global_step600/zero_pp_rank_0_mp_rank_00_model_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:03cbb159b8cccfda7676e8e19cf36f212e335b90c3a49814b8691ac21cb84dce
+size 553886
diff --git a/checkpoint-600/global_step600/zero_pp_rank_1_mp_rank_00_model_states.pt b/checkpoint-600/global_step600/zero_pp_rank_1_mp_rank_00_model_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..12ce572b21db330ebc0250f3244c14130f4d8e5b
--- /dev/null
+++ b/checkpoint-600/global_step600/zero_pp_rank_1_mp_rank_00_model_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:281036bb8a0e40d0e01e9d4a2cc15c111e4284f2e4c22cf19a78690e57bec8aa
+size 553886
diff --git a/checkpoint-600/global_step600/zero_pp_rank_2_mp_rank_00_model_states.pt b/checkpoint-600/global_step600/zero_pp_rank_2_mp_rank_00_model_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..c597300749adbe71aa1e70cbd445ff01522affd2
--- /dev/null
+++ b/checkpoint-600/global_step600/zero_pp_rank_2_mp_rank_00_model_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:566070ffca5e76ddd867c0063722365a5a8d2e3be41a423395ef2f2633cdd811
+size 553886
diff --git a/checkpoint-600/global_step600/zero_pp_rank_3_mp_rank_00_model_states.pt b/checkpoint-600/global_step600/zero_pp_rank_3_mp_rank_00_model_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..0c65299683b00bb8715f2d6934829515a2c7d21a
--- /dev/null
+++ b/checkpoint-600/global_step600/zero_pp_rank_3_mp_rank_00_model_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:639ded8d7afba064f591cd454bf89f11a982351e831a7a2e0d1cd016a1ff5bad
+size 553886
diff --git a/checkpoint-600/global_step600/zero_pp_rank_4_mp_rank_00_model_states.pt b/checkpoint-600/global_step600/zero_pp_rank_4_mp_rank_00_model_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..88aea88cd1fffa0cc1c609e9f146d1fda356b57e
--- /dev/null
+++ b/checkpoint-600/global_step600/zero_pp_rank_4_mp_rank_00_model_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ad43a6bc9c08be1440c68dd84511970f2c728f9305d442da663c8bdbaa09e4a4
+size 553886
diff --git a/checkpoint-600/global_step600/zero_pp_rank_5_mp_rank_00_model_states.pt b/checkpoint-600/global_step600/zero_pp_rank_5_mp_rank_00_model_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..3fee87f166b7cab786eb51b83a58b6f7c7498e15
--- /dev/null
+++ b/checkpoint-600/global_step600/zero_pp_rank_5_mp_rank_00_model_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a1c65be531569eb7c1f94bde7dbc5aad2203c2904761437d461e99cc409d08a2
+size 553886
diff --git a/checkpoint-600/global_step600/zero_pp_rank_6_mp_rank_00_model_states.pt b/checkpoint-600/global_step600/zero_pp_rank_6_mp_rank_00_model_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..2df271014904c5823f4c0e8d733f47e09c635feb
--- /dev/null
+++ b/checkpoint-600/global_step600/zero_pp_rank_6_mp_rank_00_model_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:25b0c162bf005bb19fac91a52ec5d554331d9d766d4c869064efae0ae7cd9ca2
+size 553886
diff --git a/checkpoint-600/latest b/checkpoint-600/latest
new file mode 100644
index 0000000000000000000000000000000000000000..12cae1adf3af8546b4141c6f62261c8e99839a54
--- /dev/null
+++ b/checkpoint-600/latest
@@ -0,0 +1 @@
+global_step600
\ No newline at end of file
diff --git a/checkpoint-600/rng_state_0.pth b/checkpoint-600/rng_state_0.pth
new file mode 100644
index 0000000000000000000000000000000000000000..82ad5d8b58573bccb5c522f193e2cfd778e6ae1b
--- /dev/null
+++ b/checkpoint-600/rng_state_0.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f9b5be461a9144680304341d499422a1226e229b465ada196b5b38902ceb9a0c
+size 15728
diff --git a/checkpoint-600/rng_state_1.pth b/checkpoint-600/rng_state_1.pth
new file mode 100644
index 0000000000000000000000000000000000000000..6b7fb7f659ec67a3bcf7e96536211b9c0e0f8f7e
--- /dev/null
+++ b/checkpoint-600/rng_state_1.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0feae1711de30f87a321998c529d6c2b1e1ed143b730c16ff0981a623a78513b
+size 15728
diff --git a/checkpoint-600/rng_state_2.pth b/checkpoint-600/rng_state_2.pth
new file mode 100644
index 0000000000000000000000000000000000000000..d22a8d8780c430d170077a08c15be231da26e1ec
--- /dev/null
+++ b/checkpoint-600/rng_state_2.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5634c41991485c64b2ce5e96d44963d5ed2d22b62a5e7225488d04f6f23526aa
+size 15728
diff --git a/checkpoint-600/rng_state_3.pth b/checkpoint-600/rng_state_3.pth
new file mode 100644
index 0000000000000000000000000000000000000000..7d2b6899f6a4c756dd9ff5d45f12d7e535706cfa
--- /dev/null
+++ b/checkpoint-600/rng_state_3.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b6988a49ed9faa2773412d8f268d7762c2316c4cb23dfc224a9df180a0143c2b
+size 15792
diff --git a/checkpoint-600/rng_state_4.pth b/checkpoint-600/rng_state_4.pth
new file mode 100644
index 0000000000000000000000000000000000000000..ae0ba3ccb1c5f7f8a17aa63d2c40d17defba9c55
--- /dev/null
+++ b/checkpoint-600/rng_state_4.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:adc2903bd0c8e5f988404aad30c7d16d49f978baddf2ebad5ff1679ed097c0c6
+size 15728
diff --git a/checkpoint-600/rng_state_5.pth b/checkpoint-600/rng_state_5.pth
new file mode 100644
index 0000000000000000000000000000000000000000..127e87f5e7261d1dda2f894b43e550fee029b635
--- /dev/null
+++ b/checkpoint-600/rng_state_5.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7bc2763f2f3a1a84b26315e60114ef857852f477e9a289d0eb70a5d1e8c03677
+size 15728
diff --git a/checkpoint-600/rng_state_6.pth b/checkpoint-600/rng_state_6.pth
new file mode 100644
index 0000000000000000000000000000000000000000..bf661e7aea31e524fd4108f4886f046a4cf61ff4
--- /dev/null
+++ b/checkpoint-600/rng_state_6.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1f5bc32e51cda2cf470e0b0816158211f03211a8ea5aa2549e1a0a6271fbc6f8
+size 15728
diff --git a/checkpoint-600/scheduler.pt b/checkpoint-600/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..adaeb5baf669fee62cee45e9cbe4d55b73f132a0
--- /dev/null
+++ b/checkpoint-600/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:842bed8b2136361e75543fcfdbbf15de4400f11948b6981f3e7627aadcb92115
+size 1064
diff --git a/checkpoint-600/special_tokens_map.json b/checkpoint-600/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..72ecfeeb7e14d244c936169d2ed139eeae235ef1
--- /dev/null
+++ b/checkpoint-600/special_tokens_map.json
@@ -0,0 +1,24 @@
+{
+  "bos_token": {
+    "content": "<s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "</s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "</s>",
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}
diff --git a/checkpoint-600/tokenizer.json b/checkpoint-600/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..b35b1c0f0bd5007afc7ad5976eeaf44988bd2067
--- /dev/null
+++ b/checkpoint-600/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9a1c103d223ee5dc2dde8307635f7d12581b40855198a38efcfeb6db4e08da69
+size 17078445
diff --git a/checkpoint-600/tokenizer_config.json b/checkpoint-600/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..550f3f7fc9144301cd0f17eeb0ad8746a5fa5354
--- /dev/null
+++ b/checkpoint-600/tokenizer_config.json
@@ -0,0 +1,8020 @@
+{
+  "add_bos_token": true,
+  "add_eos_token": false,
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "[INST]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "4": {
+      "content": "[/INST]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "5": {
+      "content": "[AVAILABLE_TOOLS]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "6": {
+      "content": "[/AVAILABLE_TOOLS]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "7": {
+      "content": "[TOOL_RESULTS]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "8": {
+      "content": "[/TOOL_RESULTS]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "9": {
+      "content": "[TOOL_CALLS]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "10": {
+      "content": "<pad>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "11": {
+      "content": "[PREFIX]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "12": {
+      "content": "[MIDDLE]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "13": {
+      "content": "[SUFFIX]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "14": {
+      "content": "<SPECIAL_14>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "15": {
+      "content": "<SPECIAL_15>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "16": {
+      "content": "<SPECIAL_16>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "17": {
+      "content": "<SPECIAL_17>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "18": {
+      "content": "<SPECIAL_18>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "19": {
+      "content": "<SPECIAL_19>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "20": {
+      "content": "<SPECIAL_20>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "21": {
+      "content": "<SPECIAL_21>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "22": {
+      "content": "<SPECIAL_22>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "23": {
+      "content": "<SPECIAL_23>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "24": {
+      "content": "<SPECIAL_24>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "25": {
+      "content": "<SPECIAL_25>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "26": {
+      "content": "<SPECIAL_26>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "27": {
+      "content": "<SPECIAL_27>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "28": {
+      "content": "<SPECIAL_28>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "29": {
+      "content": "<SPECIAL_29>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "30": {
+      "content": "<SPECIAL_30>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "31": {
+      "content": "<SPECIAL_31>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32": {
+      "content": "<SPECIAL_32>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "33": {
+      "content": "<SPECIAL_33>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "34": {
+      "content": "<SPECIAL_34>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "35": {
+      "content": "<SPECIAL_35>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "36": {
+      "content": "<SPECIAL_36>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "37": {
+      "content": "<SPECIAL_37>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "38": {
+      "content": "<SPECIAL_38>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "39": {
+      "content": "<SPECIAL_39>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "40": {
+      "content": "<SPECIAL_40>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "41": {
+      "content": "<SPECIAL_41>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "42": {
+      "content": "<SPECIAL_42>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "43": {
+      "content": "<SPECIAL_43>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "44": {
+      "content": "<SPECIAL_44>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "45": {
+      "content": "<SPECIAL_45>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "46": {
+      "content": "<SPECIAL_46>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "47": {
+      "content": "<SPECIAL_47>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "48": {
+      "content": "<SPECIAL_48>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "49": {
+      "content": "<SPECIAL_49>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "50": {
+      "content": "<SPECIAL_50>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "51": {
+      "content": "<SPECIAL_51>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "52": {
+      "content": "<SPECIAL_52>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "53": {
+      "content": "<SPECIAL_53>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "54": {
+      "content": "<SPECIAL_54>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "55": {
+      "content": "<SPECIAL_55>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "56": {
+      "content": "<SPECIAL_56>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "57": {
+      "content": "<SPECIAL_57>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "58": {
+      "content": "<SPECIAL_58>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "59": {
+      "content": "<SPECIAL_59>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "60": {
+      "content": "<SPECIAL_60>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "61": {
+      "content": "<SPECIAL_61>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "62": {
+      "content": "<SPECIAL_62>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "63": {
+      "content": "<SPECIAL_63>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "64": {
+      "content": "<SPECIAL_64>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "65": {
+      "content": "<SPECIAL_65>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "66": {
+      "content": "<SPECIAL_66>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "67": {
+      "content": "<SPECIAL_67>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "68": {
+      "content": "<SPECIAL_68>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "69": {
+      "content": "<SPECIAL_69>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "70": {
+      "content": "<SPECIAL_70>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "71": {
+      "content": "<SPECIAL_71>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "72": {
+      "content": "<SPECIAL_72>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "73": {
+      "content": "<SPECIAL_73>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "74": {
+      "content": "<SPECIAL_74>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "75": {
+      "content": "<SPECIAL_75>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "76": {
+      "content": "<SPECIAL_76>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "77": {
+      "content": "<SPECIAL_77>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "78": {
+      "content": "<SPECIAL_78>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "79": {
+      "content": "<SPECIAL_79>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "80": {
+      "content": "<SPECIAL_80>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "81": {
+      "content": "<SPECIAL_81>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "82": {
+      "content": "<SPECIAL_82>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "83": {
+      "content": "<SPECIAL_83>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "84": {
+      "content": "<SPECIAL_84>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "85": {
+      "content": "<SPECIAL_85>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "86": {
+      "content": "<SPECIAL_86>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "87": {
+      "content": "<SPECIAL_87>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "88": {
+      "content": "<SPECIAL_88>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "89": {
+      "content": "<SPECIAL_89>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "90": {
+      "content": "<SPECIAL_90>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "91": {
+      "content": "<SPECIAL_91>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "92": {
+      "content": "<SPECIAL_92>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "93": {
+      "content": "<SPECIAL_93>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "94": {
+      "content": "<SPECIAL_94>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "95": {
+      "content": "<SPECIAL_95>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "96": {
+      "content": "<SPECIAL_96>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "97": {
+      "content": "<SPECIAL_97>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "98": {
+      "content": "<SPECIAL_98>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "99": {
+      "content": "<SPECIAL_99>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "100": {
+      "content": "<SPECIAL_100>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "101": {
+      "content": "<SPECIAL_101>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "102": {
+      "content": "<SPECIAL_102>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "103": {
+      "content": "<SPECIAL_103>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "104": {
+      "content": "<SPECIAL_104>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "105": {
+      "content": "<SPECIAL_105>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "106": {
+      "content": "<SPECIAL_106>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "107": {
+      "content": "<SPECIAL_107>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "108": {
+      "content": "<SPECIAL_108>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "109": {
+      "content": "<SPECIAL_109>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "110": {
+      "content": "<SPECIAL_110>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "111": {
+      "content": "<SPECIAL_111>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "112": {
+      "content": "<SPECIAL_112>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "113": {
+      "content": "<SPECIAL_113>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "114": {
+      "content": "<SPECIAL_114>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "115": {
+      "content": "<SPECIAL_115>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "116": {
+      "content": "<SPECIAL_116>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "117": {
+      "content": "<SPECIAL_117>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "118": {
+      "content": "<SPECIAL_118>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "119": {
+      "content": "<SPECIAL_119>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "120": {
+      "content": "<SPECIAL_120>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "121": {
+      "content": "<SPECIAL_121>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "122": {
+      "content": "<SPECIAL_122>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "123": {
+      "content": "<SPECIAL_123>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "124": {
+      "content": "<SPECIAL_124>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "125": {
+      "content": "<SPECIAL_125>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "126": {
+      "content": "<SPECIAL_126>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "127": {
+      "content": "<SPECIAL_127>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128": {
+      "content": "<SPECIAL_128>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "129": {
+      "content": "<SPECIAL_129>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "130": {
+      "content": "<SPECIAL_130>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "131": {
+      "content": "<SPECIAL_131>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "132": {
+      "content": "<SPECIAL_132>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "133": {
+      "content": "<SPECIAL_133>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "134": {
+      "content": "<SPECIAL_134>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "135": {
+      "content": "<SPECIAL_135>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "136": {
+      "content": "<SPECIAL_136>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "137": {
+      "content": "<SPECIAL_137>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "138": {
+      "content": "<SPECIAL_138>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "139": {
+      "content": "<SPECIAL_139>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "140": {
+      "content": "<SPECIAL_140>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "141": {
+      "content": "<SPECIAL_141>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "142": {
+      "content": "<SPECIAL_142>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "143": {
+      "content": "<SPECIAL_143>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "144": {
+      "content": "<SPECIAL_144>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "145": {
+      "content": "<SPECIAL_145>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "146": {
+      "content": "<SPECIAL_146>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "147": {
+      "content": "<SPECIAL_147>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "148": {
+      "content": "<SPECIAL_148>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "149": {
+      "content": "<SPECIAL_149>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "150": {
+      "content": "<SPECIAL_150>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151": {
+      "content": "<SPECIAL_151>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "152": {
+      "content": "<SPECIAL_152>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "153": {
+      "content": "<SPECIAL_153>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "154": {
+      "content": "<SPECIAL_154>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "155": {
+      "content": "<SPECIAL_155>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "156": {
+      "content": "<SPECIAL_156>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "157": {
+      "content": "<SPECIAL_157>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "158": {
+      "content": "<SPECIAL_158>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "159": {
+      "content": "<SPECIAL_159>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "160": {
+      "content": "<SPECIAL_160>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "161": {
+      "content": "<SPECIAL_161>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "162": {
+      "content": "<SPECIAL_162>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "163": {
+      "content": "<SPECIAL_163>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "164": {
+      "content": "<SPECIAL_164>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "165": {
+      "content": "<SPECIAL_165>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "166": {
+      "content": "<SPECIAL_166>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "167": {
+      "content": "<SPECIAL_167>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "168": {
+      "content": "<SPECIAL_168>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "169": {
+      "content": "<SPECIAL_169>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170": {
+      "content": "<SPECIAL_170>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "171": {
+      "content": "<SPECIAL_171>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "172": {
+      "content": "<SPECIAL_172>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "173": {
+      "content": "<SPECIAL_173>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "174": {
+      "content": "<SPECIAL_174>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "175": {
+      "content": "<SPECIAL_175>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "176": {
+      "content": "<SPECIAL_176>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "177": {
+      "content": "<SPECIAL_177>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "178": {
+      "content": "<SPECIAL_178>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "179": {
+      "content": "<SPECIAL_179>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "180": {
+      "content": "<SPECIAL_180>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "181": {
+      "content": "<SPECIAL_181>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "182": {
+      "content": "<SPECIAL_182>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "183": {
+      "content": "<SPECIAL_183>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "184": {
+      "content": "<SPECIAL_184>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "185": {
+      "content": "<SPECIAL_185>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "186": {
+      "content": "<SPECIAL_186>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "187": {
+      "content": "<SPECIAL_187>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "188": {
+      "content": "<SPECIAL_188>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "189": {
+      "content": "<SPECIAL_189>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "190": {
+      "content": "<SPECIAL_190>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "191": {
+      "content": "<SPECIAL_191>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "192": {
+      "content": "<SPECIAL_192>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "193": {
+      "content": "<SPECIAL_193>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "194": {
+      "content": "<SPECIAL_194>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "195": {
+      "content": "<SPECIAL_195>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "196": {
+      "content": "<SPECIAL_196>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "197": {
+      "content": "<SPECIAL_197>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "198": {
+      "content": "<SPECIAL_198>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "199": {
+      "content": "<SPECIAL_199>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "200": {
+      "content": "<SPECIAL_200>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "201": {
+      "content": "<SPECIAL_201>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "202": {
+      "content": "<SPECIAL_202>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "203": {
+      "content": "<SPECIAL_203>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "204": {
+      "content": "<SPECIAL_204>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "205": {
+      "content": "<SPECIAL_205>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "206": {
+      "content": "<SPECIAL_206>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "207": {
+      "content": "<SPECIAL_207>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "208": {
+      "content": "<SPECIAL_208>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "209": {
+      "content": "<SPECIAL_209>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "210": {
+      "content": "<SPECIAL_210>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "211": {
+      "content": "<SPECIAL_211>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "212": {
+      "content": "<SPECIAL_212>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "213": {
+      "content": "<SPECIAL_213>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "214": {
+      "content": "<SPECIAL_214>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "215": {
+      "content": "<SPECIAL_215>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "216": {
+      "content": "<SPECIAL_216>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "217": {
+      "content": "<SPECIAL_217>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "218": {
+      "content": "<SPECIAL_218>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "219": {
+      "content": "<SPECIAL_219>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "220": {
+      "content": "<SPECIAL_220>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "221": {
+      "content": "<SPECIAL_221>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "222": {
+      "content": "<SPECIAL_222>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "223": {
+      "content": "<SPECIAL_223>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "224": {
+      "content": "<SPECIAL_224>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "225": {
+      "content": "<SPECIAL_225>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "226": {
+      "content": "<SPECIAL_226>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "227": {
+      "content": "<SPECIAL_227>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "228": {
+      "content": "<SPECIAL_228>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "229": {
+      "content": "<SPECIAL_229>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "230": {
+      "content": "<SPECIAL_230>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "231": {
+      "content": "<SPECIAL_231>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "232": {
+      "content": "<SPECIAL_232>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "233": {
+      "content": "<SPECIAL_233>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "234": {
+      "content": "<SPECIAL_234>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "235": {
+      "content": "<SPECIAL_235>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "236": {
+      "content": "<SPECIAL_236>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "237": {
+      "content": "<SPECIAL_237>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "238": {
+      "content": "<SPECIAL_238>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "239": {
+      "content": "<SPECIAL_239>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "240": {
+      "content": "<SPECIAL_240>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "241": {
+      "content": "<SPECIAL_241>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "242": {
+      "content": "<SPECIAL_242>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "243": {
+      "content": "<SPECIAL_243>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "244": {
+      "content": "<SPECIAL_244>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "245": {
+      "content": "<SPECIAL_245>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "246": {
+      "content": "<SPECIAL_246>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "247": {
+      "content": "<SPECIAL_247>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "248": {
+      "content": "<SPECIAL_248>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "249": {
+      "content": "<SPECIAL_249>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250": {
+      "content": "<SPECIAL_250>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "251": {
+      "content": "<SPECIAL_251>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "252": {
+      "content": "<SPECIAL_252>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "253": {
+      "content": "<SPECIAL_253>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "254": {
+      "content": "<SPECIAL_254>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "255": {
+      "content": "<SPECIAL_255>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "256": {
+      "content": "<SPECIAL_256>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "257": {
+      "content": "<SPECIAL_257>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "258": {
+      "content": "<SPECIAL_258>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "259": {
+      "content": "<SPECIAL_259>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "260": {
+      "content": "<SPECIAL_260>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "261": {
+      "content": "<SPECIAL_261>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "262": {
+      "content": "<SPECIAL_262>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "263": {
+      "content": "<SPECIAL_263>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "264": {
+      "content": "<SPECIAL_264>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "265": {
+      "content": "<SPECIAL_265>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "266": {
+      "content": "<SPECIAL_266>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "267": {
+      "content": "<SPECIAL_267>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "268": {
+      "content": "<SPECIAL_268>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "269": {
+      "content": "<SPECIAL_269>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "270": {
+      "content": "<SPECIAL_270>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "271": {
+      "content": "<SPECIAL_271>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "272": {
+      "content": "<SPECIAL_272>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "273": {
+      "content": "<SPECIAL_273>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "274": {
+      "content": "<SPECIAL_274>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "275": {
+      "content": "<SPECIAL_275>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "276": {
+      "content": "<SPECIAL_276>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "277": {
+      "content": "<SPECIAL_277>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "278": {
+      "content": "<SPECIAL_278>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "279": {
+      "content": "<SPECIAL_279>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "280": {
+      "content": "<SPECIAL_280>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "281": {
+      "content": "<SPECIAL_281>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "282": {
+      "content": "<SPECIAL_282>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "283": {
+      "content": "<SPECIAL_283>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "284": {
+      "content": "<SPECIAL_284>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "285": {
+      "content": "<SPECIAL_285>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "286": {
+      "content": "<SPECIAL_286>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "287": {
+      "content": "<SPECIAL_287>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "288": {
+      "content": "<SPECIAL_288>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "289": {
+      "content": "<SPECIAL_289>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "290": {
+      "content": "<SPECIAL_290>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "291": {
+      "content": "<SPECIAL_291>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "292": {
+      "content": "<SPECIAL_292>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "293": {
+      "content": "<SPECIAL_293>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "294": {
+      "content": "<SPECIAL_294>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "295": {
+      "content": "<SPECIAL_295>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "296": {
+      "content": "<SPECIAL_296>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "297": {
+      "content": "<SPECIAL_297>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "298": {
+      "content": "<SPECIAL_298>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "299": {
+      "content": "<SPECIAL_299>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "300": {
+      "content": "<SPECIAL_300>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "301": {
+      "content": "<SPECIAL_301>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "302": {
+      "content": "<SPECIAL_302>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "303": {
+      "content": "<SPECIAL_303>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "304": {
+      "content": "<SPECIAL_304>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "305": {
+      "content": "<SPECIAL_305>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "306": {
+      "content": "<SPECIAL_306>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "307": {
+      "content": "<SPECIAL_307>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "308": {
+      "content": "<SPECIAL_308>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "309": {
+      "content": "<SPECIAL_309>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "310": {
+      "content": "<SPECIAL_310>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "311": {
+      "content": "<SPECIAL_311>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "312": {
+      "content": "<SPECIAL_312>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "313": {
+      "content": "<SPECIAL_313>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "314": {
+      "content": "<SPECIAL_314>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "315": {
+      "content": "<SPECIAL_315>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "316": {
+      "content": "<SPECIAL_316>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "317": {
+      "content": "<SPECIAL_317>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "318": {
+      "content": "<SPECIAL_318>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "319": {
+      "content": "<SPECIAL_319>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "320": {
+      "content": "<SPECIAL_320>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "321": {
+      "content": "<SPECIAL_321>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "322": {
+      "content": "<SPECIAL_322>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "323": {
+      "content": "<SPECIAL_323>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "324": {
+      "content": "<SPECIAL_324>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "325": {
+      "content": "<SPECIAL_325>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "326": {
+      "content": "<SPECIAL_326>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "327": {
+      "content": "<SPECIAL_327>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "328": {
+      "content": "<SPECIAL_328>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "329": {
+      "content": "<SPECIAL_329>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "330": {
+      "content": "<SPECIAL_330>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "331": {
+      "content": "<SPECIAL_331>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "332": {
+      "content": "<SPECIAL_332>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "333": {
+      "content": "<SPECIAL_333>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "334": {
+      "content": "<SPECIAL_334>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "335": {
+      "content": "<SPECIAL_335>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "336": {
+      "content": "<SPECIAL_336>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "337": {
+      "content": "<SPECIAL_337>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "338": {
+      "content": "<SPECIAL_338>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "339": {
+      "content": "<SPECIAL_339>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "340": {
+      "content": "<SPECIAL_340>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "341": {
+      "content": "<SPECIAL_341>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "342": {
+      "content": "<SPECIAL_342>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "343": {
+      "content": "<SPECIAL_343>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "344": {
+      "content": "<SPECIAL_344>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "345": {
+      "content": "<SPECIAL_345>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "346": {
+      "content": "<SPECIAL_346>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "347": {
+      "content": "<SPECIAL_347>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "348": {
+      "content": "<SPECIAL_348>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "349": {
+      "content": "<SPECIAL_349>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "350": {
+      "content": "<SPECIAL_350>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "351": {
+      "content": "<SPECIAL_351>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "352": {
+      "content": "<SPECIAL_352>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "353": {
+      "content": "<SPECIAL_353>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "354": {
+      "content": "<SPECIAL_354>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "355": {
+      "content": "<SPECIAL_355>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "356": {
+      "content": "<SPECIAL_356>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "357": {
+      "content": "<SPECIAL_357>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "358": {
+      "content": "<SPECIAL_358>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "359": {
+      "content": "<SPECIAL_359>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "360": {
+      "content": "<SPECIAL_360>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "361": {
+      "content": "<SPECIAL_361>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "362": {
+      "content": "<SPECIAL_362>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "363": {
+      "content": "<SPECIAL_363>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "364": {
+      "content": "<SPECIAL_364>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "365": {
+      "content": "<SPECIAL_365>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "366": {
+      "content": "<SPECIAL_366>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "367": {
+      "content": "<SPECIAL_367>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "368": {
+      "content": "<SPECIAL_368>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "369": {
+      "content": "<SPECIAL_369>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "370": {
+      "content": "<SPECIAL_370>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "371": {
+      "content": "<SPECIAL_371>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "372": {
+      "content": "<SPECIAL_372>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "373": {
+      "content": "<SPECIAL_373>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "374": {
+      "content": "<SPECIAL_374>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "375": {
+      "content": "<SPECIAL_375>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "376": {
+      "content": "<SPECIAL_376>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "377": {
+      "content": "<SPECIAL_377>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "378": {
+      "content": "<SPECIAL_378>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "379": {
+      "content": "<SPECIAL_379>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "380": {
+      "content": "<SPECIAL_380>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "381": {
+      "content": "<SPECIAL_381>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "382": {
+      "content": "<SPECIAL_382>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "383": {
+      "content": "<SPECIAL_383>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "384": {
+      "content": "<SPECIAL_384>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "385": {
+      "content": "<SPECIAL_385>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "386": {
+      "content": "<SPECIAL_386>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "387": {
+      "content": "<SPECIAL_387>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "388": {
+      "content": "<SPECIAL_388>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "389": {
+      "content": "<SPECIAL_389>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "390": {
+      "content": "<SPECIAL_390>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "391": {
+      "content": "<SPECIAL_391>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "392": {
+      "content": "<SPECIAL_392>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "393": {
+      "content": "<SPECIAL_393>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "394": {
+      "content": "<SPECIAL_394>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "395": {
+      "content": "<SPECIAL_395>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "396": {
+      "content": "<SPECIAL_396>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "397": {
+      "content": "<SPECIAL_397>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "398": {
+      "content": "<SPECIAL_398>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "399": {
+      "content": "<SPECIAL_399>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "400": {
+      "content": "<SPECIAL_400>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "401": {
+      "content": "<SPECIAL_401>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "402": {
+      "content": "<SPECIAL_402>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "403": {
+      "content": "<SPECIAL_403>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "404": {
+      "content": "<SPECIAL_404>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "405": {
+      "content": "<SPECIAL_405>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "406": {
+      "content": "<SPECIAL_406>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "407": {
+      "content": "<SPECIAL_407>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "408": {
+      "content": "<SPECIAL_408>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "409": {
+      "content": "<SPECIAL_409>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "410": {
+      "content": "<SPECIAL_410>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "411": {
+      "content": "<SPECIAL_411>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "412": {
+      "content": "<SPECIAL_412>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "413": {
+      "content": "<SPECIAL_413>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "414": {
+      "content": "<SPECIAL_414>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "415": {
+      "content": "<SPECIAL_415>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "416": {
+      "content": "<SPECIAL_416>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "417": {
+      "content": "<SPECIAL_417>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "418": {
+      "content": "<SPECIAL_418>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "419": {
+      "content": "<SPECIAL_419>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "420": {
+      "content": "<SPECIAL_420>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "421": {
+      "content": "<SPECIAL_421>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "422": {
+      "content": "<SPECIAL_422>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "423": {
+      "content": "<SPECIAL_423>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "424": {
+      "content": "<SPECIAL_424>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "425": {
+      "content": "<SPECIAL_425>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "426": {
+      "content": "<SPECIAL_426>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "427": {
+      "content": "<SPECIAL_427>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "428": {
+      "content": "<SPECIAL_428>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "429": {
+      "content": "<SPECIAL_429>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "430": {
+      "content": "<SPECIAL_430>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "431": {
+      "content": "<SPECIAL_431>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "432": {
+      "content": "<SPECIAL_432>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "433": {
+      "content": "<SPECIAL_433>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "434": {
+      "content": "<SPECIAL_434>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "435": {
+      "content": "<SPECIAL_435>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "436": {
+      "content": "<SPECIAL_436>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "437": {
+      "content": "<SPECIAL_437>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "438": {
+      "content": "<SPECIAL_438>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "439": {
+      "content": "<SPECIAL_439>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "440": {
+      "content": "<SPECIAL_440>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "441": {
+      "content": "<SPECIAL_441>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "442": {
+      "content": "<SPECIAL_442>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "443": {
+      "content": "<SPECIAL_443>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "444": {
+      "content": "<SPECIAL_444>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "445": {
+      "content": "<SPECIAL_445>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "446": {
+      "content": "<SPECIAL_446>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "447": {
+      "content": "<SPECIAL_447>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "448": {
+      "content": "<SPECIAL_448>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "449": {
+      "content": "<SPECIAL_449>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "450": {
+      "content": "<SPECIAL_450>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "451": {
+      "content": "<SPECIAL_451>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "452": {
+      "content": "<SPECIAL_452>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "453": {
+      "content": "<SPECIAL_453>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "454": {
+      "content": "<SPECIAL_454>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "455": {
+      "content": "<SPECIAL_455>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "456": {
+      "content": "<SPECIAL_456>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "457": {
+      "content": "<SPECIAL_457>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "458": {
+      "content": "<SPECIAL_458>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "459": {
+      "content": "<SPECIAL_459>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "460": {
+      "content": "<SPECIAL_460>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "461": {
+      "content": "<SPECIAL_461>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "462": {
+      "content": "<SPECIAL_462>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "463": {
+      "content": "<SPECIAL_463>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "464": {
+      "content": "<SPECIAL_464>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "465": {
+      "content": "<SPECIAL_465>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "466": {
+      "content": "<SPECIAL_466>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "467": {
+      "content": "<SPECIAL_467>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "468": {
+      "content": "<SPECIAL_468>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "469": {
+      "content": "<SPECIAL_469>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "470": {
+      "content": "<SPECIAL_470>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "471": {
+      "content": "<SPECIAL_471>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "472": {
+      "content": "<SPECIAL_472>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "473": {
+      "content": "<SPECIAL_473>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "474": {
+      "content": "<SPECIAL_474>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "475": {
+      "content": "<SPECIAL_475>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "476": {
+      "content": "<SPECIAL_476>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "477": {
+      "content": "<SPECIAL_477>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "478": {
+      "content": "<SPECIAL_478>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "479": {
+      "content": "<SPECIAL_479>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "480": {
+      "content": "<SPECIAL_480>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "481": {
+      "content": "<SPECIAL_481>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "482": {
+      "content": "<SPECIAL_482>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "483": {
+      "content": "<SPECIAL_483>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "484": {
+      "content": "<SPECIAL_484>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "485": {
+      "content": "<SPECIAL_485>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "486": {
+      "content": "<SPECIAL_486>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "487": {
+      "content": "<SPECIAL_487>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "488": {
+      "content": "<SPECIAL_488>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "489": {
+      "content": "<SPECIAL_489>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "490": {
+      "content": "<SPECIAL_490>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "491": {
+      "content": "<SPECIAL_491>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "492": {
+      "content": "<SPECIAL_492>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "493": {
+      "content": "<SPECIAL_493>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "494": {
+      "content": "<SPECIAL_494>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "495": {
+      "content": "<SPECIAL_495>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "496": {
+      "content": "<SPECIAL_496>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "497": {
+      "content": "<SPECIAL_497>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "498": {
+      "content": "<SPECIAL_498>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "499": {
+      "content": "<SPECIAL_499>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "500": {
+      "content": "<SPECIAL_500>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "501": {
+      "content": "<SPECIAL_501>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "502": {
+      "content": "<SPECIAL_502>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "503": {
+      "content": "<SPECIAL_503>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "504": {
+      "content": "<SPECIAL_504>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "505": {
+      "content": "<SPECIAL_505>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "506": {
+      "content": "<SPECIAL_506>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "507": {
+      "content": "<SPECIAL_507>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "508": {
+      "content": "<SPECIAL_508>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "509": {
+      "content": "<SPECIAL_509>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "510": {
+      "content": "<SPECIAL_510>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "511": {
+      "content": "<SPECIAL_511>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "512": {
+      "content": "<SPECIAL_512>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "513": {
+      "content": "<SPECIAL_513>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "514": {
+      "content": "<SPECIAL_514>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "515": {
+      "content": "<SPECIAL_515>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "516": {
+      "content": "<SPECIAL_516>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "517": {
+      "content": "<SPECIAL_517>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "518": {
+      "content": "<SPECIAL_518>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "519": {
+      "content": "<SPECIAL_519>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "520": {
+      "content": "<SPECIAL_520>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "521": {
+      "content": "<SPECIAL_521>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "522": {
+      "content": "<SPECIAL_522>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "523": {
+      "content": "<SPECIAL_523>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "524": {
+      "content": "<SPECIAL_524>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "525": {
+      "content": "<SPECIAL_525>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "526": {
+      "content": "<SPECIAL_526>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "527": {
+      "content": "<SPECIAL_527>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "528": {
+      "content": "<SPECIAL_528>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "529": {
+      "content": "<SPECIAL_529>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "530": {
+      "content": "<SPECIAL_530>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "531": {
+      "content": "<SPECIAL_531>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "532": {
+      "content": "<SPECIAL_532>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "533": {
+      "content": "<SPECIAL_533>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "534": {
+      "content": "<SPECIAL_534>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "535": {
+      "content": "<SPECIAL_535>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "536": {
+      "content": "<SPECIAL_536>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "537": {
+      "content": "<SPECIAL_537>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "538": {
+      "content": "<SPECIAL_538>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "539": {
+      "content": "<SPECIAL_539>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "540": {
+      "content": "<SPECIAL_540>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "541": {
+      "content": "<SPECIAL_541>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "542": {
+      "content": "<SPECIAL_542>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "543": {
+      "content": "<SPECIAL_543>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "544": {
+      "content": "<SPECIAL_544>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "545": {
+      "content": "<SPECIAL_545>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "546": {
+      "content": "<SPECIAL_546>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "547": {
+      "content": "<SPECIAL_547>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "548": {
+      "content": "<SPECIAL_548>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "549": {
+      "content": "<SPECIAL_549>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "550": {
+      "content": "<SPECIAL_550>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "551": {
+      "content": "<SPECIAL_551>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "552": {
+      "content": "<SPECIAL_552>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "553": {
+      "content": "<SPECIAL_553>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "554": {
+      "content": "<SPECIAL_554>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "555": {
+      "content": "<SPECIAL_555>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "556": {
+      "content": "<SPECIAL_556>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "557": {
+      "content": "<SPECIAL_557>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "558": {
+      "content": "<SPECIAL_558>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "559": {
+      "content": "<SPECIAL_559>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "560": {
+      "content": "<SPECIAL_560>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "561": {
+      "content": "<SPECIAL_561>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "562": {
+      "content": "<SPECIAL_562>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "563": {
+      "content": "<SPECIAL_563>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "564": {
+      "content": "<SPECIAL_564>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "565": {
+      "content": "<SPECIAL_565>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "566": {
+      "content": "<SPECIAL_566>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "567": {
+      "content": "<SPECIAL_567>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "568": {
+      "content": "<SPECIAL_568>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "569": {
+      "content": "<SPECIAL_569>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "570": {
+      "content": "<SPECIAL_570>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "571": {
+      "content": "<SPECIAL_571>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "572": {
+      "content": "<SPECIAL_572>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "573": {
+      "content": "<SPECIAL_573>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "574": {
+      "content": "<SPECIAL_574>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "575": {
+      "content": "<SPECIAL_575>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "576": {
+      "content": "<SPECIAL_576>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "577": {
+      "content": "<SPECIAL_577>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "578": {
+      "content": "<SPECIAL_578>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "579": {
+      "content": "<SPECIAL_579>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "580": {
+      "content": "<SPECIAL_580>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "581": {
+      "content": "<SPECIAL_581>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "582": {
+      "content": "<SPECIAL_582>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "583": {
+      "content": "<SPECIAL_583>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "584": {
+      "content": "<SPECIAL_584>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "585": {
+      "content": "<SPECIAL_585>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "586": {
+      "content": "<SPECIAL_586>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "587": {
+      "content": "<SPECIAL_587>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "588": {
+      "content": "<SPECIAL_588>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "589": {
+      "content": "<SPECIAL_589>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "590": {
+      "content": "<SPECIAL_590>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "591": {
+      "content": "<SPECIAL_591>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "592": {
+      "content": "<SPECIAL_592>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "593": {
+      "content": "<SPECIAL_593>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "594": {
+      "content": "<SPECIAL_594>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "595": {
+      "content": "<SPECIAL_595>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "596": {
+      "content": "<SPECIAL_596>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "597": {
+      "content": "<SPECIAL_597>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "598": {
+      "content": "<SPECIAL_598>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "599": {
+      "content": "<SPECIAL_599>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "600": {
+      "content": "<SPECIAL_600>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "601": {
+      "content": "<SPECIAL_601>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "602": {
+      "content": "<SPECIAL_602>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "603": {
+      "content": "<SPECIAL_603>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "604": {
+      "content": "<SPECIAL_604>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "605": {
+      "content": "<SPECIAL_605>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "606": {
+      "content": "<SPECIAL_606>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "607": {
+      "content": "<SPECIAL_607>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "608": {
+      "content": "<SPECIAL_608>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "609": {
+      "content": "<SPECIAL_609>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "610": {
+      "content": "<SPECIAL_610>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "611": {
+      "content": "<SPECIAL_611>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "612": {
+      "content": "<SPECIAL_612>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "613": {
+      "content": "<SPECIAL_613>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "614": {
+      "content": "<SPECIAL_614>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "615": {
+      "content": "<SPECIAL_615>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "616": {
+      "content": "<SPECIAL_616>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "617": {
+      "content": "<SPECIAL_617>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "618": {
+      "content": "<SPECIAL_618>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "619": {
+      "content": "<SPECIAL_619>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "620": {
+      "content": "<SPECIAL_620>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "621": {
+      "content": "<SPECIAL_621>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "622": {
+      "content": "<SPECIAL_622>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "623": {
+      "content": "<SPECIAL_623>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "624": {
+      "content": "<SPECIAL_624>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "625": {
+      "content": "<SPECIAL_625>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "626": {
+      "content": "<SPECIAL_626>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "627": {
+      "content": "<SPECIAL_627>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "628": {
+      "content": "<SPECIAL_628>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "629": {
+      "content": "<SPECIAL_629>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "630": {
+      "content": "<SPECIAL_630>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "631": {
+      "content": "<SPECIAL_631>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "632": {
+      "content": "<SPECIAL_632>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "633": {
+      "content": "<SPECIAL_633>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "634": {
+      "content": "<SPECIAL_634>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "635": {
+      "content": "<SPECIAL_635>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "636": {
+      "content": "<SPECIAL_636>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "637": {
+      "content": "<SPECIAL_637>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "638": {
+      "content": "<SPECIAL_638>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "639": {
+      "content": "<SPECIAL_639>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "640": {
+      "content": "<SPECIAL_640>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "641": {
+      "content": "<SPECIAL_641>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "642": {
+      "content": "<SPECIAL_642>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "643": {
+      "content": "<SPECIAL_643>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "644": {
+      "content": "<SPECIAL_644>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "645": {
+      "content": "<SPECIAL_645>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "646": {
+      "content": "<SPECIAL_646>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "647": {
+      "content": "<SPECIAL_647>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "648": {
+      "content": "<SPECIAL_648>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "649": {
+      "content": "<SPECIAL_649>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "650": {
+      "content": "<SPECIAL_650>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "651": {
+      "content": "<SPECIAL_651>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "652": {
+      "content": "<SPECIAL_652>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "653": {
+      "content": "<SPECIAL_653>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "654": {
+      "content": "<SPECIAL_654>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "655": {
+      "content": "<SPECIAL_655>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "656": {
+      "content": "<SPECIAL_656>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "657": {
+      "content": "<SPECIAL_657>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "658": {
+      "content": "<SPECIAL_658>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "659": {
+      "content": "<SPECIAL_659>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "660": {
+      "content": "<SPECIAL_660>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "661": {
+      "content": "<SPECIAL_661>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "662": {
+      "content": "<SPECIAL_662>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "663": {
+      "content": "<SPECIAL_663>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "664": {
+      "content": "<SPECIAL_664>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "665": {
+      "content": "<SPECIAL_665>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "666": {
+      "content": "<SPECIAL_666>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "667": {
+      "content": "<SPECIAL_667>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "668": {
+      "content": "<SPECIAL_668>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "669": {
+      "content": "<SPECIAL_669>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "670": {
+      "content": "<SPECIAL_670>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "671": {
+      "content": "<SPECIAL_671>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "672": {
+      "content": "<SPECIAL_672>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "673": {
+      "content": "<SPECIAL_673>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "674": {
+      "content": "<SPECIAL_674>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "675": {
+      "content": "<SPECIAL_675>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "676": {
+      "content": "<SPECIAL_676>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "677": {
+      "content": "<SPECIAL_677>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "678": {
+      "content": "<SPECIAL_678>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "679": {
+      "content": "<SPECIAL_679>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "680": {
+      "content": "<SPECIAL_680>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "681": {
+      "content": "<SPECIAL_681>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "682": {
+      "content": "<SPECIAL_682>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "683": {
+      "content": "<SPECIAL_683>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "684": {
+      "content": "<SPECIAL_684>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "685": {
+      "content": "<SPECIAL_685>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "686": {
+      "content": "<SPECIAL_686>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "687": {
+      "content": "<SPECIAL_687>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "688": {
+      "content": "<SPECIAL_688>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "689": {
+      "content": "<SPECIAL_689>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "690": {
+      "content": "<SPECIAL_690>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "691": {
+      "content": "<SPECIAL_691>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "692": {
+      "content": "<SPECIAL_692>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "693": {
+      "content": "<SPECIAL_693>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "694": {
+      "content": "<SPECIAL_694>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "695": {
+      "content": "<SPECIAL_695>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "696": {
+      "content": "<SPECIAL_696>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "697": {
+      "content": "<SPECIAL_697>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "698": {
+      "content": "<SPECIAL_698>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "699": {
+      "content": "<SPECIAL_699>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "700": {
+      "content": "<SPECIAL_700>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "701": {
+      "content": "<SPECIAL_701>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "702": {
+      "content": "<SPECIAL_702>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "703": {
+      "content": "<SPECIAL_703>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "704": {
+      "content": "<SPECIAL_704>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "705": {
+      "content": "<SPECIAL_705>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "706": {
+      "content": "<SPECIAL_706>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "707": {
+      "content": "<SPECIAL_707>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "708": {
+      "content": "<SPECIAL_708>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "709": {
+      "content": "<SPECIAL_709>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "710": {
+      "content": "<SPECIAL_710>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "711": {
+      "content": "<SPECIAL_711>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "712": {
+      "content": "<SPECIAL_712>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "713": {
+      "content": "<SPECIAL_713>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "714": {
+      "content": "<SPECIAL_714>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "715": {
+      "content": "<SPECIAL_715>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "716": {
+      "content": "<SPECIAL_716>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "717": {
+      "content": "<SPECIAL_717>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "718": {
+      "content": "<SPECIAL_718>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "719": {
+      "content": "<SPECIAL_719>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "720": {
+      "content": "<SPECIAL_720>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "721": {
+      "content": "<SPECIAL_721>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "722": {
+      "content": "<SPECIAL_722>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "723": {
+      "content": "<SPECIAL_723>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "724": {
+      "content": "<SPECIAL_724>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "725": {
+      "content": "<SPECIAL_725>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "726": {
+      "content": "<SPECIAL_726>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "727": {
+      "content": "<SPECIAL_727>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "728": {
+      "content": "<SPECIAL_728>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "729": {
+      "content": "<SPECIAL_729>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "730": {
+      "content": "<SPECIAL_730>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "731": {
+      "content": "<SPECIAL_731>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "732": {
+      "content": "<SPECIAL_732>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "733": {
+      "content": "<SPECIAL_733>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "734": {
+      "content": "<SPECIAL_734>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "735": {
+      "content": "<SPECIAL_735>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "736": {
+      "content": "<SPECIAL_736>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "737": {
+      "content": "<SPECIAL_737>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "738": {
+      "content": "<SPECIAL_738>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "739": {
+      "content": "<SPECIAL_739>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "740": {
+      "content": "<SPECIAL_740>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "741": {
+      "content": "<SPECIAL_741>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "742": {
+      "content": "<SPECIAL_742>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "743": {
+      "content": "<SPECIAL_743>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "744": {
+      "content": "<SPECIAL_744>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "745": {
+      "content": "<SPECIAL_745>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "746": {
+      "content": "<SPECIAL_746>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "747": {
+      "content": "<SPECIAL_747>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "748": {
+      "content": "<SPECIAL_748>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "749": {
+      "content": "<SPECIAL_749>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "750": {
+      "content": "<SPECIAL_750>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "751": {
+      "content": "<SPECIAL_751>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "752": {
+      "content": "<SPECIAL_752>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "753": {
+      "content": "<SPECIAL_753>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "754": {
+      "content": "<SPECIAL_754>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "755": {
+      "content": "<SPECIAL_755>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "756": {
+      "content": "<SPECIAL_756>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "757": {
+      "content": "<SPECIAL_757>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "758": {
+      "content": "<SPECIAL_758>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "759": {
+      "content": "<SPECIAL_759>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "760": {
+      "content": "<SPECIAL_760>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "761": {
+      "content": "<SPECIAL_761>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "762": {
+      "content": "<SPECIAL_762>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "763": {
+      "content": "<SPECIAL_763>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "764": {
+      "content": "<SPECIAL_764>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "765": {
+      "content": "<SPECIAL_765>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "766": {
+      "content": "<SPECIAL_766>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "767": {
+      "content": "<SPECIAL_767>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "768": {
+      "content": "<SPECIAL_768>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "769": {
+      "content": "<SPECIAL_769>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "770": {
+      "content": "<SPECIAL_770>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "771": {
+      "content": "<SPECIAL_771>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "772": {
+      "content": "<SPECIAL_772>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "773": {
+      "content": "<SPECIAL_773>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "774": {
+      "content": "<SPECIAL_774>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "775": {
+      "content": "<SPECIAL_775>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "776": {
+      "content": "<SPECIAL_776>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "777": {
+      "content": "<SPECIAL_777>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "778": {
+      "content": "<SPECIAL_778>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "779": {
+      "content": "<SPECIAL_779>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "780": {
+      "content": "<SPECIAL_780>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "781": {
+      "content": "<SPECIAL_781>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "782": {
+      "content": "<SPECIAL_782>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "783": {
+      "content": "<SPECIAL_783>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "784": {
+      "content": "<SPECIAL_784>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "785": {
+      "content": "<SPECIAL_785>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "786": {
+      "content": "<SPECIAL_786>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "787": {
+      "content": "<SPECIAL_787>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "788": {
+      "content": "<SPECIAL_788>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "789": {
+      "content": "<SPECIAL_789>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "790": {
+      "content": "<SPECIAL_790>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "791": {
+      "content": "<SPECIAL_791>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "792": {
+      "content": "<SPECIAL_792>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "793": {
+      "content": "<SPECIAL_793>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "794": {
+      "content": "<SPECIAL_794>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "795": {
+      "content": "<SPECIAL_795>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "796": {
+      "content": "<SPECIAL_796>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "797": {
+      "content": "<SPECIAL_797>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "798": {
+      "content": "<SPECIAL_798>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "799": {
+      "content": "<SPECIAL_799>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "800": {
+      "content": "<SPECIAL_800>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "801": {
+      "content": "<SPECIAL_801>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "802": {
+      "content": "<SPECIAL_802>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "803": {
+      "content": "<SPECIAL_803>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "804": {
+      "content": "<SPECIAL_804>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "805": {
+      "content": "<SPECIAL_805>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "806": {
+      "content": "<SPECIAL_806>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "807": {
+      "content": "<SPECIAL_807>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "808": {
+      "content": "<SPECIAL_808>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "809": {
+      "content": "<SPECIAL_809>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "810": {
+      "content": "<SPECIAL_810>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "811": {
+      "content": "<SPECIAL_811>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "812": {
+      "content": "<SPECIAL_812>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "813": {
+      "content": "<SPECIAL_813>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "814": {
+      "content": "<SPECIAL_814>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "815": {
+      "content": "<SPECIAL_815>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "816": {
+      "content": "<SPECIAL_816>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "817": {
+      "content": "<SPECIAL_817>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "818": {
+      "content": "<SPECIAL_818>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "819": {
+      "content": "<SPECIAL_819>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "820": {
+      "content": "<SPECIAL_820>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "821": {
+      "content": "<SPECIAL_821>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "822": {
+      "content": "<SPECIAL_822>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "823": {
+      "content": "<SPECIAL_823>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "824": {
+      "content": "<SPECIAL_824>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "825": {
+      "content": "<SPECIAL_825>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "826": {
+      "content": "<SPECIAL_826>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "827": {
+      "content": "<SPECIAL_827>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "828": {
+      "content": "<SPECIAL_828>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "829": {
+      "content": "<SPECIAL_829>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "830": {
+      "content": "<SPECIAL_830>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "831": {
+      "content": "<SPECIAL_831>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "832": {
+      "content": "<SPECIAL_832>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "833": {
+      "content": "<SPECIAL_833>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "834": {
+      "content": "<SPECIAL_834>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "835": {
+      "content": "<SPECIAL_835>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "836": {
+      "content": "<SPECIAL_836>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "837": {
+      "content": "<SPECIAL_837>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "838": {
+      "content": "<SPECIAL_838>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "839": {
+      "content": "<SPECIAL_839>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "840": {
+      "content": "<SPECIAL_840>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "841": {
+      "content": "<SPECIAL_841>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "842": {
+      "content": "<SPECIAL_842>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "843": {
+      "content": "<SPECIAL_843>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "844": {
+      "content": "<SPECIAL_844>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "845": {
+      "content": "<SPECIAL_845>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "846": {
+      "content": "<SPECIAL_846>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "847": {
+      "content": "<SPECIAL_847>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "848": {
+      "content": "<SPECIAL_848>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "849": {
+      "content": "<SPECIAL_849>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "850": {
+      "content": "<SPECIAL_850>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "851": {
+      "content": "<SPECIAL_851>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "852": {
+      "content": "<SPECIAL_852>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "853": {
+      "content": "<SPECIAL_853>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "854": {
+      "content": "<SPECIAL_854>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "855": {
+      "content": "<SPECIAL_855>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "856": {
+      "content": "<SPECIAL_856>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "857": {
+      "content": "<SPECIAL_857>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "858": {
+      "content": "<SPECIAL_858>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "859": {
+      "content": "<SPECIAL_859>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "860": {
+      "content": "<SPECIAL_860>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "861": {
+      "content": "<SPECIAL_861>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "862": {
+      "content": "<SPECIAL_862>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "863": {
+      "content": "<SPECIAL_863>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "864": {
+      "content": "<SPECIAL_864>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "865": {
+      "content": "<SPECIAL_865>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "866": {
+      "content": "<SPECIAL_866>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "867": {
+      "content": "<SPECIAL_867>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "868": {
+      "content": "<SPECIAL_868>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "869": {
+      "content": "<SPECIAL_869>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "870": {
+      "content": "<SPECIAL_870>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "871": {
+      "content": "<SPECIAL_871>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "872": {
+      "content": "<SPECIAL_872>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "873": {
+      "content": "<SPECIAL_873>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "874": {
+      "content": "<SPECIAL_874>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "875": {
+      "content": "<SPECIAL_875>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "876": {
+      "content": "<SPECIAL_876>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "877": {
+      "content": "<SPECIAL_877>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "878": {
+      "content": "<SPECIAL_878>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "879": {
+      "content": "<SPECIAL_879>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "880": {
+      "content": "<SPECIAL_880>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "881": {
+      "content": "<SPECIAL_881>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "882": {
+      "content": "<SPECIAL_882>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "883": {
+      "content": "<SPECIAL_883>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "884": {
+      "content": "<SPECIAL_884>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "885": {
+      "content": "<SPECIAL_885>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "886": {
+      "content": "<SPECIAL_886>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "887": {
+      "content": "<SPECIAL_887>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "888": {
+      "content": "<SPECIAL_888>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "889": {
+      "content": "<SPECIAL_889>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "890": {
+      "content": "<SPECIAL_890>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "891": {
+      "content": "<SPECIAL_891>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "892": {
+      "content": "<SPECIAL_892>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "893": {
+      "content": "<SPECIAL_893>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "894": {
+      "content": "<SPECIAL_894>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "895": {
+      "content": "<SPECIAL_895>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "896": {
+      "content": "<SPECIAL_896>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "897": {
+      "content": "<SPECIAL_897>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "898": {
+      "content": "<SPECIAL_898>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "899": {
+      "content": "<SPECIAL_899>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "900": {
+      "content": "<SPECIAL_900>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "901": {
+      "content": "<SPECIAL_901>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "902": {
+      "content": "<SPECIAL_902>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "903": {
+      "content": "<SPECIAL_903>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "904": {
+      "content": "<SPECIAL_904>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "905": {
+      "content": "<SPECIAL_905>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "906": {
+      "content": "<SPECIAL_906>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "907": {
+      "content": "<SPECIAL_907>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "908": {
+      "content": "<SPECIAL_908>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "909": {
+      "content": "<SPECIAL_909>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "910": {
+      "content": "<SPECIAL_910>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "911": {
+      "content": "<SPECIAL_911>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "912": {
+      "content": "<SPECIAL_912>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "913": {
+      "content": "<SPECIAL_913>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "914": {
+      "content": "<SPECIAL_914>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "915": {
+      "content": "<SPECIAL_915>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "916": {
+      "content": "<SPECIAL_916>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "917": {
+      "content": "<SPECIAL_917>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "918": {
+      "content": "<SPECIAL_918>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "919": {
+      "content": "<SPECIAL_919>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "920": {
+      "content": "<SPECIAL_920>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "921": {
+      "content": "<SPECIAL_921>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "922": {
+      "content": "<SPECIAL_922>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "923": {
+      "content": "<SPECIAL_923>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "924": {
+      "content": "<SPECIAL_924>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "925": {
+      "content": "<SPECIAL_925>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "926": {
+      "content": "<SPECIAL_926>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "927": {
+      "content": "<SPECIAL_927>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "928": {
+      "content": "<SPECIAL_928>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "929": {
+      "content": "<SPECIAL_929>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "930": {
+      "content": "<SPECIAL_930>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "931": {
+      "content": "<SPECIAL_931>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "932": {
+      "content": "<SPECIAL_932>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "933": {
+      "content": "<SPECIAL_933>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "934": {
+      "content": "<SPECIAL_934>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "935": {
+      "content": "<SPECIAL_935>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "936": {
+      "content": "<SPECIAL_936>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "937": {
+      "content": "<SPECIAL_937>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "938": {
+      "content": "<SPECIAL_938>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "939": {
+      "content": "<SPECIAL_939>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "940": {
+      "content": "<SPECIAL_940>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "941": {
+      "content": "<SPECIAL_941>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "942": {
+      "content": "<SPECIAL_942>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "943": {
+      "content": "<SPECIAL_943>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "944": {
+      "content": "<SPECIAL_944>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "945": {
+      "content": "<SPECIAL_945>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "946": {
+      "content": "<SPECIAL_946>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "947": {
+      "content": "<SPECIAL_947>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "948": {
+      "content": "<SPECIAL_948>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "949": {
+      "content": "<SPECIAL_949>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "950": {
+      "content": "<SPECIAL_950>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "951": {
+      "content": "<SPECIAL_951>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "952": {
+      "content": "<SPECIAL_952>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "953": {
+      "content": "<SPECIAL_953>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "954": {
+      "content": "<SPECIAL_954>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "955": {
+      "content": "<SPECIAL_955>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "956": {
+      "content": "<SPECIAL_956>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "957": {
+      "content": "<SPECIAL_957>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "958": {
+      "content": "<SPECIAL_958>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "959": {
+      "content": "<SPECIAL_959>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "960": {
+      "content": "<SPECIAL_960>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "961": {
+      "content": "<SPECIAL_961>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "962": {
+      "content": "<SPECIAL_962>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "963": {
+      "content": "<SPECIAL_963>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "964": {
+      "content": "<SPECIAL_964>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "965": {
+      "content": "<SPECIAL_965>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "966": {
+      "content": "<SPECIAL_966>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "967": {
+      "content": "<SPECIAL_967>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "968": {
+      "content": "<SPECIAL_968>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "969": {
+      "content": "<SPECIAL_969>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "970": {
+      "content": "<SPECIAL_970>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "971": {
+      "content": "<SPECIAL_971>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "972": {
+      "content": "<SPECIAL_972>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "973": {
+      "content": "<SPECIAL_973>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "974": {
+      "content": "<SPECIAL_974>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "975": {
+      "content": "<SPECIAL_975>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "976": {
+      "content": "<SPECIAL_976>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "977": {
+      "content": "<SPECIAL_977>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "978": {
+      "content": "<SPECIAL_978>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "979": {
+      "content": "<SPECIAL_979>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "980": {
+      "content": "<SPECIAL_980>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "981": {
+      "content": "<SPECIAL_981>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "982": {
+      "content": "<SPECIAL_982>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "983": {
+      "content": "<SPECIAL_983>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "984": {
+      "content": "<SPECIAL_984>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "985": {
+      "content": "<SPECIAL_985>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "986": {
+      "content": "<SPECIAL_986>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "987": {
+      "content": "<SPECIAL_987>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "988": {
+      "content": "<SPECIAL_988>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "989": {
+      "content": "<SPECIAL_989>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "990": {
+      "content": "<SPECIAL_990>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "991": {
+      "content": "<SPECIAL_991>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "992": {
+      "content": "<SPECIAL_992>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "993": {
+      "content": "<SPECIAL_993>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "994": {
+      "content": "<SPECIAL_994>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "995": {
+      "content": "<SPECIAL_995>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "996": {
+      "content": "<SPECIAL_996>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "997": {
+      "content": "<SPECIAL_997>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "998": {
+      "content": "<SPECIAL_998>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "999": {
+      "content": "<SPECIAL_999>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<s>",
+  "chat_template": "{%- if messages[0][\"role\"] == \"system\" %}\n    {%- set system_message = messages[0][\"content\"] %}\n    {%- set loop_messages = messages[1:] %}\n{%- else %}\n    {%- set loop_messages = messages %}\n{%- endif %}\n{%- if not tools is defined %}\n    {%- set tools = none %}\n{%- endif %}\n{%- set user_messages = loop_messages | selectattr(\"role\", \"equalto\", \"user\") | list %}\n\n{#- This block checks for alternating user/assistant messages, skipping tool calling messages #}\n{%- set ns = namespace() %}\n{%- set ns.index = 0 %}\n{%- for message in loop_messages %}\n    {%- if not (message.role == \"tool\" or message.role == \"tool_results\" or (message.tool_calls is defined and message.tool_calls is not none)) %}\n        {%- if (message[\"role\"] == \"user\") != (ns.index % 2 == 0) %}\n            {{- raise_exception(\"After the optional system message, conversation roles must alternate user/assistant/user/assistant/...\") }}\n        {%- endif %}\n        {%- set ns.index = ns.index + 1 %}\n    {%- endif %}\n{%- endfor %}\n\n{{- bos_token }}\n{%- for message in loop_messages %}\n    {%- if message[\"role\"] == \"user\" %}\n        {%- if tools is not none and (message == user_messages[-1]) %}\n            {{- \"[AVAILABLE_TOOLS][\" }}\n            {%- for tool in tools %}\n                {%- set tool = tool.function %}\n                {{- '{\"type\": \"function\", \"function\": {' }}\n                {%- for key, val in tool.items() if key != \"return\" %}\n                    {%- if val is string %}\n                        {{- '\"' + key + '\": \"' + val + '\"' }}\n                    {%- else %}\n                        {{- '\"' + key + '\": ' + val|tojson }}\n                    {%- endif %}\n                    {%- if not loop.last %}\n                        {{- \", \" }}\n                    {%- endif %}\n                {%- endfor %}\n                {{- \"}}\" }}\n                {%- if not loop.last %}\n                    {{- \", \" }}\n                {%- else %}\n                    {{- \"]\" }}\n                {%- endif %}\n            {%- endfor %}\n            {{- \"[/AVAILABLE_TOOLS]\" }}\n            {%- endif %}\n        {%- if loop.last and system_message is defined %}\n            {{- \"[INST]\" + system_message + \"\\n\\n\" + message[\"content\"] + \"[/INST]\" }}\n        {%- else %}\n            {{- \"[INST]\" + message[\"content\"] + \"[/INST]\" }}\n        {%- endif %}\n    {%- elif (message.tool_calls is defined and message.tool_calls is not none) %}\n        {{- \"[TOOL_CALLS][\" }}\n        {%- for tool_call in message.tool_calls %}\n            {%- set out = tool_call.function|tojson %}\n            {{- out[:-1] }}\n            {%- if not tool_call.id is defined or tool_call.id|length != 9 %}\n                {{- raise_exception(\"Tool call IDs should be alphanumeric strings with length 9!\") }}\n            {%- endif %}\n            {{- ', \"id\": \"' + tool_call.id + '\"}' }}\n            {%- if not loop.last %}\n                {{- \", \" }}\n            {%- else %}\n                {{- \"]\" + eos_token }}\n            {%- endif %}\n        {%- endfor %}\n    {%- elif message[\"role\"] == \"assistant\" %}\n        {{- message[\"content\"] + eos_token}}\n    {%- elif message[\"role\"] == \"tool_results\" or message[\"role\"] == \"tool\" %}\n        {%- if message.content is defined and message.content.content is defined %}\n            {%- set content = message.content.content %}\n        {%- else %}\n            {%- set content = message.content %}\n        {%- endif %}\n        {{- '[TOOL_RESULTS]{\"content\": ' + content|string + \", \" }}\n        {%- if not message.tool_call_id is defined or message.tool_call_id|length != 9 %}\n            {{- raise_exception(\"Tool call IDs should be alphanumeric strings with length 9!\") }}\n        {%- endif %}\n        {{- '\"call_id\": \"' + message.tool_call_id + '\"}[/TOOL_RESULTS]' }}\n    {%- else %}\n        {{- raise_exception(\"Only user and assistant roles are supported, with the exception of an initial optional system message!\") }}\n    {%- endif %}\n{%- endfor %}\n",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "</s>",
+  "extra_special_tokens": {},
+  "model_input_names": [
+    "input_ids",
+    "attention_mask"
+  ],
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "</s>",
+  "tokenizer_class": "PreTrainedTokenizer",
+  "unk_token": "<unk>"
+}
diff --git a/checkpoint-600/trainer_state.json b/checkpoint-600/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..5dfafb883edfdbbe519eebfd12989d09164661ed
--- /dev/null
+++ b/checkpoint-600/trainer_state.json
@@ -0,0 +1,7534 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 0.8391608391608392,
+  "eval_steps": 500,
+  "global_step": 600,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 94.14286041259766,
+      "epoch": 0.0013986013986013986,
+      "grad_norm": 0.09098726750251449,
+      "kl": 0.0003147125244140625,
+      "learning_rate": 3.496503496503497e-07,
+      "loss": -0.0001,
+      "num_tokens": 19338.0,
+      "reward": 1.3853150606155396,
+      "reward_std": 0.4827312231063843,
+      "rewards/check_gptzero_func": 0.3214285671710968,
+      "rewards/check_perplexity_diff_func": 0.6428571343421936,
+      "rewards/check_winston_local_func": 0.42102929949760437,
+      "step": 1
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.002797202797202797,
+      "grad_norm": 0.09100260462019598,
+      "kl": 0.0003147125244140625,
+      "learning_rate": 6.993006993006994e-07,
+      "loss": -0.0001,
+      "step": 2
+    },
+    {
+      "clip_ratio": 0.0031990089919418097,
+      "completion_length": 97.73214721679688,
+      "epoch": 0.004195804195804196,
+      "grad_norm": 0.16978190242842756,
+      "kl": 0.0003662109375,
+      "learning_rate": 1.0489510489510491e-06,
+      "loss": -0.0017,
+      "num_tokens": 39487.0,
+      "reward": 0.9710169434547424,
+      "reward_std": 0.6376833319664001,
+      "rewards/check_gptzero_func": 0.125,
+      "rewards/check_perplexity_diff_func": 0.5,
+      "rewards/check_winston_local_func": 0.34601688385009766,
+      "step": 3
+    },
+    {
+      "clip_ratio": 0.0018373647471889853,
+      "epoch": 0.005594405594405594,
+      "grad_norm": 0.17663660241053314,
+      "kl": 0.000423431396484375,
+      "learning_rate": 1.3986013986013987e-06,
+      "loss": -0.0013,
+      "step": 4
+    },
+    {
+      "clip_ratio": 0.0016871786210685968,
+      "completion_length": 72.83928680419922,
+      "epoch": 0.006993006993006993,
+      "grad_norm": 0.09466520543424979,
+      "kl": 0.00034332275390625,
+      "learning_rate": 1.7482517482517485e-06,
+      "loss": 0.0008,
+      "num_tokens": 56082.0,
+      "reward": 1.0736479759216309,
+      "reward_std": 0.6862176656723022,
+      "rewards/check_gptzero_func": 0.2321428507566452,
+      "rewards/check_perplexity_diff_func": 0.3928571343421936,
+      "rewards/check_winston_local_func": 0.4486479163169861,
+      "step": 5
+    },
+    {
+      "clip_ratio": 0.0010098000057041645,
+      "epoch": 0.008391608391608392,
+      "grad_norm": 0.09539550614656178,
+      "kl": 0.0003376007080078125,
+      "learning_rate": 2.0979020979020983e-06,
+      "loss": 0.001,
+      "step": 6
+    },
+    {
+      "clip_ratio": 0.00184189947322011,
+      "completion_length": 95.92857360839844,
+      "epoch": 0.009790209790209791,
+      "grad_norm": 0.10573631695411624,
+      "kl": 0.0003757476806640625,
+      "learning_rate": 2.4475524475524477e-06,
+      "loss": 0.001,
+      "num_tokens": 75906.0,
+      "reward": 1.135968804359436,
+      "reward_std": 0.48974505066871643,
+      "rewards/check_gptzero_func": 0.2142857164144516,
+      "rewards/check_perplexity_diff_func": 0.5,
+      "rewards/check_winston_local_func": 0.42168304324150085,
+      "step": 7
+    },
+    {
+      "clip_ratio": 0.0018530809320509434,
+      "epoch": 0.011188811188811189,
+      "grad_norm": 0.10614532740192534,
+      "kl": 0.0004119873046875,
+      "learning_rate": 2.7972027972027974e-06,
+      "loss": 0.0011,
+      "step": 8
+    },
+    {
+      "clip_ratio": 0.0017129909247159958,
+      "completion_length": 106.76786041259766,
+      "epoch": 0.012587412587412588,
+      "grad_norm": 0.07916450383171113,
+      "kl": 0.000377655029296875,
+      "learning_rate": 3.1468531468531472e-06,
+      "loss": -0.0012,
+      "num_tokens": 96989.0,
+      "reward": 0.9590713977813721,
+      "reward_std": 0.5206417441368103,
+      "rewards/check_gptzero_func": 0.2321428507566452,
+      "rewards/check_perplexity_diff_func": 0.3928571343421936,
+      "rewards/check_winston_local_func": 0.3340713679790497,
+      "step": 9
+    },
+    {
+      "clip_ratio": 0.0034840807784348726,
+      "epoch": 0.013986013986013986,
+      "grad_norm": 0.08061029037170311,
+      "kl": 0.00037384033203125,
+      "learning_rate": 3.496503496503497e-06,
+      "loss": -0.0008,
+      "step": 10
+    },
+    {
+      "clip_ratio": 0.0017933619674295187,
+      "completion_length": 102.85714721679688,
+      "epoch": 0.015384615384615385,
+      "grad_norm": 0.07930375667041344,
+      "kl": 0.00030517578125,
+      "learning_rate": 3.846153846153847e-06,
+      "loss": -0.0038,
+      "num_tokens": 117271.0,
+      "reward": 1.1903148889541626,
+      "reward_std": 0.4722224473953247,
+      "rewards/check_gptzero_func": 0.2857142984867096,
+      "rewards/check_perplexity_diff_func": 0.5357142686843872,
+      "rewards/check_winston_local_func": 0.368886262178421,
+      "step": 11
+    },
+    {
+      "clip_ratio": 0.0015239757485687733,
+      "epoch": 0.016783216783216783,
+      "grad_norm": 0.07898151410343454,
+      "kl": 0.0003871917724609375,
+      "learning_rate": 4.195804195804197e-06,
+      "loss": -0.0041,
+      "step": 12
+    },
+    {
+      "clip_ratio": 0.0029670600779354572,
+      "completion_length": 104.37500762939453,
+      "epoch": 0.01818181818181818,
+      "grad_norm": 0.09247854491588844,
+      "kl": 0.0003986358642578125,
+      "learning_rate": 4.5454545454545455e-06,
+      "loss": -0.0049,
+      "num_tokens": 138174.0,
+      "reward": 0.9787324070930481,
+      "reward_std": 0.63657546043396,
+      "rewards/check_gptzero_func": 0.1964285671710968,
+      "rewards/check_perplexity_diff_func": 0.5714285969734192,
+      "rewards/check_winston_local_func": 0.21087531745433807,
+      "step": 13
+    },
+    {
+      "clip_ratio": 0.003217793768271804,
+      "epoch": 0.019580419580419582,
+      "grad_norm": 0.09293749933034297,
+      "kl": 0.0003204345703125,
+      "learning_rate": 4.895104895104895e-06,
+      "loss": -0.0045,
+      "step": 14
+    },
+    {
+      "clip_ratio": 0.002677519340068102,
+      "completion_length": 67.41072082519531,
+      "epoch": 0.02097902097902098,
+      "grad_norm": 0.14020673218962584,
+      "kl": 0.000499725341796875,
+      "learning_rate": 5.244755244755245e-06,
+      "loss": -0.008,
+      "num_tokens": 154385.0,
+      "reward": 1.0573337078094482,
+      "reward_std": 0.5388314127922058,
+      "rewards/check_gptzero_func": 0.2857142984867096,
+      "rewards/check_perplexity_diff_func": 0.3571428656578064,
+      "rewards/check_winston_local_func": 0.41447654366493225,
+      "step": 15
+    },
+    {
+      "clip_ratio": 0.0021408216562122107,
+      "epoch": 0.022377622377622378,
+      "grad_norm": 0.16110555979669544,
+      "kl": 0.000530242919921875,
+      "learning_rate": 5.594405594405595e-06,
+      "loss": -0.0081,
+      "step": 16
+    },
+    {
+      "clip_ratio": 0.002852825215086341,
+      "completion_length": 89.83928680419922,
+      "epoch": 0.023776223776223775,
+      "grad_norm": 0.1066932843002634,
+      "kl": 0.000415802001953125,
+      "learning_rate": 5.944055944055945e-06,
+      "loss": 0.0166,
+      "num_tokens": 173158.0,
+      "reward": 1.2226747274398804,
+      "reward_std": 0.45578521490097046,
+      "rewards/check_gptzero_func": 0.3392857015132904,
+      "rewards/check_perplexity_diff_func": 0.5,
+      "rewards/check_winston_local_func": 0.3833889663219452,
+      "step": 17
+    },
+    {
+      "clip_ratio": 0.002018069615587592,
+      "epoch": 0.025174825174825177,
+      "grad_norm": 0.10578254537451186,
+      "kl": 0.0003833770751953125,
+      "learning_rate": 6.2937062937062944e-06,
+      "loss": 0.0164,
+      "step": 18
+    },
+    {
+      "clip_ratio": 0.0032663617748767138,
+      "completion_length": 105.83928680419922,
+      "epoch": 0.026573426573426574,
+      "grad_norm": 0.11554707163597037,
+      "kl": 0.0003719329833984375,
+      "learning_rate": 6.643356643356643e-06,
+      "loss": -0.0005,
+      "num_tokens": 194649.0,
+      "reward": 0.8957912921905518,
+      "reward_std": 0.47702187299728394,
+      "rewards/check_gptzero_func": 0.1785714328289032,
+      "rewards/check_perplexity_diff_func": 0.3571428656578064,
+      "rewards/check_winston_local_func": 0.3600769340991974,
+      "step": 19
+    },
+    {
+      "clip_ratio": 0.0024422537535429,
+      "epoch": 0.027972027972027972,
+      "grad_norm": 0.11417666172406592,
+      "kl": 0.0003814697265625,
+      "learning_rate": 6.993006993006994e-06,
+      "loss": -0.0008,
+      "step": 20
+    },
+    {
+      "clip_ratio": 0.0013639701064676046,
+      "completion_length": 107.39286041259766,
+      "epoch": 0.02937062937062937,
+      "grad_norm": 0.09634196153318929,
+      "kl": 0.000396728515625,
+      "learning_rate": 7.342657342657343e-06,
+      "loss": -0.0011,
+      "num_tokens": 215987.0,
+      "reward": 0.9197577238082886,
+      "reward_std": 0.39100831747055054,
+      "rewards/check_gptzero_func": 0.2321428507566452,
+      "rewards/check_perplexity_diff_func": 0.3571428656578064,
+      "rewards/check_winston_local_func": 0.33047202229499817,
+      "step": 21
+    },
+    {
+      "clip_ratio": 0.003595889313146472,
+      "epoch": 0.03076923076923077,
+      "grad_norm": 0.09779857975187953,
+      "kl": 0.00052642822265625,
+      "learning_rate": 7.692307692307694e-06,
+      "loss": -0.0013,
+      "step": 22
+    },
+    {
+      "clip_ratio": 0.0024316231720149517,
+      "completion_length": 89.08928680419922,
+      "epoch": 0.032167832167832165,
+      "grad_norm": 0.13189477135999747,
+      "kl": 0.0003414154052734375,
+      "learning_rate": 8.041958041958042e-06,
+      "loss": -0.0009,
+      "num_tokens": 234678.0,
+      "reward": 0.9593304395675659,
+      "reward_std": 0.5972030758857727,
+      "rewards/check_gptzero_func": 0.1964285671710968,
+      "rewards/check_perplexity_diff_func": 0.4285714328289032,
+      "rewards/check_winston_local_func": 0.33433040976524353,
+      "step": 23
+    },
+    {
+      "clip_ratio": 0.000728283659555018,
+      "epoch": 0.033566433566433566,
+      "grad_norm": 0.12953291972672934,
+      "kl": 0.000335693359375,
+      "learning_rate": 8.391608391608393e-06,
+      "loss": -0.0011,
+      "step": 24
+    },
+    {
+      "clip_ratio": 0.0021546650677919388,
+      "completion_length": 90.98214721679688,
+      "epoch": 0.03496503496503497,
+      "grad_norm": 0.07682474826469807,
+      "kl": 0.0004558563232421875,
+      "learning_rate": 8.741258741258741e-06,
+      "loss": 0.0077,
+      "num_tokens": 253833.0,
+      "reward": 0.8255766034126282,
+      "reward_std": 0.3988816440105438,
+      "rewards/check_gptzero_func": 0.2678571343421936,
+      "rewards/check_perplexity_diff_func": 0.2857142984867096,
+      "rewards/check_winston_local_func": 0.272005170583725,
+      "step": 25
+    },
+    {
+      "clip_ratio": 0.0019250252516940236,
+      "epoch": 0.03636363636363636,
+      "grad_norm": 0.07873511079065725,
+      "kl": 0.0003490447998046875,
+      "learning_rate": 9.090909090909091e-06,
+      "loss": 0.0075,
+      "step": 26
+    },
+    {
+      "clip_ratio": 0.001886485842987895,
+      "completion_length": 74.4464340209961,
+      "epoch": 0.03776223776223776,
+      "grad_norm": 0.06765277577538054,
+      "kl": 0.000354766845703125,
+      "learning_rate": 9.44055944055944e-06,
+      "loss": -0.0003,
+      "num_tokens": 270908.0,
+      "reward": 1.3086735010147095,
+      "reward_std": 0.3881511986255646,
+      "rewards/check_gptzero_func": 0.4107142984867096,
+      "rewards/check_perplexity_diff_func": 0.4642857015132904,
+      "rewards/check_winston_local_func": 0.4336733818054199,
+      "step": 27
+    },
+    {
+      "clip_ratio": 0.0020898371003568172,
+      "epoch": 0.039160839160839164,
+      "grad_norm": 0.06722581226675682,
+      "kl": 0.000453948974609375,
+      "learning_rate": 9.79020979020979e-06,
+      "loss": -0.0006,
+      "step": 28
+    },
+    {
+      "clip_ratio": 0.002855924190953374,
+      "completion_length": 103.64286041259766,
+      "epoch": 0.04055944055944056,
+      "grad_norm": 0.16301390061045865,
+      "kl": 0.0003948211669921875,
+      "learning_rate": 1.013986013986014e-05,
+      "loss": 0.0063,
+      "num_tokens": 292060.0,
+      "reward": 1.22488272190094,
+      "reward_std": 0.8093323111534119,
+      "rewards/check_gptzero_func": 0.2678571343421936,
+      "rewards/check_perplexity_diff_func": 0.6071428656578064,
+      "rewards/check_winston_local_func": 0.34988275170326233,
+      "step": 29
+    },
+    {
+      "clip_ratio": 0.0016208746237680316,
+      "epoch": 0.04195804195804196,
+      "grad_norm": 0.17335125351231856,
+      "kl": 0.000507354736328125,
+      "learning_rate": 1.048951048951049e-05,
+      "loss": 0.0059,
+      "step": 30
+    },
+    {
+      "clip_ratio": 0.0011572305811569095,
+      "completion_length": 82.91072082519531,
+      "epoch": 0.043356643356643354,
+      "grad_norm": 0.11830097508590764,
+      "kl": 0.0006256103515625,
+      "learning_rate": 1.0839160839160838e-05,
+      "loss": 0.002,
+      "num_tokens": 310163.0,
+      "reward": 0.9069231748580933,
+      "reward_std": 0.4621748924255371,
+      "rewards/check_gptzero_func": 0.25,
+      "rewards/check_perplexity_diff_func": 0.3571428656578064,
+      "rewards/check_winston_local_func": 0.29978030920028687,
+      "step": 31
+    },
+    {
+      "clip_ratio": 0.00402362085878849,
+      "epoch": 0.044755244755244755,
+      "grad_norm": 0.11861334422671065,
+      "kl": 0.000762939453125,
+      "learning_rate": 1.118881118881119e-05,
+      "loss": 0.0019,
+      "step": 32
+    },
+    {
+      "clip_ratio": 0.0017474278574809432,
+      "completion_length": 90.9464340209961,
+      "epoch": 0.046153846153846156,
+      "grad_norm": 0.1025496804751534,
+      "kl": 0.000701904296875,
+      "learning_rate": 1.153846153846154e-05,
+      "loss": -0.0014,
+      "num_tokens": 328922.0,
+      "reward": 1.1198338270187378,
+      "reward_std": 0.5045111775398254,
+      "rewards/check_gptzero_func": 0.0892857164144516,
+      "rewards/check_perplexity_diff_func": 0.7857142686843872,
+      "rewards/check_winston_local_func": 0.24483375251293182,
+      "step": 33
+    },
+    {
+      "clip_ratio": 0.0012307984288781881,
+      "epoch": 0.04755244755244755,
+      "grad_norm": 0.10274008519956918,
+      "kl": 0.00087738037109375,
+      "learning_rate": 1.188811188811189e-05,
+      "loss": -0.0015,
+      "step": 34
+    },
+    {
+      "clip_ratio": 0.0029886537231504917,
+      "completion_length": 93.00000762939453,
+      "epoch": 0.04895104895104895,
+      "grad_norm": 0.1311802256012859,
+      "kl": 0.001190185546875,
+      "learning_rate": 1.2237762237762239e-05,
+      "loss": -0.0141,
+      "num_tokens": 348852.0,
+      "reward": 0.9456299543380737,
+      "reward_std": 0.6374980807304382,
+      "rewards/check_gptzero_func": 0.2321428507566452,
+      "rewards/check_perplexity_diff_func": 0.4285714328289032,
+      "rewards/check_winston_local_func": 0.28491565585136414,
+      "step": 35
+    },
+    {
+      "clip_ratio": 0.0028984802775084972,
+      "epoch": 0.05034965034965035,
+      "grad_norm": 0.13267305607921692,
+      "kl": 0.00145721435546875,
+      "learning_rate": 1.2587412587412589e-05,
+      "loss": -0.0147,
+      "step": 36
+    },
+    {
+      "clip_ratio": 0.0018746532732620835,
+      "completion_length": 89.85714721679688,
+      "epoch": 0.05174825174825175,
+      "grad_norm": 0.1497182815907709,
+      "kl": 0.002044677734375,
+      "learning_rate": 1.2937062937062939e-05,
+      "loss": -0.0253,
+      "num_tokens": 367482.0,
+      "reward": 1.2605940103530884,
+      "reward_std": 0.66156405210495,
+      "rewards/check_gptzero_func": 0.1785714328289032,
+      "rewards/check_perplexity_diff_func": 0.75,
+      "rewards/check_winston_local_func": 0.3320225179195404,
+      "step": 37
+    },
+    {
+      "clip_ratio": 0.0020914783235639334,
+      "epoch": 0.05314685314685315,
+      "grad_norm": 0.15027543310666272,
+      "kl": 0.0026092529296875,
+      "learning_rate": 1.3286713286713287e-05,
+      "loss": -0.0257,
+      "step": 38
+    },
+    {
+      "clip_ratio": 0.0017028081929311156,
+      "completion_length": 101.58928680419922,
+      "epoch": 0.05454545454545454,
+      "grad_norm": 0.1306748742984494,
+      "kl": 0.0027008056640625,
+      "learning_rate": 1.3636363636363637e-05,
+      "loss": -0.0091,
+      "num_tokens": 388365.0,
+      "reward": 1.1187876462936401,
+      "reward_std": 0.4044211804866791,
+      "rewards/check_gptzero_func": 0.2321428507566452,
+      "rewards/check_perplexity_diff_func": 0.5357142686843872,
+      "rewards/check_winston_local_func": 0.3509305417537689,
+      "step": 39
+    },
+    {
+      "clip_ratio": 0.003386714030057192,
+      "epoch": 0.055944055944055944,
+      "grad_norm": 0.12880289464580882,
+      "kl": 0.003265380859375,
+      "learning_rate": 1.3986013986013988e-05,
+      "loss": -0.0086,
+      "step": 40
+    },
+    {
+      "clip_ratio": 0.0024523374158889055,
+      "completion_length": 100.41072082519531,
+      "epoch": 0.057342657342657345,
+      "grad_norm": 0.12385816200639352,
+      "kl": 0.004547119140625,
+      "learning_rate": 1.4335664335664336e-05,
+      "loss": -0.0161,
+      "num_tokens": 409074.0,
+      "reward": 0.804709792137146,
+      "reward_std": 0.5129754543304443,
+      "rewards/check_gptzero_func": 0.1428571492433548,
+      "rewards/check_perplexity_diff_func": 0.3928571343421936,
+      "rewards/check_winston_local_func": 0.2689954936504364,
+      "step": 41
+    },
+    {
+      "clip_ratio": 0.004267544951289892,
+      "epoch": 0.05874125874125874,
+      "grad_norm": 0.12301160152684054,
+      "kl": 0.005218505859375,
+      "learning_rate": 1.4685314685314686e-05,
+      "loss": -0.0161,
+      "step": 42
+    },
+    {
+      "clip_ratio": 0.0019341636216267943,
+      "completion_length": 90.46428680419922,
+      "epoch": 0.06013986013986014,
+      "grad_norm": 0.1390908713431379,
+      "kl": 0.0048828125,
+      "learning_rate": 1.5034965034965034e-05,
+      "loss": 0.0068,
+      "num_tokens": 428262.0,
+      "reward": 1.087609887123108,
+      "reward_std": 0.6448312997817993,
+      "rewards/check_gptzero_func": 0.2321428507566452,
+      "rewards/check_perplexity_diff_func": 0.3928571343421936,
+      "rewards/check_winston_local_func": 0.46260982751846313,
+      "step": 43
+    },
+    {
+      "clip_ratio": 0.0016832423862069845,
+      "epoch": 0.06153846153846154,
+      "grad_norm": 0.13873080950458352,
+      "kl": 0.00567626953125,
+      "learning_rate": 1.5384615384615387e-05,
+      "loss": 0.0065,
+      "step": 44
+    },
+    {
+      "clip_ratio": 0.0019954824820160866,
+      "completion_length": 96.92857360839844,
+      "epoch": 0.06293706293706294,
+      "grad_norm": 0.11630124487143968,
+      "kl": 0.00897216796875,
+      "learning_rate": 1.5734265734265734e-05,
+      "loss": 0.008,
+      "num_tokens": 448480.0,
+      "reward": 1.0604994297027588,
+      "reward_std": 0.5378546118736267,
+      "rewards/check_gptzero_func": 0.2321428507566452,
+      "rewards/check_perplexity_diff_func": 0.4642857015132904,
+      "rewards/check_winston_local_func": 0.36407074332237244,
+      "step": 45
+    },
+    {
+      "clip_ratio": 0.002663462422788143,
+      "epoch": 0.06433566433566433,
+      "grad_norm": 0.1149796219701467,
+      "kl": 0.01007080078125,
+      "learning_rate": 1.6083916083916083e-05,
+      "loss": 0.0081,
+      "step": 46
+    },
+    {
+      "clip_ratio": 0.0018039485439658165,
+      "completion_length": 79.51786041259766,
+      "epoch": 0.06573426573426573,
+      "grad_norm": 0.12721969778213826,
+      "kl": 0.01123046875,
+      "learning_rate": 1.6433566433566433e-05,
+      "loss": 0.0112,
+      "num_tokens": 465671.0,
+      "reward": 0.9921315312385559,
+      "reward_std": 0.38579052686691284,
+      "rewards/check_gptzero_func": 0.1607142835855484,
+      "rewards/check_perplexity_diff_func": 0.3928571343421936,
+      "rewards/check_winston_local_func": 0.43856000900268555,
+      "step": 47
+    },
+    {
+      "clip_ratio": 0.0027954555116593838,
+      "epoch": 0.06713286713286713,
+      "grad_norm": 0.12930272448754576,
+      "kl": 0.01300048828125,
+      "learning_rate": 1.6783216783216786e-05,
+      "loss": 0.0107,
+      "step": 48
+    },
+    {
+      "clip_ratio": 0.0028269642498344183,
+      "completion_length": 116.5714340209961,
+      "epoch": 0.06853146853146853,
+      "grad_norm": 0.12430052891449103,
+      "kl": 0.01226806640625,
+      "learning_rate": 1.7132867132867133e-05,
+      "loss": 0.025,
+      "num_tokens": 488367.0,
+      "reward": 0.8830849528312683,
+      "reward_std": 0.5276607871055603,
+      "rewards/check_gptzero_func": 0.25,
+      "rewards/check_perplexity_diff_func": 0.3214285671710968,
+      "rewards/check_winston_local_func": 0.31165632605552673,
+      "step": 49
+    },
+    {
+      "clip_ratio": 0.0023628019262105227,
+      "epoch": 0.06993006993006994,
+      "grad_norm": 0.12293590613717971,
+      "kl": 0.0142822265625,
+      "learning_rate": 1.7482517482517483e-05,
+      "loss": 0.0249,
+      "step": 50
+    },
+    {
+      "clip_ratio": 0.0033425339497625828,
+      "completion_length": 83.39286041259766,
+      "epoch": 0.07132867132867132,
+      "grad_norm": 0.11106847343390151,
+      "kl": 0.01177978515625,
+      "learning_rate": 1.7832167832167836e-05,
+      "loss": -0.0041,
+      "num_tokens": 506551.0,
+      "reward": 1.029296875,
+      "reward_std": 0.4901208281517029,
+      "rewards/check_gptzero_func": 0.2857142984867096,
+      "rewards/check_perplexity_diff_func": 0.3214285671710968,
+      "rewards/check_winston_local_func": 0.42215394973754883,
+      "step": 51
+    },
+    {
+      "clip_ratio": 0.0022109400015324354,
+      "epoch": 0.07272727272727272,
+      "grad_norm": 0.11017795132415774,
+      "kl": 0.013671875,
+      "learning_rate": 1.8181818181818182e-05,
+      "loss": -0.0043,
+      "step": 52
+    },
+    {
+      "clip_ratio": 0.0019256824161857367,
+      "completion_length": 86.5714340209961,
+      "epoch": 0.07412587412587412,
+      "grad_norm": 0.11010712395827207,
+      "kl": 0.0208740234375,
+      "learning_rate": 1.8531468531468532e-05,
+      "loss": -0.0032,
+      "num_tokens": 524863.0,
+      "reward": 0.962996244430542,
+      "reward_std": 0.6607940196990967,
+      "rewards/check_gptzero_func": 0.1071428582072258,
+      "rewards/check_perplexity_diff_func": 0.5,
+      "rewards/check_winston_local_func": 0.355853408575058,
+      "step": 53
+    },
+    {
+      "clip_ratio": 0.0025832760147750378,
+      "epoch": 0.07552447552447553,
+      "grad_norm": 0.10695378216267973,
+      "kl": 0.0250244140625,
+      "learning_rate": 1.888111888111888e-05,
+      "loss": -0.0034,
+      "step": 54
+    },
+    {
+      "clip_ratio": 0.0017639577854424715,
+      "completion_length": 96.08928680419922,
+      "epoch": 0.07692307692307693,
+      "grad_norm": 0.17472941592839772,
+      "kl": 0.0322265625,
+      "learning_rate": 1.923076923076923e-05,
+      "loss": -0.0141,
+      "num_tokens": 544736.0,
+      "reward": 1.3478963375091553,
+      "reward_std": 0.8002303242683411,
+      "rewards/check_gptzero_func": 0.2857142984867096,
+      "rewards/check_perplexity_diff_func": 0.7142857313156128,
+      "rewards/check_winston_local_func": 0.34789615869522095,
+      "step": 55
+    },
+    {
+      "clip_ratio": 0.0034089265391230583,
+      "epoch": 0.07832167832167833,
+      "grad_norm": 0.14666884899599142,
+      "kl": 0.041748046875,
+      "learning_rate": 1.958041958041958e-05,
+      "loss": -0.0155,
+      "step": 56
+    },
+    {
+      "clip_ratio": 0.0026615143287926912,
+      "completion_length": 96.91072082519531,
+      "epoch": 0.07972027972027972,
+      "grad_norm": 0.1400337476406121,
+      "kl": 0.035400390625,
+      "learning_rate": 1.993006993006993e-05,
+      "loss": -0.027,
+      "num_tokens": 564803.0,
+      "reward": 1.0594887733459473,
+      "reward_std": 0.6912091970443726,
+      "rewards/check_gptzero_func": 0.2857142984867096,
+      "rewards/check_perplexity_diff_func": 0.4285714328289032,
+      "rewards/check_winston_local_func": 0.34520307183265686,
+      "step": 57
+    },
+    {
+      "clip_ratio": 0.003153084311634302,
+      "epoch": 0.08111888111888112,
+      "grad_norm": 0.138816359725178,
+      "kl": 0.042236328125,
+      "learning_rate": 2.027972027972028e-05,
+      "loss": -0.0274,
+      "step": 58
+    },
+    {
+      "clip_ratio": 0.0012626759707927704,
+      "completion_length": 76.35714721679688,
+      "epoch": 0.08251748251748252,
+      "grad_norm": 0.15940476459351663,
+      "kl": 0.0556640625,
+      "learning_rate": 2.062937062937063e-05,
+      "loss": -0.0057,
+      "num_tokens": 582265.0,
+      "reward": 0.992591381072998,
+      "reward_std": 0.5938137769699097,
+      "rewards/check_gptzero_func": 0.125,
+      "rewards/check_perplexity_diff_func": 0.4642857015132904,
+      "rewards/check_winston_local_func": 0.4033055901527405,
+      "step": 59
+    },
+    {
+      "clip_ratio": 0.004643784370273352,
+      "epoch": 0.08391608391608392,
+      "grad_norm": 0.15305834024157844,
+      "kl": 0.07421875,
+      "learning_rate": 2.097902097902098e-05,
+      "loss": -0.0064,
+      "step": 60
+    },
+    {
+      "clip_ratio": 0.0032869603019207716,
+      "completion_length": 103.33928680419922,
+      "epoch": 0.08531468531468532,
+      "grad_norm": 0.12260760902895611,
+      "kl": 0.0810546875,
+      "learning_rate": 2.132867132867133e-05,
+      "loss": 0.0066,
+      "num_tokens": 603454.0,
+      "reward": 1.191124439239502,
+      "reward_std": 0.5557684898376465,
+      "rewards/check_gptzero_func": 0.2321428507566452,
+      "rewards/check_perplexity_diff_func": 0.5357142686843872,
+      "rewards/check_winston_local_func": 0.4232672154903412,
+      "step": 61
+    },
+    {
+      "clip_ratio": 0.008058370091021061,
+      "epoch": 0.08671328671328671,
+      "grad_norm": 0.10224499878092017,
+      "kl": 0.1162109375,
+      "learning_rate": 2.1678321678321677e-05,
+      "loss": 0.0058,
+      "step": 62
+    },
+    {
+      "clip_ratio": 0.002202474046498537,
+      "completion_length": 91.4464340209961,
+      "epoch": 0.08811188811188811,
+      "grad_norm": 0.09776882919444682,
+      "kl": 0.2138671875,
+      "learning_rate": 2.202797202797203e-05,
+      "loss": 0.0021,
+      "num_tokens": 622705.0,
+      "reward": 1.1320958137512207,
+      "reward_std": 0.4060821831226349,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 0.3214285671710968,
+      "rewards/check_winston_local_func": 0.41781002283096313,
+      "step": 63
+    },
+    {
+      "clip_ratio": 0.0025903189089149237,
+      "epoch": 0.08951048951048951,
+      "grad_norm": 0.0963149975780195,
+      "kl": 0.2109375,
+      "learning_rate": 2.237762237762238e-05,
+      "loss": 0.0014,
+      "step": 64
+    },
+    {
+      "clip_ratio": 0.0016666523879393935,
+      "completion_length": 100.00000762939453,
+      "epoch": 0.09090909090909091,
+      "grad_norm": 0.12490001498293682,
+      "kl": 0.1220703125,
+      "learning_rate": 2.272727272727273e-05,
+      "loss": 0.0081,
+      "num_tokens": 643191.0,
+      "reward": 1.274580478668213,
+      "reward_std": 0.5277792811393738,
+      "rewards/check_gptzero_func": 0.2321428507566452,
+      "rewards/check_perplexity_diff_func": 0.5357142686843872,
+      "rewards/check_winston_local_func": 0.5067232847213745,
+      "step": 65
+    },
+    {
+      "clip_ratio": 0.0032784033101052046,
+      "epoch": 0.09230769230769231,
+      "grad_norm": 0.12114304348028067,
+      "kl": 0.134765625,
+      "learning_rate": 2.307692307692308e-05,
+      "loss": 0.0074,
+      "step": 66
+    },
+    {
+      "clip_ratio": 0.0017520035617053509,
+      "completion_length": 95.16072082519531,
+      "epoch": 0.0937062937062937,
+      "grad_norm": 0.14292853522837679,
+      "kl": 0.216796875,
+      "learning_rate": 2.342657342657343e-05,
+      "loss": -0.0018,
+      "num_tokens": 662736.0,
+      "reward": 1.1007850170135498,
+      "reward_std": 0.45985397696495056,
+      "rewards/check_gptzero_func": 0.25,
+      "rewards/check_perplexity_diff_func": 0.4642857015132904,
+      "rewards/check_winston_local_func": 0.38649922609329224,
+      "step": 67
+    },
+    {
+      "clip_ratio": 0.00579434959217906,
+      "epoch": 0.0951048951048951,
+      "grad_norm": 0.14340586759573087,
+      "kl": 0.2431640625,
+      "learning_rate": 2.377622377622378e-05,
+      "loss": -0.0029,
+      "step": 68
+    },
+    {
+      "clip_ratio": 0.0022546499967575073,
+      "completion_length": 86.17857360839844,
+      "epoch": 0.0965034965034965,
+      "grad_norm": 0.17287042335415018,
+      "kl": 0.26171875,
+      "learning_rate": 2.4125874125874125e-05,
+      "loss": 0.0059,
+      "num_tokens": 681030.0,
+      "reward": 1.4391711950302124,
+      "reward_std": 0.6516181230545044,
+      "rewards/check_gptzero_func": 0.5178571343421936,
+      "rewards/check_perplexity_diff_func": 0.4642857015132904,
+      "rewards/check_winston_local_func": 0.45702821016311646,
+      "step": 69
+    },
+    {
+      "clip_ratio": 0.004240955226123333,
+      "epoch": 0.0979020979020979,
+      "grad_norm": 0.16371494613893736,
+      "kl": 0.28125,
+      "learning_rate": 2.4475524475524478e-05,
+      "loss": 0.0048,
+      "step": 70
+    },
+    {
+      "clip_ratio": 0.003548440057784319,
+      "completion_length": 102.42857360839844,
+      "epoch": 0.0993006993006993,
+      "grad_norm": 0.1615418523026911,
+      "kl": 0.1708984375,
+      "learning_rate": 2.4825174825174828e-05,
+      "loss": 0.0015,
+      "num_tokens": 701874.0,
+      "reward": 1.4929367303848267,
+      "reward_std": 0.6083499193191528,
+      "rewards/check_gptzero_func": 0.3214285671710968,
+      "rewards/check_perplexity_diff_func": 0.7142857313156128,
+      "rewards/check_winston_local_func": 0.4572224020957947,
+      "step": 71
+    },
+    {
+      "clip_ratio": 0.004331245087087154,
+      "epoch": 0.1006993006993007,
+      "grad_norm": 0.12876017762018707,
+      "kl": 0.216796875,
+      "learning_rate": 2.5174825174825178e-05,
+      "loss": 0.0,
+      "step": 72
+    },
+    {
+      "clip_ratio": 0.003951852675527334,
+      "completion_length": 73.46428680419922,
+      "epoch": 0.1020979020979021,
+      "grad_norm": 0.20791748542634808,
+      "kl": 0.365234375,
+      "learning_rate": 2.5524475524475528e-05,
+      "loss": -0.0037,
+      "num_tokens": 718488.0,
+      "reward": 1.7953797578811646,
+      "reward_std": 0.5393027067184448,
+      "rewards/check_gptzero_func": 0.4464285671710968,
+      "rewards/check_perplexity_diff_func": 0.8571428656578064,
+      "rewards/check_winston_local_func": 0.49180838465690613,
+      "step": 73
+    },
+    {
+      "clip_ratio": 0.008312534540891647,
+      "epoch": 0.1034965034965035,
+      "grad_norm": 0.19232465302245624,
+      "kl": 0.5390625,
+      "learning_rate": 2.5874125874125877e-05,
+      "loss": -0.0053,
+      "step": 74
+    },
+    {
+      "clip_ratio": 0.0021736263297498226,
+      "completion_length": 97.73214721679688,
+      "epoch": 0.1048951048951049,
+      "grad_norm": 0.18504738647098173,
+      "kl": 0.2421875,
+      "learning_rate": 2.6223776223776224e-05,
+      "loss": -0.0023,
+      "num_tokens": 738619.0,
+      "reward": 1.284185767173767,
+      "reward_std": 0.6959513425827026,
+      "rewards/check_gptzero_func": 0.4464285671710968,
+      "rewards/check_perplexity_diff_func": 0.3928571343421936,
+      "rewards/check_winston_local_func": 0.4449000358581543,
+      "step": 75
+    },
+    {
+      "clip_ratio": 0.0033756059128791094,
+      "epoch": 0.1062937062937063,
+      "grad_norm": 0.17585639043073228,
+      "kl": 0.255859375,
+      "learning_rate": 2.6573426573426574e-05,
+      "loss": -0.0043,
+      "step": 76
+    },
+    {
+      "clip_ratio": 0.003740633837878704,
+      "completion_length": 95.67857360839844,
+      "epoch": 0.1076923076923077,
+      "grad_norm": 0.21573775826044048,
+      "kl": 0.2451171875,
+      "learning_rate": 2.6923076923076923e-05,
+      "loss": 0.0105,
+      "num_tokens": 758111.0,
+      "reward": 1.4672702550888062,
+      "reward_std": 0.85318922996521,
+      "rewards/check_gptzero_func": 0.25,
+      "rewards/check_perplexity_diff_func": 0.7857142686843872,
+      "rewards/check_winston_local_func": 0.43155592679977417,
+      "step": 77
+    },
+    {
+      "clip_ratio": 0.004517109598964453,
+      "epoch": 0.10909090909090909,
+      "grad_norm": 0.2071235680791843,
+      "kl": 0.302734375,
+      "learning_rate": 2.7272727272727273e-05,
+      "loss": 0.009,
+      "step": 78
+    },
+    {
+      "clip_ratio": 0.004641580395400524,
+      "completion_length": 117.75000762939453,
+      "epoch": 0.11048951048951049,
+      "grad_norm": 0.19278645376555428,
+      "kl": 0.259765625,
+      "learning_rate": 2.762237762237762e-05,
+      "loss": -0.0004,
+      "num_tokens": 780403.0,
+      "reward": 1.4974778890609741,
+      "reward_std": 0.59557044506073,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 0.6785714030265808,
+      "rewards/check_winston_local_func": 0.39033493399620056,
+      "step": 79
+    },
+    {
+      "clip_ratio": 0.006305322516709566,
+      "epoch": 0.11188811188811189,
+      "grad_norm": 0.13786850726930885,
+      "kl": 0.3125,
+      "learning_rate": 2.7972027972027976e-05,
+      "loss": -0.0019,
+      "step": 80
+    },
+    {
+      "clip_ratio": 0.0017778041074052453,
+      "completion_length": 101.03572082519531,
+      "epoch": 0.11328671328671329,
+      "grad_norm": 0.18099911272050398,
+      "kl": 1.7578125,
+      "learning_rate": 2.8321678321678326e-05,
+      "loss": 0.0053,
+      "num_tokens": 800231.0,
+      "reward": 1.420699954032898,
+      "reward_std": 0.5104399919509888,
+      "rewards/check_gptzero_func": 0.2678571343421936,
+      "rewards/check_perplexity_diff_func": 0.6428571343421936,
+      "rewards/check_winston_local_func": 0.5099858045578003,
+      "step": 81
+    },
+    {
+      "clip_ratio": 0.0030425102449953556,
+      "epoch": 0.11468531468531469,
+      "grad_norm": 0.10725025403922342,
+      "kl": 0.9921875,
+      "learning_rate": 2.8671328671328672e-05,
+      "loss": 0.0043,
+      "step": 82
+    },
+    {
+      "clip_ratio": 0.0030319676734507084,
+      "completion_length": 93.3214340209961,
+      "epoch": 0.11608391608391608,
+      "grad_norm": 0.22858957476463732,
+      "kl": 0.35546875,
+      "learning_rate": 2.9020979020979022e-05,
+      "loss": 0.0021,
+      "num_tokens": 819773.0,
+      "reward": 1.4201573133468628,
+      "reward_std": 0.761705756187439,
+      "rewards/check_gptzero_func": 0.3392857015132904,
+      "rewards/check_perplexity_diff_func": 0.5714285969734192,
+      "rewards/check_winston_local_func": 0.5094431042671204,
+      "step": 83
+    },
+    {
+      "clip_ratio": 0.0027519434224814177,
+      "epoch": 0.11748251748251748,
+      "grad_norm": 0.180287994009811,
+      "kl": 0.392578125,
+      "learning_rate": 2.9370629370629372e-05,
+      "loss": -0.0002,
+      "step": 84
+    },
+    {
+      "clip_ratio": 0.0034523813519626856,
+      "completion_length": 102.66072082519531,
+      "epoch": 0.11888111888111888,
+      "grad_norm": 0.25571068617285075,
+      "kl": 0.494140625,
+      "learning_rate": 2.972027972027972e-05,
+      "loss": -0.0103,
+      "num_tokens": 840280.0,
+      "reward": 1.7226576805114746,
+      "reward_std": 0.7545516490936279,
+      "rewards/check_gptzero_func": 0.5714285969734192,
+      "rewards/check_perplexity_diff_func": 0.6785714030265808,
+      "rewards/check_winston_local_func": 0.4726576805114746,
+      "step": 85
+    },
+    {
+      "clip_ratio": 0.00493992306292057,
+      "epoch": 0.12027972027972028,
+      "grad_norm": 0.2172932526081326,
+      "kl": 0.490234375,
+      "learning_rate": 3.0069930069930068e-05,
+      "loss": -0.013,
+      "step": 86
+    },
+    {
+      "clip_ratio": 0.003154533449560404,
+      "completion_length": 98.96428680419922,
+      "epoch": 0.12167832167832168,
+      "grad_norm": 0.3651883135439141,
+      "kl": 0.5859375,
+      "learning_rate": 3.0419580419580425e-05,
+      "loss": -0.0129,
+      "num_tokens": 860422.0,
+      "reward": 1.6449967622756958,
+      "reward_std": 0.7163441777229309,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 0.7142857313156128,
+      "rewards/check_winston_local_func": 0.5378537178039551,
+      "step": 87
+    },
+    {
+      "clip_ratio": 0.007247431669384241,
+      "epoch": 0.12307692307692308,
+      "grad_norm": 0.2167770735292921,
+      "kl": 0.61328125,
+      "learning_rate": 3.0769230769230774e-05,
+      "loss": -0.0168,
+      "step": 88
+    },
+    {
+      "clip_ratio": 0.0027036736719310284,
+      "completion_length": 108.03572082519531,
+      "epoch": 0.12447552447552447,
+      "grad_norm": 0.12604084880729077,
+      "kl": 0.291015625,
+      "learning_rate": 3.111888111888112e-05,
+      "loss": -0.0039,
+      "num_tokens": 881364.0,
+      "reward": 1.3799673318862915,
+      "reward_std": 0.6881024837493896,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 0.5714285969734192,
+      "rewards/check_winston_local_func": 0.4156815707683563,
+      "step": 89
+    },
+    {
+      "clip_ratio": 0.005238416139036417,
+      "epoch": 0.1258741258741259,
+      "grad_norm": 0.12103303780018655,
+      "kl": 0.30859375,
+      "learning_rate": 3.146853146853147e-05,
+      "loss": -0.005,
+      "step": 90
+    },
+    {
+      "clip_ratio": 0.003768512513488531,
+      "completion_length": 122.80357360839844,
+      "epoch": 0.12727272727272726,
+      "grad_norm": 0.18423815092289067,
+      "kl": 0.67578125,
+      "learning_rate": 3.181818181818182e-05,
+      "loss": -0.0171,
+      "num_tokens": 904487.0,
+      "reward": 1.6101170778274536,
+      "reward_std": 0.6241902709007263,
+      "rewards/check_gptzero_func": 0.3392857015132904,
+      "rewards/check_perplexity_diff_func": 0.7857142686843872,
+      "rewards/check_winston_local_func": 0.48511695861816406,
+      "step": 91
+    },
+    {
+      "clip_ratio": 0.005407070741057396,
+      "epoch": 0.12867132867132866,
+      "grad_norm": 0.15457101650134836,
+      "kl": 0.66796875,
+      "learning_rate": 3.216783216783217e-05,
+      "loss": -0.0182,
+      "step": 92
+    },
+    {
+      "clip_ratio": 0.002042532665655017,
+      "completion_length": 108.17857360839844,
+      "epoch": 0.13006993006993006,
+      "grad_norm": 0.19588538280583437,
+      "kl": 1.75,
+      "learning_rate": 3.251748251748252e-05,
+      "loss": -0.0023,
+      "num_tokens": 925633.0,
+      "reward": 1.3932582139968872,
+      "reward_std": 0.5087055563926697,
+      "rewards/check_gptzero_func": 0.4107142984867096,
+      "rewards/check_perplexity_diff_func": 0.4642857015132904,
+      "rewards/check_winston_local_func": 0.5182580351829529,
+      "step": 93
+    },
+    {
+      "clip_ratio": 0.00876330491155386,
+      "epoch": 0.13146853146853146,
+      "grad_norm": 0.1506256243597478,
+      "kl": 0.80078125,
+      "learning_rate": 3.2867132867132866e-05,
+      "loss": -0.0035,
+      "step": 94
+    },
+    {
+      "clip_ratio": 0.003118924330919981,
+      "completion_length": 77.4464340209961,
+      "epoch": 0.13286713286713286,
+      "grad_norm": 0.34784170699852607,
+      "kl": 0.72265625,
+      "learning_rate": 3.321678321678322e-05,
+      "loss": -0.0134,
+      "num_tokens": 942892.0,
+      "reward": 1.832355260848999,
+      "reward_std": 1.064102292060852,
+      "rewards/check_gptzero_func": 0.4821428656578064,
+      "rewards/check_perplexity_diff_func": 0.8214285969734192,
+      "rewards/check_winston_local_func": 0.5287837982177734,
+      "step": 95
+    },
+    {
+      "clip_ratio": 0.00509096821770072,
+      "epoch": 0.13426573426573427,
+      "grad_norm": 0.25968424971577175,
+      "kl": 0.96875,
+      "learning_rate": 3.356643356643357e-05,
+      "loss": -0.0176,
+      "step": 96
+    },
+    {
+      "clip_ratio": 0.0037272910121828318,
+      "completion_length": 95.26786041259766,
+      "epoch": 0.13566433566433567,
+      "grad_norm": 0.16914051199341906,
+      "kl": 0.349609375,
+      "learning_rate": 3.391608391608392e-05,
+      "loss": 0.0037,
+      "num_tokens": 962317.0,
+      "reward": 1.805726170539856,
+      "reward_std": 0.831261396408081,
+      "rewards/check_gptzero_func": 0.4107142984867096,
+      "rewards/check_perplexity_diff_func": 0.9285714030265808,
+      "rewards/check_winston_local_func": 0.46644020080566406,
+      "step": 97
+    },
+    {
+      "clip_ratio": 0.0034601751249283552,
+      "epoch": 0.13706293706293707,
+      "grad_norm": 0.16538870434198547,
+      "kl": 0.388671875,
+      "learning_rate": 3.4265734265734265e-05,
+      "loss": 0.0019,
+      "step": 98
+    },
+    {
+      "clip_ratio": 0.004487877711653709,
+      "completion_length": 111.8214340209961,
+      "epoch": 0.13846153846153847,
+      "grad_norm": 0.12415078751808917,
+      "kl": 0.384765625,
+      "learning_rate": 3.461538461538462e-05,
+      "loss": -0.001,
+      "num_tokens": 983513.0,
+      "reward": 1.2062333822250366,
+      "reward_std": 0.591974139213562,
+      "rewards/check_gptzero_func": 0.3214285671710968,
+      "rewards/check_perplexity_diff_func": 0.3571428656578064,
+      "rewards/check_winston_local_func": 0.5276618599891663,
+      "step": 99
+    },
+    {
+      "clip_ratio": 0.00284260674379766,
+      "epoch": 0.13986013986013987,
+      "grad_norm": 0.1197068572237121,
+      "kl": 0.376953125,
+      "learning_rate": 3.4965034965034965e-05,
+      "loss": -0.0017,
+      "step": 100
+    },
+    {
+      "clip_ratio": 0.0023206709884107113,
+      "completion_length": 101.08928680419922,
+      "epoch": 0.14125874125874127,
+      "grad_norm": 0.16400691003183243,
+      "kl": 0.30078125,
+      "learning_rate": 3.531468531468531e-05,
+      "loss": 0.0033,
+      "num_tokens": 1003878.0,
+      "reward": 1.631854772567749,
+      "reward_std": 0.5223387479782104,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 0.7142857313156128,
+      "rewards/check_winston_local_func": 0.5247119665145874,
+      "step": 101
+    },
+    {
+      "clip_ratio": 0.0016618981026113033,
+      "epoch": 0.14265734265734265,
+      "grad_norm": 0.1573929247787687,
+      "kl": 0.326171875,
+      "learning_rate": 3.566433566433567e-05,
+      "loss": 0.0012,
+      "step": 102
+    },
+    {
+      "clip_ratio": 0.0022201864048838615,
+      "completion_length": 124.46429443359375,
+      "epoch": 0.14405594405594405,
+      "grad_norm": 0.1353861256020412,
+      "kl": 0.3828125,
+      "learning_rate": 3.601398601398602e-05,
+      "loss": 0.005,
+      "num_tokens": 1027066.0,
+      "reward": 1.6920486688613892,
+      "reward_std": 0.48560601472854614,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 0.75,
+      "rewards/check_winston_local_func": 0.5491914749145508,
+      "step": 103
+    },
+    {
+      "clip_ratio": 0.0029075751081109047,
+      "epoch": 0.14545454545454545,
+      "grad_norm": 0.13328591024866146,
+      "kl": 0.39453125,
+      "learning_rate": 3.6363636363636364e-05,
+      "loss": 0.0034,
+      "step": 104
+    },
+    {
+      "clip_ratio": 0.0019614642951637506,
+      "completion_length": 85.26786041259766,
+      "epoch": 0.14685314685314685,
+      "grad_norm": 0.19674478002774554,
+      "kl": 0.333984375,
+      "learning_rate": 3.671328671328672e-05,
+      "loss": -0.0188,
+      "num_tokens": 1045567.0,
+      "reward": 1.6822034120559692,
+      "reward_std": 0.7559517621994019,
+      "rewards/check_gptzero_func": 0.5357142686843872,
+      "rewards/check_perplexity_diff_func": 0.6071428656578064,
+      "rewards/check_winston_local_func": 0.5393460988998413,
+      "step": 105
+    },
+    {
+      "clip_ratio": 0.004534203093498945,
+      "epoch": 0.14825174825174825,
+      "grad_norm": 0.16787611299305724,
+      "kl": 0.3671875,
+      "learning_rate": 3.7062937062937064e-05,
+      "loss": -0.0217,
+      "step": 106
+    },
+    {
+      "clip_ratio": 0.003009920008480549,
+      "completion_length": 106.55357360839844,
+      "epoch": 0.14965034965034965,
+      "grad_norm": 0.19751132540015684,
+      "kl": 0.455078125,
+      "learning_rate": 3.741258741258741e-05,
+      "loss": 0.0111,
+      "num_tokens": 1066466.0,
+      "reward": 2.178619861602783,
+      "reward_std": 0.7016831040382385,
+      "rewards/check_gptzero_func": 0.6071428656578064,
+      "rewards/check_perplexity_diff_func": 0.8928571343421936,
+      "rewards/check_winston_local_func": 0.6786197423934937,
+      "step": 107
+    },
+    {
+      "clip_ratio": 0.004473494831472635,
+      "epoch": 0.15104895104895105,
+      "grad_norm": 0.1431470008572649,
+      "kl": 0.4375,
+      "learning_rate": 3.776223776223776e-05,
+      "loss": 0.0095,
+      "step": 108
+    },
+    {
+      "clip_ratio": 0.0026541000697761774,
+      "completion_length": 100.6964340209961,
+      "epoch": 0.15244755244755245,
+      "grad_norm": 0.33135604940901486,
+      "kl": 0.310546875,
+      "learning_rate": 3.811188811188811e-05,
+      "loss": 0.0138,
+      "num_tokens": 1086751.0,
+      "reward": 1.6315226554870605,
+      "reward_std": 0.7806248068809509,
+      "rewards/check_gptzero_func": 0.4642857015132904,
+      "rewards/check_perplexity_diff_func": 0.6071428656578064,
+      "rewards/check_winston_local_func": 0.5600939393043518,
+      "step": 109
+    },
+    {
+      "clip_ratio": 0.0028892713598906994,
+      "epoch": 0.15384615384615385,
+      "grad_norm": 0.24217207616877234,
+      "kl": 0.361328125,
+      "learning_rate": 3.846153846153846e-05,
+      "loss": 0.0074,
+      "step": 110
+    },
+    {
+      "clip_ratio": 0.0021747422870248556,
+      "completion_length": 84.76786041259766,
+      "epoch": 0.15524475524475526,
+      "grad_norm": 0.15803622964815373,
+      "kl": 0.66796875,
+      "learning_rate": 3.8811188811188816e-05,
+      "loss": -0.0045,
+      "num_tokens": 1104518.0,
+      "reward": 1.6692500114440918,
+      "reward_std": 0.5965060591697693,
+      "rewards/check_gptzero_func": 0.5535714030265808,
+      "rewards/check_perplexity_diff_func": 0.6071428656578064,
+      "rewards/check_winston_local_func": 0.5085356831550598,
+      "step": 111
+    },
+    {
+      "clip_ratio": 0.004626331850886345,
+      "epoch": 0.15664335664335666,
+      "grad_norm": 0.14920406573741435,
+      "kl": 0.7578125,
+      "learning_rate": 3.916083916083916e-05,
+      "loss": -0.0064,
+      "step": 112
+    },
+    {
+      "clip_ratio": 0.002694516209885478,
+      "completion_length": 92.17857360839844,
+      "epoch": 0.15804195804195803,
+      "grad_norm": 0.161427063551978,
+      "kl": 0.361328125,
+      "learning_rate": 3.9510489510489516e-05,
+      "loss": 0.0058,
+      "num_tokens": 1123466.0,
+      "reward": 1.353637456893921,
+      "reward_std": 0.5352396965026855,
+      "rewards/check_gptzero_func": 0.4821428656578064,
+      "rewards/check_perplexity_diff_func": 0.3571428656578064,
+      "rewards/check_winston_local_func": 0.5143517255783081,
+      "step": 113
+    },
+    {
+      "clip_ratio": 0.00391918933019042,
+      "epoch": 0.15944055944055943,
+      "grad_norm": 0.14561571783883442,
+      "kl": 0.39453125,
+      "learning_rate": 3.986013986013986e-05,
+      "loss": 0.0036,
+      "step": 114
+    },
+    {
+      "clip_ratio": 0.001278667594306171,
+      "completion_length": 107.12500762939453,
+      "epoch": 0.16083916083916083,
+      "grad_norm": 0.1805624176716803,
+      "kl": 0.3046875,
+      "learning_rate": 4.020979020979021e-05,
+      "loss": -0.001,
+      "num_tokens": 1144619.0,
+      "reward": 1.6282455921173096,
+      "reward_std": 0.7325619459152222,
+      "rewards/check_gptzero_func": 0.4464285671710968,
+      "rewards/check_perplexity_diff_func": 0.5357142686843872,
+      "rewards/check_winston_local_func": 0.6461027264595032,
+      "step": 115
+    },
+    {
+      "clip_ratio": 0.0017050534952431917,
+      "epoch": 0.16223776223776223,
+      "grad_norm": 0.16942268791902212,
+      "kl": 0.337890625,
+      "learning_rate": 4.055944055944056e-05,
+      "loss": -0.0027,
+      "step": 116
+    },
+    {
+      "clip_ratio": 0.0015462420415133238,
+      "completion_length": 114.8214340209961,
+      "epoch": 0.16363636363636364,
+      "grad_norm": 0.1777133538888581,
+      "kl": 0.322265625,
+      "learning_rate": 4.0909090909090915e-05,
+      "loss": -0.0028,
+      "num_tokens": 1166185.0,
+      "reward": 1.566156268119812,
+      "reward_std": 0.38663557171821594,
+      "rewards/check_gptzero_func": 0.4642857015132904,
+      "rewards/check_perplexity_diff_func": 0.5,
+      "rewards/check_winston_local_func": 0.6018705368041992,
+      "step": 117
+    },
+    {
+      "clip_ratio": 0.0030793240293860435,
+      "epoch": 0.16503496503496504,
+      "grad_norm": 0.15084856022622706,
+      "kl": 0.2890625,
+      "learning_rate": 4.125874125874126e-05,
+      "loss": -0.0048,
+      "step": 118
+    },
+    {
+      "clip_ratio": 0.0018360918620601296,
+      "completion_length": 100.25000762939453,
+      "epoch": 0.16643356643356644,
+      "grad_norm": 0.10943003413749244,
+      "kl": 0.83984375,
+      "learning_rate": 4.1608391608391614e-05,
+      "loss": 0.0043,
+      "num_tokens": 1186317.0,
+      "reward": 1.7977957725524902,
+      "reward_std": 0.5172301530838013,
+      "rewards/check_gptzero_func": 0.5714285969734192,
+      "rewards/check_perplexity_diff_func": 0.7142857313156128,
+      "rewards/check_winston_local_func": 0.5120813250541687,
+      "step": 119
+    },
+    {
+      "clip_ratio": 0.0025589358992874622,
+      "epoch": 0.16783216783216784,
+      "grad_norm": 0.10431032316313237,
+      "kl": 0.68359375,
+      "learning_rate": 4.195804195804196e-05,
+      "loss": 0.0029,
+      "step": 120
+    },
+    {
+      "clip_ratio": 0.002800893737003207,
+      "completion_length": 120.91072082519531,
+      "epoch": 0.16923076923076924,
+      "grad_norm": 0.1567960287886374,
+      "kl": 0.37890625,
+      "learning_rate": 4.230769230769231e-05,
+      "loss": -0.0016,
+      "num_tokens": 1208530.0,
+      "reward": 1.5369055271148682,
+      "reward_std": 0.7073518633842468,
+      "rewards/check_gptzero_func": 0.3214285671710968,
+      "rewards/check_perplexity_diff_func": 0.6785714030265808,
+      "rewards/check_winston_local_func": 0.5369054079055786,
+      "step": 121
+    },
+    {
+      "clip_ratio": 0.0026409339625388384,
+      "epoch": 0.17062937062937064,
+      "grad_norm": 0.15013150065673506,
+      "kl": 0.37890625,
+      "learning_rate": 4.265734265734266e-05,
+      "loss": -0.0035,
+      "step": 122
+    },
+    {
+      "clip_ratio": 0.002211854327470064,
+      "completion_length": 110.66072082519531,
+      "epoch": 0.17202797202797201,
+      "grad_norm": 0.17274409069851862,
+      "kl": 0.380859375,
+      "learning_rate": 4.300699300699301e-05,
+      "loss": -0.0062,
+      "num_tokens": 1229171.0,
+      "reward": 1.6140996217727661,
+      "reward_std": 0.7221139669418335,
+      "rewards/check_gptzero_func": 0.4642857015132904,
+      "rewards/check_perplexity_diff_func": 0.5357142686843872,
+      "rewards/check_winston_local_func": 0.6140995025634766,
+      "step": 123
+    },
+    {
+      "clip_ratio": 0.005159804597496986,
+      "epoch": 0.17342657342657342,
+      "grad_norm": 0.13306053466541726,
+      "kl": 0.443359375,
+      "learning_rate": 4.335664335664335e-05,
+      "loss": -0.009,
+      "step": 124
+    },
+    {
+      "clip_ratio": 0.0031610180158168077,
+      "completion_length": 104.30357360839844,
+      "epoch": 0.17482517482517482,
+      "grad_norm": 0.19261275892706695,
+      "kl": 0.30078125,
+      "learning_rate": 4.370629370629371e-05,
+      "loss": -0.0129,
+      "num_tokens": 1249350.0,
+      "reward": 1.9047484397888184,
+      "reward_std": 0.6767317652702332,
+      "rewards/check_gptzero_func": 0.6071428656578064,
+      "rewards/check_perplexity_diff_func": 0.6071428656578064,
+      "rewards/check_winston_local_func": 0.6904626488685608,
+      "step": 125
+    },
+    {
+      "clip_ratio": 0.004231306724250317,
+      "epoch": 0.17622377622377622,
+      "grad_norm": 0.1752737652506695,
+      "kl": 0.33203125,
+      "learning_rate": 4.405594405594406e-05,
+      "loss": -0.0162,
+      "step": 126
+    },
+    {
+      "clip_ratio": 0.003981932066380978,
+      "completion_length": 109.53572082519531,
+      "epoch": 0.17762237762237762,
+      "grad_norm": 0.1423250496935692,
+      "kl": 0.48828125,
+      "learning_rate": 4.4405594405594406e-05,
+      "loss": -0.0102,
+      "num_tokens": 1269848.0,
+      "reward": 1.7591207027435303,
+      "reward_std": 0.5321380496025085,
+      "rewards/check_gptzero_func": 0.5,
+      "rewards/check_perplexity_diff_func": 0.6785714030265808,
+      "rewards/check_winston_local_func": 0.5805493593215942,
+      "step": 127
+    },
+    {
+      "clip_ratio": 0.005287667270749807,
+      "epoch": 0.17902097902097902,
+      "grad_norm": 0.13255848380623775,
+      "kl": 0.498046875,
+      "learning_rate": 4.475524475524476e-05,
+      "loss": -0.0121,
+      "step": 128
+    },
+    {
+      "clip_ratio": 0.0025025398936122656,
+      "completion_length": 105.30357360839844,
+      "epoch": 0.18041958041958042,
+      "grad_norm": 0.1177341259986552,
+      "kl": 0.283203125,
+      "learning_rate": 4.5104895104895105e-05,
+      "loss": -0.0016,
+      "num_tokens": 1290033.0,
+      "reward": 1.7174798250198364,
+      "reward_std": 0.5096268057823181,
+      "rewards/check_gptzero_func": 0.5357142686843872,
+      "rewards/check_perplexity_diff_func": 0.5357142686843872,
+      "rewards/check_winston_local_func": 0.6460510492324829,
+      "step": 129
+    },
+    {
+      "clip_ratio": 0.0033291254658252,
+      "epoch": 0.18181818181818182,
+      "grad_norm": 0.11141937591016414,
+      "kl": 0.294921875,
+      "learning_rate": 4.545454545454546e-05,
+      "loss": -0.0032,
+      "step": 130
+    },
+    {
+      "clip_ratio": 0.0024425899609923363,
+      "completion_length": 96.30357360839844,
+      "epoch": 0.18321678321678322,
+      "grad_norm": 0.23593816935965856,
+      "kl": 0.341796875,
+      "learning_rate": 4.5804195804195805e-05,
+      "loss": -0.006,
+      "num_tokens": 1309356.0,
+      "reward": 1.7568891048431396,
+      "reward_std": 0.7255779504776001,
+      "rewards/check_gptzero_func": 0.5892857313156128,
+      "rewards/check_perplexity_diff_func": 0.5357142686843872,
+      "rewards/check_winston_local_func": 0.6318890452384949,
+      "step": 131
+    },
+    {
+      "clip_ratio": 0.004331678152084351,
+      "epoch": 0.18461538461538463,
+      "grad_norm": 0.2135627059179365,
+      "kl": 0.349609375,
+      "learning_rate": 4.615384615384616e-05,
+      "loss": -0.0089,
+      "step": 132
+    },
+    {
+      "clip_ratio": 0.0028827630449086428,
+      "completion_length": 111.6964340209961,
+      "epoch": 0.18601398601398603,
+      "grad_norm": 0.1622976058137894,
+      "kl": 0.369140625,
+      "learning_rate": 4.6503496503496505e-05,
+      "loss": 0.014,
+      "num_tokens": 1330017.0,
+      "reward": 1.6692737340927124,
+      "reward_std": 0.628279983997345,
+      "rewards/check_gptzero_func": 0.4464285671710968,
+      "rewards/check_perplexity_diff_func": 0.6428571343421936,
+      "rewards/check_winston_local_func": 0.5799878835678101,
+      "step": 133
+    },
+    {
+      "clip_ratio": 0.004608546383678913,
+      "epoch": 0.1874125874125874,
+      "grad_norm": 0.16503328062657166,
+      "kl": 0.4140625,
+      "learning_rate": 4.685314685314686e-05,
+      "loss": 0.0121,
+      "step": 134
+    },
+    {
+      "clip_ratio": 0.0027426625601947308,
+      "completion_length": 107.58928680419922,
+      "epoch": 0.1888111888111888,
+      "grad_norm": 0.27926954484170907,
+      "kl": 0.5859375,
+      "learning_rate": 4.7202797202797204e-05,
+      "loss": 0.0139,
+      "num_tokens": 1350164.0,
+      "reward": 1.9428951740264893,
+      "reward_std": 0.6565932035446167,
+      "rewards/check_gptzero_func": 0.6785714030265808,
+      "rewards/check_perplexity_diff_func": 0.5714285969734192,
+      "rewards/check_winston_local_func": 0.6928950548171997,
+      "step": 135
+    },
+    {
+      "clip_ratio": 0.00439854059368372,
+      "epoch": 0.1902097902097902,
+      "grad_norm": 0.1925514269864402,
+      "kl": 0.5546875,
+      "learning_rate": 4.755244755244756e-05,
+      "loss": 0.0074,
+      "step": 136
+    },
+    {
+      "clip_ratio": 0.0018738384824246168,
+      "completion_length": 116.4464340209961,
+      "epoch": 0.1916083916083916,
+      "grad_norm": 0.14918417137816656,
+      "kl": 0.58984375,
+      "learning_rate": 4.7902097902097904e-05,
+      "loss": 0.0037,
+      "num_tokens": 1371507.0,
+      "reward": 2.0137577056884766,
+      "reward_std": 0.5378711819648743,
+      "rewards/check_gptzero_func": 0.6071428656578064,
+      "rewards/check_perplexity_diff_func": 0.6785714030265808,
+      "rewards/check_winston_local_func": 0.7280434370040894,
+      "step": 137
+    },
+    {
+      "clip_ratio": 0.003529219189658761,
+      "epoch": 0.193006993006993,
+      "grad_norm": 0.1301060608041501,
+      "kl": 0.6171875,
+      "learning_rate": 4.825174825174825e-05,
+      "loss": 0.0016,
+      "step": 138
+    },
+    {
+      "clip_ratio": 0.0011985624441877007,
+      "completion_length": 112.6964340209961,
+      "epoch": 0.1944055944055944,
+      "grad_norm": 0.13034934146859325,
+      "kl": 0.322265625,
+      "learning_rate": 4.86013986013986e-05,
+      "loss": -0.0024,
+      "num_tokens": 1392414.0,
+      "reward": 1.9355616569519043,
+      "reward_std": 0.4726971983909607,
+      "rewards/check_gptzero_func": 0.5892857313156128,
+      "rewards/check_perplexity_diff_func": 0.75,
+      "rewards/check_winston_local_func": 0.596275806427002,
+      "step": 139
+    },
+    {
+      "clip_ratio": 0.0032699662260711193,
+      "epoch": 0.1958041958041958,
+      "grad_norm": 0.12091032490165694,
+      "kl": 0.318359375,
+      "learning_rate": 4.8951048951048956e-05,
+      "loss": -0.0039,
+      "step": 140
+    },
+    {
+      "clip_ratio": 0.0013384540798142552,
+      "completion_length": 132.05357360839844,
+      "epoch": 0.1972027972027972,
+      "grad_norm": 0.12728573210952063,
+      "kl": 0.59765625,
+      "learning_rate": 4.93006993006993e-05,
+      "loss": 0.0045,
+      "num_tokens": 1416177.0,
+      "reward": 1.7924094200134277,
+      "reward_std": 0.7202263474464417,
+      "rewards/check_gptzero_func": 0.5714285969734192,
+      "rewards/check_perplexity_diff_func": 0.7142857313156128,
+      "rewards/check_winston_local_func": 0.5066950917243958,
+      "step": 141
+    },
+    {
+      "clip_ratio": 0.0017050639726221561,
+      "epoch": 0.1986013986013986,
+      "grad_norm": 0.12231361158637834,
+      "kl": 0.65234375,
+      "learning_rate": 4.9650349650349656e-05,
+      "loss": 0.0025,
+      "step": 142
+    },
+    {
+      "clip_ratio": 0.0015860958956182003,
+      "completion_length": 99.73214721679688,
+      "epoch": 0.2,
+      "grad_norm": 0.1853946177277763,
+      "kl": 0.58984375,
+      "learning_rate": 5e-05,
+      "loss": 0.0021,
+      "num_tokens": 1435732.0,
+      "reward": 2.0557591915130615,
+      "reward_std": 0.503829836845398,
+      "rewards/check_gptzero_func": 0.6964285969734192,
+      "rewards/check_perplexity_diff_func": 0.75,
+      "rewards/check_winston_local_func": 0.6093305945396423,
+      "step": 143
+    },
+    {
+      "clip_ratio": 0.002123563550412655,
+      "epoch": 0.2013986013986014,
+      "grad_norm": 0.13470126319919157,
+      "kl": 0.423828125,
+      "learning_rate": 5.0349650349650356e-05,
+      "loss": -0.0009,
+      "step": 144
+    },
+    {
+      "clip_ratio": 0.0018757216166704893,
+      "completion_length": 95.73214721679688,
+      "epoch": 0.20279720279720279,
+      "grad_norm": 0.14930190515465938,
+      "kl": 0.388671875,
+      "learning_rate": 5.06993006993007e-05,
+      "loss": -0.0026,
+      "num_tokens": 1454705.0,
+      "reward": 2.0560998916625977,
+      "reward_std": 0.5541732311248779,
+      "rewards/check_gptzero_func": 0.6964285969734192,
+      "rewards/check_perplexity_diff_func": 0.5714285969734192,
+      "rewards/check_winston_local_func": 0.788242518901825,
+      "step": 145
+    },
+    {
+      "clip_ratio": 0.003540371311828494,
+      "epoch": 0.2041958041958042,
+      "grad_norm": 0.1304251509197979,
+      "kl": 0.41796875,
+      "learning_rate": 5.1048951048951055e-05,
+      "loss": -0.0046,
+      "step": 146
+    },
+    {
+      "clip_ratio": 0.0031265008728951216,
+      "completion_length": 80.9464340209961,
+      "epoch": 0.2055944055944056,
+      "grad_norm": 0.2087929657840581,
+      "kl": 1.1328125,
+      "learning_rate": 5.1398601398601395e-05,
+      "loss": 0.0026,
+      "num_tokens": 1472182.0,
+      "reward": 1.9076076745986938,
+      "reward_std": 0.46633273363113403,
+      "rewards/check_gptzero_func": 0.6428571343421936,
+      "rewards/check_perplexity_diff_func": 0.5,
+      "rewards/check_winston_local_func": 0.7647504210472107,
+      "step": 147
+    },
+    {
+      "clip_ratio": 0.0029583375435322523,
+      "epoch": 0.206993006993007,
+      "grad_norm": 0.178151823438605,
+      "kl": 0.66015625,
+      "learning_rate": 5.1748251748251755e-05,
+      "loss": -0.0014,
+      "step": 148
+    },
+    {
+      "clip_ratio": 0.0020277961157262325,
+      "completion_length": 126.58929443359375,
+      "epoch": 0.2083916083916084,
+      "grad_norm": 0.13736153026044381,
+      "kl": 0.373046875,
+      "learning_rate": 5.2097902097902094e-05,
+      "loss": 0.0127,
+      "num_tokens": 1494581.0,
+      "reward": 1.870171308517456,
+      "reward_std": 0.4412032663822174,
+      "rewards/check_gptzero_func": 0.4464285671710968,
+      "rewards/check_perplexity_diff_func": 0.8214285969734192,
+      "rewards/check_winston_local_func": 0.6023141741752625,
+      "step": 149
+    },
+    {
+      "clip_ratio": 0.0022643795236945152,
+      "epoch": 0.2097902097902098,
+      "grad_norm": 0.12942677801199462,
+      "kl": 0.40625,
+      "learning_rate": 5.244755244755245e-05,
+      "loss": 0.0106,
+      "step": 150
+    },
+    {
+      "clip_ratio": 0.001886948710307479,
+      "completion_length": 95.14286041259766,
+      "epoch": 0.2111888111888112,
+      "grad_norm": 0.15555312890089615,
+      "kl": 0.62890625,
+      "learning_rate": 5.279720279720281e-05,
+      "loss": 0.0013,
+      "num_tokens": 1513403.0,
+      "reward": 1.9735006093978882,
+      "reward_std": 0.5218394994735718,
+      "rewards/check_gptzero_func": 0.5535714030265808,
+      "rewards/check_perplexity_diff_func": 0.6071428656578064,
+      "rewards/check_winston_local_func": 0.8127861022949219,
+      "step": 151
+    },
+    {
+      "clip_ratio": 0.003476199461147189,
+      "epoch": 0.2125874125874126,
+      "grad_norm": 0.13833602394582134,
+      "kl": 0.5390625,
+      "learning_rate": 5.314685314685315e-05,
+      "loss": -0.0015,
+      "step": 152
+    },
+    {
+      "clip_ratio": 0.0029247915372252464,
+      "completion_length": 104.1964340209961,
+      "epoch": 0.213986013986014,
+      "grad_norm": 0.18332277821774023,
+      "kl": 0.451171875,
+      "learning_rate": 5.34965034965035e-05,
+      "loss": -0.0047,
+      "num_tokens": 1533154.0,
+      "reward": 1.977178931236267,
+      "reward_std": 0.5758928656578064,
+      "rewards/check_gptzero_func": 0.6785714030265808,
+      "rewards/check_perplexity_diff_func": 0.6785714030265808,
+      "rewards/check_winston_local_func": 0.6200361251831055,
+      "step": 153
+    },
+    {
+      "clip_ratio": 0.002894646255299449,
+      "epoch": 0.2153846153846154,
+      "grad_norm": 0.12417162413759981,
+      "kl": 0.48828125,
+      "learning_rate": 5.384615384615385e-05,
+      "loss": -0.0082,
+      "step": 154
+    },
+    {
+      "clip_ratio": 0.002533347113057971,
+      "completion_length": 106.6964340209961,
+      "epoch": 0.21678321678321677,
+      "grad_norm": 0.2165608004438418,
+      "kl": 0.4140625,
+      "learning_rate": 5.41958041958042e-05,
+      "loss": 0.0367,
+      "num_tokens": 1554005.0,
+      "reward": 2.254620313644409,
+      "reward_std": 0.6860859394073486,
+      "rewards/check_gptzero_func": 0.7857142686843872,
+      "rewards/check_perplexity_diff_func": 0.7142857313156128,
+      "rewards/check_winston_local_func": 0.7546200752258301,
+      "step": 155
+    },
+    {
+      "clip_ratio": 0.005156368017196655,
+      "epoch": 0.21818181818181817,
+      "grad_norm": 0.16185524517286934,
+      "kl": 0.4609375,
+      "learning_rate": 5.4545454545454546e-05,
+      "loss": 0.0322,
+      "step": 156
+    },
+    {
+      "clip_ratio": 0.002192203886806965,
+      "completion_length": 129.55357360839844,
+      "epoch": 0.21958041958041957,
+      "grad_norm": 0.13686573639431518,
+      "kl": 0.421875,
+      "learning_rate": 5.48951048951049e-05,
+      "loss": 0.0095,
+      "num_tokens": 1576988.0,
+      "reward": 2.202296495437622,
+      "reward_std": 0.4840867817401886,
+      "rewards/check_gptzero_func": 0.6428571343421936,
+      "rewards/check_perplexity_diff_func": 0.8928571343421936,
+      "rewards/check_winston_local_func": 0.6665821075439453,
+      "step": 157
+    },
+    {
+      "clip_ratio": 0.0037054666317999363,
+      "epoch": 0.22097902097902097,
+      "grad_norm": 0.12040805378810834,
+      "kl": 0.45703125,
+      "learning_rate": 5.524475524475524e-05,
+      "loss": 0.0077,
+      "step": 158
+    },
+    {
+      "clip_ratio": 0.002324200002476573,
+      "completion_length": 141.10714721679688,
+      "epoch": 0.22237762237762237,
+      "grad_norm": 0.1722210877266639,
+      "kl": 0.9765625,
+      "learning_rate": 5.55944055944056e-05,
+      "loss": -0.0049,
+      "num_tokens": 1600826.0,
+      "reward": 2.0245378017425537,
+      "reward_std": 0.45877185463905334,
+      "rewards/check_gptzero_func": 0.5892857313156128,
+      "rewards/check_perplexity_diff_func": 0.75,
+      "rewards/check_winston_local_func": 0.6852518916130066,
+      "step": 159
+    },
+    {
+      "clip_ratio": 0.003470373572781682,
+      "epoch": 0.22377622377622378,
+      "grad_norm": 0.13356363193434526,
+      "kl": 0.65234375,
+      "learning_rate": 5.594405594405595e-05,
+      "loss": -0.0085,
+      "step": 160
+    },
+    {
+      "clip_ratio": 0.0026535126380622387,
+      "completion_length": 109.76786041259766,
+      "epoch": 0.22517482517482518,
+      "grad_norm": 0.409404915148595,
+      "kl": 0.515625,
+      "learning_rate": 5.629370629370629e-05,
+      "loss": -0.0051,
+      "num_tokens": 1621103.0,
+      "reward": 2.0040969848632812,
+      "reward_std": 0.5942196249961853,
+      "rewards/check_gptzero_func": 0.5178571343421936,
+      "rewards/check_perplexity_diff_func": 0.7857142686843872,
+      "rewards/check_winston_local_func": 0.7005256414413452,
+      "step": 161
+    },
+    {
+      "clip_ratio": 0.0058495416305959225,
+      "epoch": 0.22657342657342658,
+      "grad_norm": 0.16376328147020022,
+      "kl": 0.5078125,
+      "learning_rate": 5.664335664335665e-05,
+      "loss": -0.0129,
+      "step": 162
+    },
+    {
+      "clip_ratio": 0.002924378262832761,
+      "completion_length": 114.3214340209961,
+      "epoch": 0.22797202797202798,
+      "grad_norm": 0.17698045766822795,
+      "kl": 0.62109375,
+      "learning_rate": 5.699300699300699e-05,
+      "loss": -0.0053,
+      "num_tokens": 1642629.0,
+      "reward": 2.0913121700286865,
+      "reward_std": 0.533393144607544,
+      "rewards/check_gptzero_func": 0.6785714030265808,
+      "rewards/check_perplexity_diff_func": 0.6785714030265808,
+      "rewards/check_winston_local_func": 0.7341693639755249,
+      "step": 163
+    },
+    {
+      "clip_ratio": 0.006533453240990639,
+      "epoch": 0.22937062937062938,
+      "grad_norm": 0.15566114152311913,
+      "kl": 0.66796875,
+      "learning_rate": 5.7342657342657345e-05,
+      "loss": -0.0085,
+      "step": 164
+    },
+    {
+      "clip_ratio": 0.002033352619037032,
+      "completion_length": 110.48214721679688,
+      "epoch": 0.23076923076923078,
+      "grad_norm": 0.19340316318251602,
+      "kl": 0.4453125,
+      "learning_rate": 5.769230769230769e-05,
+      "loss": 0.0054,
+      "num_tokens": 1663000.0,
+      "reward": 1.809409499168396,
+      "reward_std": 0.5209853053092957,
+      "rewards/check_gptzero_func": 0.6428571343421936,
+      "rewards/check_perplexity_diff_func": 0.3571428656578064,
+      "rewards/check_winston_local_func": 0.809409499168396,
+      "step": 165
+    },
+    {
+      "clip_ratio": 0.005329853855073452,
+      "epoch": 0.23216783216783216,
+      "grad_norm": 0.15299329983895626,
+      "kl": 0.55078125,
+      "learning_rate": 5.8041958041958044e-05,
+      "loss": 0.0018,
+      "step": 166
+    },
+    {
+      "clip_ratio": 0.0025733087677508593,
+      "completion_length": 95.73214721679688,
+      "epoch": 0.23356643356643356,
+      "grad_norm": 0.16102285972903455,
+      "kl": 0.458984375,
+      "learning_rate": 5.83916083916084e-05,
+      "loss": -0.0008,
+      "num_tokens": 1682123.0,
+      "reward": 1.9705681800842285,
+      "reward_std": 0.5465096235275269,
+      "rewards/check_gptzero_func": 0.6428571343421936,
+      "rewards/check_perplexity_diff_func": 0.6428571343421936,
+      "rewards/check_winston_local_func": 0.6848538517951965,
+      "step": 167
+    },
+    {
+      "clip_ratio": 0.0034487086813896894,
+      "epoch": 0.23496503496503496,
+      "grad_norm": 0.12896214559568192,
+      "kl": 0.474609375,
+      "learning_rate": 5.8741258741258744e-05,
+      "loss": -0.0039,
+      "step": 168
+    },
+    {
+      "clip_ratio": 0.0012721805833280087,
+      "completion_length": 72.125,
+      "epoch": 0.23636363636363636,
+      "grad_norm": 0.19976827264055144,
+      "kl": 0.5703125,
+      "learning_rate": 5.90909090909091e-05,
+      "loss": -0.0046,
+      "num_tokens": 1698078.0,
+      "reward": 1.8698246479034424,
+      "reward_std": 0.4430284798145294,
+      "rewards/check_gptzero_func": 0.7142857313156128,
+      "rewards/check_perplexity_diff_func": 0.3571428656578064,
+      "rewards/check_winston_local_func": 0.7983959913253784,
+      "step": 169
+    },
+    {
+      "clip_ratio": 0.005269515328109264,
+      "epoch": 0.23776223776223776,
+      "grad_norm": 0.14855282442633314,
+      "kl": 0.61328125,
+      "learning_rate": 5.944055944055944e-05,
+      "loss": -0.0079,
+      "step": 170
+    },
+    {
+      "clip_ratio": 0.0027731256559491158,
+      "completion_length": 117.83929443359375,
+      "epoch": 0.23916083916083916,
+      "grad_norm": 0.23058249563979205,
+      "kl": 0.412109375,
+      "learning_rate": 5.9790209790209796e-05,
+      "loss": 0.0055,
+      "num_tokens": 1719673.0,
+      "reward": 2.2663450241088867,
+      "reward_std": 0.46630793809890747,
+      "rewards/check_gptzero_func": 0.6785714030265808,
+      "rewards/check_perplexity_diff_func": 0.7142857313156128,
+      "rewards/check_winston_local_func": 0.8734878301620483,
+      "step": 171
+    },
+    {
+      "clip_ratio": 0.0034511894918978214,
+      "epoch": 0.24055944055944056,
+      "grad_norm": 0.17368207545699044,
+      "kl": 0.373046875,
+      "learning_rate": 6.0139860139860136e-05,
+      "loss": -0.0007,
+      "step": 172
+    },
+    {
+      "clip_ratio": 0.0032431341242045164,
+      "completion_length": 126.85714721679688,
+      "epoch": 0.24195804195804196,
+      "grad_norm": 0.17995690811350934,
+      "kl": 0.353515625,
+      "learning_rate": 6.048951048951049e-05,
+      "loss": -0.0011,
+      "num_tokens": 1742569.0,
+      "reward": 2.258990526199341,
+      "reward_std": 0.6008436679840088,
+      "rewards/check_gptzero_func": 0.75,
+      "rewards/check_perplexity_diff_func": 0.75,
+      "rewards/check_winston_local_func": 0.7589904069900513,
+      "step": 173
+    },
+    {
+      "clip_ratio": 0.006950656417757273,
+      "epoch": 0.24335664335664337,
+      "grad_norm": 0.13892848528506246,
+      "kl": 0.359375,
+      "learning_rate": 6.083916083916085e-05,
+      "loss": -0.0033,
+      "step": 174
+    },
+    {
+      "clip_ratio": 0.0027662119828164577,
+      "completion_length": 130.8928680419922,
+      "epoch": 0.24475524475524477,
+      "grad_norm": 0.10040261340554028,
+      "kl": 0.345703125,
+      "learning_rate": 6.118881118881119e-05,
+      "loss": -0.0032,
+      "num_tokens": 1765643.0,
+      "reward": 2.074061632156372,
+      "reward_std": 0.3745954632759094,
+      "rewards/check_gptzero_func": 0.6607142686843872,
+      "rewards/check_perplexity_diff_func": 0.6785714030265808,
+      "rewards/check_winston_local_func": 0.7347758412361145,
+      "step": 175
+    },
+    {
+      "clip_ratio": 0.0030742601957172155,
+      "epoch": 0.24615384615384617,
+      "grad_norm": 0.0921134439221548,
+      "kl": 0.3359375,
+      "learning_rate": 6.153846153846155e-05,
+      "loss": -0.0044,
+      "step": 176
+    },
+    {
+      "clip_ratio": 0.002027927665039897,
+      "completion_length": 89.73214721679688,
+      "epoch": 0.24755244755244754,
+      "grad_norm": 0.1709976566266231,
+      "kl": 0.50390625,
+      "learning_rate": 6.188811188811188e-05,
+      "loss": 0.0051,
+      "num_tokens": 1783674.0,
+      "reward": 1.7810758352279663,
+      "reward_std": 0.4943030774593353,
+      "rewards/check_gptzero_func": 0.6428571343421936,
+      "rewards/check_perplexity_diff_func": 0.4285714328289032,
+      "rewards/check_winston_local_func": 0.7096471786499023,
+      "step": 177
+    },
+    {
+      "clip_ratio": 0.0020612890366464853,
+      "epoch": 0.24895104895104894,
+      "grad_norm": 0.1477275348028994,
+      "kl": 0.50390625,
+      "learning_rate": 6.223776223776224e-05,
+      "loss": 0.0011,
+      "step": 178
+    },
+    {
+      "clip_ratio": 0.0021104796323925257,
+      "completion_length": 132.10714721679688,
+      "epoch": 0.25034965034965034,
+      "grad_norm": 0.08842848405697774,
+      "kl": 0.39453125,
+      "learning_rate": 6.258741258741259e-05,
+      "loss": 0.0005,
+      "num_tokens": 1806754.0,
+      "reward": 2.107767105102539,
+      "reward_std": 0.269815057516098,
+      "rewards/check_gptzero_func": 0.75,
+      "rewards/check_perplexity_diff_func": 0.5357142686843872,
+      "rewards/check_winston_local_func": 0.8220529556274414,
+      "step": 179
+    },
+    {
+      "clip_ratio": 0.004407648928463459,
+      "epoch": 0.2517482517482518,
+      "grad_norm": 0.07542595290297384,
+      "kl": 0.416015625,
+      "learning_rate": 6.293706293706293e-05,
+      "loss": -0.0006,
+      "step": 180
+    },
+    {
+      "clip_ratio": 0.0018615310546010733,
+      "completion_length": 89.30357360839844,
+      "epoch": 0.25314685314685315,
+      "grad_norm": 0.2895549842864223,
+      "kl": 0.498046875,
+      "learning_rate": 6.32867132867133e-05,
+      "loss": -0.0082,
+      "num_tokens": 1824795.0,
+      "reward": 1.8801069259643555,
+      "reward_std": 0.5523244738578796,
+      "rewards/check_gptzero_func": 0.6071428656578064,
+      "rewards/check_perplexity_diff_func": 0.5,
+      "rewards/check_winston_local_func": 0.7729640603065491,
+      "step": 181
+    },
+    {
+      "clip_ratio": 0.007555535528808832,
+      "epoch": 0.2545454545454545,
+      "grad_norm": 0.20874610098022578,
+      "kl": 0.59375,
+      "learning_rate": 6.363636363636364e-05,
+      "loss": -0.016,
+      "step": 182
+    },
+    {
+      "clip_ratio": 0.0016618422232568264,
+      "completion_length": 84.46428680419922,
+      "epoch": 0.25594405594405595,
+      "grad_norm": 0.20653753743630154,
+      "kl": 0.5625,
+      "learning_rate": 6.398601398601399e-05,
+      "loss": 0.0058,
+      "num_tokens": 1842573.0,
+      "reward": 2.1319751739501953,
+      "reward_std": 0.5030335187911987,
+      "rewards/check_gptzero_func": 0.7321428656578064,
+      "rewards/check_perplexity_diff_func": 0.5714285969734192,
+      "rewards/check_winston_local_func": 0.8284037709236145,
+      "step": 183
+    },
+    {
+      "clip_ratio": 0.00579726742580533,
+      "epoch": 0.2573426573426573,
+      "grad_norm": 0.15782001907156346,
+      "kl": 0.578125,
+      "learning_rate": 6.433566433566433e-05,
+      "loss": 0.0001,
+      "step": 184
+    },
+    {
+      "clip_ratio": 0.0016176491044461727,
+      "completion_length": 125.39286041259766,
+      "epoch": 0.25874125874125875,
+      "grad_norm": 0.12385736188741701,
+      "kl": 0.4921875,
+      "learning_rate": 6.46853146853147e-05,
+      "loss": -0.0027,
+      "num_tokens": 1864297.0,
+      "reward": 1.7732529640197754,
+      "reward_std": 0.3774341642856598,
+      "rewards/check_gptzero_func": 0.625,
+      "rewards/check_perplexity_diff_func": 0.4285714328289032,
+      "rewards/check_winston_local_func": 0.7196813821792603,
+      "step": 185
+    },
+    {
+      "clip_ratio": 0.00623862212523818,
+      "epoch": 0.2601398601398601,
+      "grad_norm": 0.10877531810926387,
+      "kl": 0.53125,
+      "learning_rate": 6.503496503496504e-05,
+      "loss": -0.0048,
+      "step": 186
+    },
+    {
+      "clip_ratio": 0.0031279984395951033,
+      "completion_length": 98.00000762939453,
+      "epoch": 0.26153846153846155,
+      "grad_norm": 0.3661429387824046,
+      "kl": 0.6328125,
+      "learning_rate": 6.538461538461539e-05,
+      "loss": -0.0063,
+      "num_tokens": 1882931.0,
+      "reward": 2.0694050788879395,
+      "reward_std": 0.6090536117553711,
+      "rewards/check_gptzero_func": 0.6785714030265808,
+      "rewards/check_perplexity_diff_func": 0.5714285969734192,
+      "rewards/check_winston_local_func": 0.819405198097229,
+      "step": 187
+    },
+    {
+      "clip_ratio": 0.010017934255301952,
+      "epoch": 0.2629370629370629,
+      "grad_norm": 0.34977841113727764,
+      "kl": 0.65234375,
+      "learning_rate": 6.573426573426573e-05,
+      "loss": -0.0185,
+      "step": 188
+    },
+    {
+      "clip_ratio": 0.00222155568189919,
+      "completion_length": 133.44644165039062,
+      "epoch": 0.26433566433566436,
+      "grad_norm": 0.18715243199492715,
+      "kl": 0.46484375,
+      "learning_rate": 6.608391608391609e-05,
+      "loss": 0.0062,
+      "num_tokens": 1906084.0,
+      "reward": 2.364468812942505,
+      "reward_std": 0.5593475699424744,
+      "rewards/check_gptzero_func": 0.6964285969734192,
+      "rewards/check_perplexity_diff_func": 0.8571428656578064,
+      "rewards/check_winston_local_func": 0.8108974099159241,
+      "step": 189
+    },
+    {
+      "clip_ratio": 0.008103223517537117,
+      "epoch": 0.26573426573426573,
+      "grad_norm": 0.14058389303467,
+      "kl": 0.45703125,
+      "learning_rate": 6.643356643356644e-05,
+      "loss": 0.0025,
+      "step": 190
+    },
+    {
+      "clip_ratio": 0.0019668787717819214,
+      "completion_length": 142.07144165039062,
+      "epoch": 0.26713286713286716,
+      "grad_norm": 0.14399812178023874,
+      "kl": 0.43359375,
+      "learning_rate": 6.678321678321679e-05,
+      "loss": -0.0031,
+      "num_tokens": 1930224.0,
+      "reward": 2.1098926067352295,
+      "reward_std": 0.5390018820762634,
+      "rewards/check_gptzero_func": 0.8214285969734192,
+      "rewards/check_perplexity_diff_func": 0.4642857015132904,
+      "rewards/check_winston_local_func": 0.8241782784461975,
+      "step": 191
+    },
+    {
+      "clip_ratio": 0.005082913674414158,
+      "epoch": 0.26853146853146853,
+      "grad_norm": 0.12024703070308183,
+      "kl": 0.447265625,
+      "learning_rate": 6.713286713286715e-05,
+      "loss": -0.0055,
+      "step": 192
+    },
+    {
+      "clip_ratio": 0.0032251765951514244,
+      "completion_length": 84.98214721679688,
+      "epoch": 0.2699300699300699,
+      "grad_norm": 0.32527529059153654,
+      "kl": 0.6640625,
+      "learning_rate": 6.748251748251748e-05,
+      "loss": 0.0186,
+      "num_tokens": 1947399.0,
+      "reward": 1.9443087577819824,
+      "reward_std": 0.6417390704154968,
+      "rewards/check_gptzero_func": 0.7678571343421936,
+      "rewards/check_perplexity_diff_func": 0.2857142984867096,
+      "rewards/check_winston_local_func": 0.8907372355461121,
+      "step": 193
+    },
+    {
+      "clip_ratio": 0.008085422217845917,
+      "epoch": 0.27132867132867133,
+      "grad_norm": 0.21806984416977268,
+      "kl": 0.69140625,
+      "learning_rate": 6.783216783216784e-05,
+      "loss": 0.0068,
+      "step": 194
+    },
+    {
+      "clip_ratio": 0.0025381618179380894,
+      "completion_length": 91.66072082519531,
+      "epoch": 0.2727272727272727,
+      "grad_norm": 0.1831840025015104,
+      "kl": 0.53125,
+      "learning_rate": 6.818181818181818e-05,
+      "loss": 0.0014,
+      "num_tokens": 1965402.0,
+      "reward": 2.3979477882385254,
+      "reward_std": 0.45264866948127747,
+      "rewards/check_gptzero_func": 0.7678571343421936,
+      "rewards/check_perplexity_diff_func": 0.7857142686843872,
+      "rewards/check_winston_local_func": 0.8443759083747864,
+      "step": 195
+    },
+    {
+      "clip_ratio": 0.005082620773464441,
+      "epoch": 0.27412587412587414,
+      "grad_norm": 0.14399845631168656,
+      "kl": 0.515625,
+      "learning_rate": 6.853146853146853e-05,
+      "loss": -0.0023,
+      "step": 196
+    },
+    {
+      "clip_ratio": 0.0025359569117426872,
+      "completion_length": 118.80357360839844,
+      "epoch": 0.2755244755244755,
+      "grad_norm": 0.1617404225919446,
+      "kl": 0.5625,
+      "learning_rate": 6.888111888111889e-05,
+      "loss": 0.0146,
+      "num_tokens": 1986731.0,
+      "reward": 2.078895330429077,
+      "reward_std": 0.31289052963256836,
+      "rewards/check_gptzero_func": 0.8035714030265808,
+      "rewards/check_perplexity_diff_func": 0.3928571343421936,
+      "rewards/check_winston_local_func": 0.882466733455658,
+      "step": 197
+    },
+    {
+      "clip_ratio": 0.0062335156835615635,
+      "epoch": 0.27692307692307694,
+      "grad_norm": 0.12006362566688389,
+      "kl": 0.69921875,
+      "learning_rate": 6.923076923076924e-05,
+      "loss": 0.0125,
+      "step": 198
+    },
+    {
+      "clip_ratio": 0.0023628328926861286,
+      "completion_length": 113.76786041259766,
+      "epoch": 0.2783216783216783,
+      "grad_norm": 0.19235514876641358,
+      "kl": 0.58984375,
+      "learning_rate": 6.958041958041958e-05,
+      "loss": -0.0069,
+      "num_tokens": 2007822.0,
+      "reward": 2.0453529357910156,
+      "reward_std": 0.576248288154602,
+      "rewards/check_gptzero_func": 0.7321428656578064,
+      "rewards/check_perplexity_diff_func": 0.5714285969734192,
+      "rewards/check_winston_local_func": 0.7417814135551453,
+      "step": 199
+    },
+    {
+      "clip_ratio": 0.0030245708767324686,
+      "epoch": 0.27972027972027974,
+      "grad_norm": 0.15320753759426498,
+      "kl": 0.578125,
+      "learning_rate": 6.993006993006993e-05,
+      "loss": -0.0115,
+      "step": 200
+    },
+    {
+      "clip_ratio": 0.0028001146856695414,
+      "completion_length": 121.51786041259766,
+      "epoch": 0.2811188811188811,
+      "grad_norm": 0.2770839119590159,
+      "kl": 0.5859375,
+      "learning_rate": 7.027972027972029e-05,
+      "loss": -0.0321,
+      "num_tokens": 2030153.0,
+      "reward": 2.223505735397339,
+      "reward_std": 0.580276370048523,
+      "rewards/check_gptzero_func": 0.8392857313156128,
+      "rewards/check_perplexity_diff_func": 0.5,
+      "rewards/check_winston_local_func": 0.8842198252677917,
+      "step": 201
+    },
+    {
+      "clip_ratio": 0.004435424692928791,
+      "epoch": 0.28251748251748254,
+      "grad_norm": 0.1752123363810394,
+      "kl": 0.61328125,
+      "learning_rate": 7.062937062937062e-05,
+      "loss": -0.0398,
+      "step": 202
+    },
+    {
+      "clip_ratio": 0.003775001736357808,
+      "completion_length": 132.96429443359375,
+      "epoch": 0.2839160839160839,
+      "grad_norm": 0.20528144857768493,
+      "kl": 0.5390625,
+      "learning_rate": 7.097902097902098e-05,
+      "loss": -0.0054,
+      "num_tokens": 2053611.0,
+      "reward": 2.4398131370544434,
+      "reward_std": 0.6120996475219727,
+      "rewards/check_gptzero_func": 0.8214285969734192,
+      "rewards/check_perplexity_diff_func": 0.8571428656578064,
+      "rewards/check_winston_local_func": 0.761241614818573,
+      "step": 203
+    },
+    {
+      "clip_ratio": 0.0037133130244910717,
+      "epoch": 0.2853146853146853,
+      "grad_norm": 0.14916355666745199,
+      "kl": 0.55859375,
+      "learning_rate": 7.132867132867134e-05,
+      "loss": -0.0112,
+      "step": 204
+    },
+    {
+      "clip_ratio": 0.002573356730863452,
+      "completion_length": 109.51786041259766,
+      "epoch": 0.2867132867132867,
+      "grad_norm": 0.15672915743695726,
+      "kl": 1.4921875,
+      "learning_rate": 7.167832167832168e-05,
+      "loss": 0.0032,
+      "num_tokens": 2073846.0,
+      "reward": 2.221467971801758,
+      "reward_std": 0.492183119058609,
+      "rewards/check_gptzero_func": 0.6964285969734192,
+      "rewards/check_perplexity_diff_func": 0.6428571343421936,
+      "rewards/check_winston_local_func": 0.8821821212768555,
+      "step": 205
+    },
+    {
+      "clip_ratio": 0.0036338225472718477,
+      "epoch": 0.2881118881118881,
+      "grad_norm": 0.27785877634425765,
+      "kl": 0.74609375,
+      "learning_rate": 7.202797202797204e-05,
+      "loss": 0.0013,
+      "step": 206
+    },
+    {
+      "clip_ratio": 0.004580673761665821,
+      "completion_length": 95.05357360839844,
+      "epoch": 0.2895104895104895,
+      "grad_norm": 0.1746155777114636,
+      "kl": 0.671875,
+      "learning_rate": 7.237762237762238e-05,
+      "loss": -0.0068,
+      "num_tokens": 2092355.0,
+      "reward": 2.0429205894470215,
+      "reward_std": 0.36572587490081787,
+      "rewards/check_gptzero_func": 0.6964285969734192,
+      "rewards/check_perplexity_diff_func": 0.4642857015132904,
+      "rewards/check_winston_local_func": 0.8822061419487,
+      "step": 207
+    },
+    {
+      "clip_ratio": 0.007289381232112646,
+      "epoch": 0.2909090909090909,
+      "grad_norm": 0.12704093759252294,
+      "kl": 0.75390625,
+      "learning_rate": 7.272727272727273e-05,
+      "loss": -0.0111,
+      "step": 208
+    },
+    {
+      "clip_ratio": 0.0038105440326035023,
+      "completion_length": 113.4464340209961,
+      "epoch": 0.2923076923076923,
+      "grad_norm": 0.20958397036994925,
+      "kl": 0.6015625,
+      "learning_rate": 7.307692307692307e-05,
+      "loss": 0.0082,
+      "num_tokens": 2113542.0,
+      "reward": 2.162785530090332,
+      "reward_std": 0.49807849526405334,
+      "rewards/check_gptzero_func": 0.75,
+      "rewards/check_perplexity_diff_func": 0.5714285969734192,
+      "rewards/check_winston_local_func": 0.8413568139076233,
+      "step": 209
+    },
+    {
+      "clip_ratio": 0.007606918923556805,
+      "epoch": 0.2937062937062937,
+      "grad_norm": 0.1766466418371376,
+      "kl": 0.63671875,
+      "learning_rate": 7.342657342657343e-05,
+      "loss": 0.0025,
+      "step": 210
+    },
+    {
+      "clip_ratio": 0.0019986790139228106,
+      "completion_length": 92.73214721679688,
+      "epoch": 0.2951048951048951,
+      "grad_norm": 0.2716166901527731,
+      "kl": 0.6171875,
+      "learning_rate": 7.377622377622378e-05,
+      "loss": -0.0005,
+      "num_tokens": 2131873.0,
+      "reward": 2.099912405014038,
+      "reward_std": 0.5661742687225342,
+      "rewards/check_gptzero_func": 0.8214285969734192,
+      "rewards/check_perplexity_diff_func": 0.4285714328289032,
+      "rewards/check_winston_local_func": 0.8499122858047485,
+      "step": 211
+    },
+    {
+      "clip_ratio": 0.005917501635849476,
+      "epoch": 0.2965034965034965,
+      "grad_norm": 0.19270405158731238,
+      "kl": 0.62890625,
+      "learning_rate": 7.412587412587413e-05,
+      "loss": -0.0079,
+      "step": 212
+    },
+    {
+      "clip_ratio": 0.002530448604375124,
+      "completion_length": 96.51786041259766,
+      "epoch": 0.29790209790209793,
+      "grad_norm": 0.22723256705117592,
+      "kl": 0.69921875,
+      "learning_rate": 7.447552447552449e-05,
+      "loss": -0.009,
+      "num_tokens": 2150462.0,
+      "reward": 2.1355700492858887,
+      "reward_std": 0.3738899827003479,
+      "rewards/check_gptzero_func": 0.8571428656578064,
+      "rewards/check_perplexity_diff_func": 0.3571428656578064,
+      "rewards/check_winston_local_func": 0.9212842583656311,
+      "step": 213
+    },
+    {
+      "clip_ratio": 0.011397747322916985,
+      "epoch": 0.2993006993006993,
+      "grad_norm": 0.18189402035580596,
+      "kl": 0.7265625,
+      "learning_rate": 7.482517482517482e-05,
+      "loss": -0.0139,
+      "step": 214
+    },
+    {
+      "clip_ratio": 0.004513947293162346,
+      "completion_length": 94.66072082519531,
+      "epoch": 0.3006993006993007,
+      "grad_norm": 0.2675295151019955,
+      "kl": 0.67578125,
+      "learning_rate": 7.517482517482518e-05,
+      "loss": 0.0014,
+      "num_tokens": 2169243.0,
+      "reward": 2.2817916870117188,
+      "reward_std": 0.4118366539478302,
+      "rewards/check_gptzero_func": 0.8571428656578064,
+      "rewards/check_perplexity_diff_func": 0.4642857015132904,
+      "rewards/check_winston_local_func": 0.9603630900382996,
+      "step": 215
+    },
+    {
+      "clip_ratio": 0.010964194312691689,
+      "epoch": 0.3020979020979021,
+      "grad_norm": 0.17031623307626814,
+      "kl": 0.875,
+      "learning_rate": 7.552447552447553e-05,
+      "loss": -0.0057,
+      "step": 216
+    },
+    {
+      "clip_ratio": 0.003371638245880604,
+      "completion_length": 118.17857360839844,
+      "epoch": 0.3034965034965035,
+      "grad_norm": 0.1847364030092435,
+      "kl": 0.60546875,
+      "learning_rate": 7.587412587412587e-05,
+      "loss": -0.0022,
+      "num_tokens": 2190571.0,
+      "reward": 2.1225900650024414,
+      "reward_std": 0.5403507947921753,
+      "rewards/check_gptzero_func": 0.75,
+      "rewards/check_perplexity_diff_func": 0.4642857015132904,
+      "rewards/check_winston_local_func": 0.9083043336868286,
+      "step": 217
+    },
+    {
+      "clip_ratio": 0.01002263929694891,
+      "epoch": 0.3048951048951049,
+      "grad_norm": 0.1449002041728082,
+      "kl": 0.61328125,
+      "learning_rate": 7.622377622377622e-05,
+      "loss": -0.0067,
+      "step": 218
+    },
+    {
+      "clip_ratio": 0.0013253266224637628,
+      "completion_length": 129.33929443359375,
+      "epoch": 0.3062937062937063,
+      "grad_norm": 0.1504069021149342,
+      "kl": 0.578125,
+      "learning_rate": 7.657342657342658e-05,
+      "loss": -0.0075,
+      "num_tokens": 2213408.0,
+      "reward": 2.1974618434906006,
+      "reward_std": 0.4602964520454407,
+      "rewards/check_gptzero_func": 0.7857142686843872,
+      "rewards/check_perplexity_diff_func": 0.5,
+      "rewards/check_winston_local_func": 0.911747395992279,
+      "step": 219
+    },
+    {
+      "clip_ratio": 0.006876418832689524,
+      "epoch": 0.3076923076923077,
+      "grad_norm": 0.11992971275284438,
+      "kl": 0.57421875,
+      "learning_rate": 7.692307692307693e-05,
+      "loss": -0.0101,
+      "step": 220
+    },
+    {
+      "clip_ratio": 0.0024474726524204016,
+      "completion_length": 89.67857360839844,
+      "epoch": 0.3090909090909091,
+      "grad_norm": 0.46915411380362765,
+      "kl": 0.73046875,
+      "learning_rate": 7.727272727272727e-05,
+      "loss": 0.0333,
+      "num_tokens": 2231248.0,
+      "reward": 2.161350727081299,
+      "reward_std": 0.47136664390563965,
+      "rewards/check_gptzero_func": 0.8214285969734192,
+      "rewards/check_perplexity_diff_func": 0.4642857015132904,
+      "rewards/check_winston_local_func": 0.8756363987922668,
+      "step": 221
+    },
+    {
+      "clip_ratio": 0.02464432455599308,
+      "epoch": 0.3104895104895105,
+      "grad_norm": 0.30204889343007607,
+      "kl": 0.8046875,
+      "learning_rate": 7.762237762237763e-05,
+      "loss": 0.0211,
+      "step": 222
+    },
+    {
+      "clip_ratio": 0.0017527465242892504,
+      "completion_length": 134.17857360839844,
+      "epoch": 0.3118881118881119,
+      "grad_norm": 0.18948229022832336,
+      "kl": 0.6328125,
+      "learning_rate": 7.797202797202798e-05,
+      "loss": -0.0049,
+      "num_tokens": 2255162.0,
+      "reward": 2.4446465969085693,
+      "reward_std": 0.5171672105789185,
+      "rewards/check_gptzero_func": 0.7678571343421936,
+      "rewards/check_perplexity_diff_func": 0.7857142686843872,
+      "rewards/check_winston_local_func": 0.8910752534866333,
+      "step": 223
+    },
+    {
+      "clip_ratio": 0.006593752186745405,
+      "epoch": 0.3132867132867133,
+      "grad_norm": 0.15013602627394823,
+      "kl": 0.640625,
+      "learning_rate": 7.832167832167832e-05,
+      "loss": -0.0097,
+      "step": 224
+    },
+    {
+      "clip_ratio": 0.0019926591776311398,
+      "completion_length": 127.00000762939453,
+      "epoch": 0.3146853146853147,
+      "grad_norm": 0.22875481051435598,
+      "kl": 0.6875,
+      "learning_rate": 7.867132867132867e-05,
+      "loss": 0.0021,
+      "num_tokens": 2277320.0,
+      "reward": 2.2040371894836426,
+      "reward_std": 0.5528277158737183,
+      "rewards/check_gptzero_func": 0.7321428656578064,
+      "rewards/check_perplexity_diff_func": 0.5357142686843872,
+      "rewards/check_winston_local_func": 0.936180055141449,
+      "step": 225
+    },
+    {
+      "clip_ratio": 0.005839359946548939,
+      "epoch": 0.31608391608391606,
+      "grad_norm": 0.17829200080123941,
+      "kl": 0.68359375,
+      "learning_rate": 7.902097902097903e-05,
+      "loss": -0.0053,
+      "step": 226
+    },
+    {
+      "clip_ratio": 0.0026252593379467726,
+      "completion_length": 130.0357208251953,
+      "epoch": 0.3174825174825175,
+      "grad_norm": 0.16076101957107508,
+      "kl": 0.66015625,
+      "learning_rate": 7.937062937062938e-05,
+      "loss": -0.0033,
+      "num_tokens": 2299818.0,
+      "reward": 2.3278117179870605,
+      "reward_std": 0.48680880665779114,
+      "rewards/check_gptzero_func": 0.8035714030265808,
+      "rewards/check_perplexity_diff_func": 0.6428571343421936,
+      "rewards/check_winston_local_func": 0.8813830614089966,
+      "step": 227
+    },
+    {
+      "clip_ratio": 0.004954828415066004,
+      "epoch": 0.31888111888111886,
+      "grad_norm": 0.12734264227867195,
+      "kl": 0.6875,
+      "learning_rate": 7.972027972027972e-05,
+      "loss": -0.0071,
+      "step": 228
+    },
+    {
+      "clip_ratio": 0.0034341278951615095,
+      "completion_length": 133.7857208251953,
+      "epoch": 0.3202797202797203,
+      "grad_norm": 0.24065024754984402,
+      "kl": 0.6875,
+      "learning_rate": 8.006993006993007e-05,
+      "loss": -0.0052,
+      "num_tokens": 2322948.0,
+      "reward": 2.5632615089416504,
+      "reward_std": 0.6206622123718262,
+      "rewards/check_gptzero_func": 0.9642857313156128,
+      "rewards/check_perplexity_diff_func": 0.6785714030265808,
+      "rewards/check_winston_local_func": 0.9204041361808777,
+      "step": 229
+    },
+    {
+      "clip_ratio": 0.006521092262119055,
+      "epoch": 0.32167832167832167,
+      "grad_norm": 0.18602361781154253,
+      "kl": 0.6796875,
+      "learning_rate": 8.041958041958042e-05,
+      "loss": -0.013,
+      "step": 230
+    },
+    {
+      "clip_ratio": 0.004207184072583914,
+      "completion_length": 116.53572082519531,
+      "epoch": 0.3230769230769231,
+      "grad_norm": 0.36704569865090053,
+      "kl": 0.69921875,
+      "learning_rate": 8.076923076923078e-05,
+      "loss": 0.0055,
+      "num_tokens": 2344106.0,
+      "reward": 2.4917171001434326,
+      "reward_std": 0.6073458790779114,
+      "rewards/check_gptzero_func": 0.8392857313156128,
+      "rewards/check_perplexity_diff_func": 0.75,
+      "rewards/check_winston_local_func": 0.902431309223175,
+      "step": 231
+    },
+    {
+      "clip_ratio": 0.013317321427166462,
+      "epoch": 0.32447552447552447,
+      "grad_norm": 0.13844634287775134,
+      "kl": 0.73046875,
+      "learning_rate": 8.111888111888112e-05,
+      "loss": -0.0039,
+      "step": 232
+    },
+    {
+      "clip_ratio": 0.004024635534733534,
+      "completion_length": 134.92857360839844,
+      "epoch": 0.3258741258741259,
+      "grad_norm": 0.34168858248313216,
+      "kl": 0.64453125,
+      "learning_rate": 8.146853146853147e-05,
+      "loss": -0.0057,
+      "num_tokens": 2367724.0,
+      "reward": 2.20210599899292,
+      "reward_std": 0.5888614654541016,
+      "rewards/check_gptzero_func": 0.7321428656578064,
+      "rewards/check_perplexity_diff_func": 0.6071428656578064,
+      "rewards/check_winston_local_func": 0.8628200888633728,
+      "step": 233
+    },
+    {
+      "clip_ratio": 0.009708845987915993,
+      "epoch": 0.32727272727272727,
+      "grad_norm": 0.19135959991877635,
+      "kl": 0.67578125,
+      "learning_rate": 8.181818181818183e-05,
+      "loss": -0.0146,
+      "step": 234
+    },
+    {
+      "clip_ratio": 0.005571329966187477,
+      "completion_length": 98.75000762939453,
+      "epoch": 0.32867132867132864,
+      "grad_norm": 0.32328761096657005,
+      "kl": 0.77734375,
+      "learning_rate": 8.216783216783218e-05,
+      "loss": 0.0128,
+      "num_tokens": 2386636.0,
+      "reward": 2.2358410358428955,
+      "reward_std": 0.599101185798645,
+      "rewards/check_gptzero_func": 0.8214285969734192,
+      "rewards/check_perplexity_diff_func": 0.5357142686843872,
+      "rewards/check_winston_local_func": 0.8786982297897339,
+      "step": 235
+    },
+    {
+      "clip_ratio": 0.009436835534870625,
+      "epoch": 0.3300699300699301,
+      "grad_norm": 0.19557085227388898,
+      "kl": 0.8515625,
+      "learning_rate": 8.251748251748252e-05,
+      "loss": 0.0022,
+      "step": 236
+    },
+    {
+      "clip_ratio": 0.006164635997265577,
+      "completion_length": 79.75,
+      "epoch": 0.33146853146853145,
+      "grad_norm": 0.3250433111655094,
+      "kl": 0.98828125,
+      "learning_rate": 8.286713286713287e-05,
+      "loss": -0.0126,
+      "num_tokens": 2403400.0,
+      "reward": 2.1074206829071045,
+      "reward_std": 0.41642776131629944,
+      "rewards/check_gptzero_func": 0.9107142686843872,
+      "rewards/check_perplexity_diff_func": 0.25,
+      "rewards/check_winston_local_func": 0.946706235408783,
+      "step": 237
+    },
+    {
+      "clip_ratio": 0.0066314926370978355,
+      "epoch": 0.3328671328671329,
+      "grad_norm": 0.1612667131666411,
+      "kl": 0.98828125,
+      "learning_rate": 8.321678321678323e-05,
+      "loss": -0.0225,
+      "step": 238
+    },
+    {
+      "clip_ratio": 0.003079179208725691,
+      "completion_length": 117.10714721679688,
+      "epoch": 0.33426573426573425,
+      "grad_norm": 0.2505534626607256,
+      "kl": 0.6796875,
+      "learning_rate": 8.356643356643356e-05,
+      "loss": -0.0103,
+      "num_tokens": 2425354.0,
+      "reward": 2.3166799545288086,
+      "reward_std": 0.5270359516143799,
+      "rewards/check_gptzero_func": 0.8214285969734192,
+      "rewards/check_perplexity_diff_func": 0.5714285969734192,
+      "rewards/check_winston_local_func": 0.9238227605819702,
+      "step": 239
+    },
+    {
+      "clip_ratio": 0.006801524665206671,
+      "epoch": 0.3356643356643357,
+      "grad_norm": 0.16826419365505382,
+      "kl": 0.671875,
+      "learning_rate": 8.391608391608392e-05,
+      "loss": -0.0182,
+      "step": 240
+    },
+    {
+      "clip_ratio": 0.004480496048927307,
+      "completion_length": 125.05357360839844,
+      "epoch": 0.33706293706293705,
+      "grad_norm": 0.26807661638826913,
+      "kl": 0.60546875,
+      "learning_rate": 8.426573426573428e-05,
+      "loss": -0.0086,
+      "num_tokens": 2447333.0,
+      "reward": 2.2939677238464355,
+      "reward_std": 0.39558398723602295,
+      "rewards/check_gptzero_func": 0.9107142686843872,
+      "rewards/check_perplexity_diff_func": 0.5357142686843872,
+      "rewards/check_winston_local_func": 0.8475390672683716,
+      "step": 241
+    },
+    {
+      "clip_ratio": 0.007587091531604528,
+      "epoch": 0.3384615384615385,
+      "grad_norm": 0.13602673124742348,
+      "kl": 0.57421875,
+      "learning_rate": 8.461538461538461e-05,
+      "loss": -0.0152,
+      "step": 242
+    },
+    {
+      "clip_ratio": 0.0034923183266073465,
+      "completion_length": 98.71428680419922,
+      "epoch": 0.33986013986013985,
+      "grad_norm": 0.7773422467135167,
+      "kl": 6.5,
+      "learning_rate": 8.496503496503497e-05,
+      "loss": 0.0071,
+      "num_tokens": 2466435.0,
+      "reward": 2.4026196002960205,
+      "reward_std": 0.4264836013317108,
+      "rewards/check_gptzero_func": 0.8571428656578064,
+      "rewards/check_perplexity_diff_func": 0.6785714030265808,
+      "rewards/check_winston_local_func": 0.8669052124023438,
+      "step": 243
+    },
+    {
+      "clip_ratio": 0.026189187541604042,
+      "epoch": 0.3412587412587413,
+      "grad_norm": 130.9971047023429,
+      "kl": 0.70703125,
+      "learning_rate": 8.531468531468532e-05,
+      "loss": 1.8998,
+      "step": 244
+    },
+    {
+      "clip_ratio": 0.004780410788953304,
+      "completion_length": 135.0,
+      "epoch": 0.34265734265734266,
+      "grad_norm": 0.2329479243774761,
+      "kl": 0.87109375,
+      "learning_rate": 8.566433566433567e-05,
+      "loss": 0.0011,
+      "num_tokens": 2489651.0,
+      "reward": 2.2166569232940674,
+      "reward_std": 0.5797281265258789,
+      "rewards/check_gptzero_func": 0.9107142686843872,
+      "rewards/check_perplexity_diff_func": 0.4285714328289032,
+      "rewards/check_winston_local_func": 0.8773713111877441,
+      "step": 245
+    },
+    {
+      "clip_ratio": 0.021273447200655937,
+      "epoch": 0.34405594405594403,
+      "grad_norm": 0.22672502397660207,
+      "kl": 0.9609375,
+      "learning_rate": 8.601398601398601e-05,
+      "loss": -0.0045,
+      "step": 246
+    },
+    {
+      "clip_ratio": 0.002856873208656907,
+      "completion_length": 122.66072082519531,
+      "epoch": 0.34545454545454546,
+      "grad_norm": 0.19772681964626543,
+      "kl": 0.984375,
+      "learning_rate": 8.636363636363637e-05,
+      "loss": 0.0028,
+      "num_tokens": 2510506.0,
+      "reward": 2.323245048522949,
+      "reward_std": 0.46726977825164795,
+      "rewards/check_gptzero_func": 0.9107142686843872,
+      "rewards/check_perplexity_diff_func": 0.5,
+      "rewards/check_winston_local_func": 0.9125306010246277,
+      "step": 247
+    },
+    {
+      "clip_ratio": 0.02299562282860279,
+      "epoch": 0.34685314685314683,
+      "grad_norm": 0.1812005085005865,
+      "kl": 0.87890625,
+      "learning_rate": 8.67132867132867e-05,
+      "loss": -0.001,
+      "step": 248
+    },
+    {
+      "clip_ratio": 0.0026335662696510553,
+      "completion_length": 116.85714721679688,
+      "epoch": 0.34825174825174826,
+      "grad_norm": 0.28731547808119623,
+      "kl": 0.76953125,
+      "learning_rate": 8.706293706293707e-05,
+      "loss": 0.024,
+      "num_tokens": 2531254.0,
+      "reward": 2.5312695503234863,
+      "reward_std": 0.5851892828941345,
+      "rewards/check_gptzero_func": 0.8571428656578064,
+      "rewards/check_perplexity_diff_func": 0.8214285969734192,
+      "rewards/check_winston_local_func": 0.8526979088783264,
+      "step": 249
+    },
+    {
+      "clip_ratio": 0.012148472480475903,
+      "epoch": 0.34965034965034963,
+      "grad_norm": 0.21250398409595742,
+      "kl": 0.734375,
+      "learning_rate": 8.741258741258743e-05,
+      "loss": 0.0156,
+      "step": 250
+    },
+    {
+      "clip_ratio": 0.0036881309933960438,
+      "completion_length": 109.12500762939453,
+      "epoch": 0.35104895104895106,
+      "grad_norm": 0.278787459589393,
+      "kl": 1.3671875,
+      "learning_rate": 8.776223776223776e-05,
+      "loss": -0.0127,
+      "num_tokens": 2551265.0,
+      "reward": 2.24423885345459,
+      "reward_std": 0.6221095323562622,
+      "rewards/check_gptzero_func": 0.875,
+      "rewards/check_perplexity_diff_func": 0.5,
+      "rewards/check_winston_local_func": 0.8692389130592346,
+      "step": 251
+    },
+    {
+      "clip_ratio": 0.011438491754233837,
+      "epoch": 0.35244755244755244,
+      "grad_norm": 0.21324998809887244,
+      "kl": 1.0859375,
+      "learning_rate": 8.811188811188812e-05,
+      "loss": -0.023,
+      "step": 252
+    },
+    {
+      "clip_ratio": 0.0039367591962218285,
+      "completion_length": 112.14286041259766,
+      "epoch": 0.35384615384615387,
+      "grad_norm": 0.31069817037713615,
+      "kl": 0.91015625,
+      "learning_rate": 8.846153846153847e-05,
+      "loss": 0.0167,
+      "num_tokens": 2572119.0,
+      "reward": 2.492161989212036,
+      "reward_std": 0.6407585144042969,
+      "rewards/check_gptzero_func": 0.875,
+      "rewards/check_perplexity_diff_func": 0.75,
+      "rewards/check_winston_local_func": 0.8671619296073914,
+      "step": 253
+    },
+    {
+      "clip_ratio": 0.005880096461623907,
+      "epoch": 0.35524475524475524,
+      "grad_norm": 0.19008007229653096,
+      "kl": 0.90234375,
+      "learning_rate": 8.881118881118881e-05,
+      "loss": 0.0057,
+      "step": 254
+    },
+    {
+      "clip_ratio": 0.003151817014440894,
+      "completion_length": 120.78572082519531,
+      "epoch": 0.35664335664335667,
+      "grad_norm": 0.4113945542101958,
+      "kl": 0.890625,
+      "learning_rate": 8.916083916083916e-05,
+      "loss": 0.0079,
+      "num_tokens": 2593081.0,
+      "reward": 2.3228745460510254,
+      "reward_std": 0.5329480171203613,
+      "rewards/check_gptzero_func": 0.875,
+      "rewards/check_perplexity_diff_func": 0.6428571343421936,
+      "rewards/check_winston_local_func": 0.8050170540809631,
+      "step": 255
+    },
+    {
+      "clip_ratio": 0.01224872563034296,
+      "epoch": 0.35804195804195804,
+      "grad_norm": 0.24677625747594412,
+      "kl": 1.0078125,
+      "learning_rate": 8.951048951048952e-05,
+      "loss": 0.0006,
+      "step": 256
+    },
+    {
+      "clip_ratio": 0.003666934324428439,
+      "completion_length": 146.5178680419922,
+      "epoch": 0.3594405594405594,
+      "grad_norm": 0.21537926236506497,
+      "kl": 0.484375,
+      "learning_rate": 8.986013986013986e-05,
+      "loss": 0.0068,
+      "num_tokens": 2618758.0,
+      "reward": 2.198406934738159,
+      "reward_std": 0.3875991106033325,
+      "rewards/check_gptzero_func": 0.9285714030265808,
+      "rewards/check_perplexity_diff_func": 0.4285714328289032,
+      "rewards/check_winston_local_func": 0.8412641882896423,
+      "step": 257
+    },
+    {
+      "clip_ratio": 0.012731654569506645,
+      "epoch": 0.36083916083916084,
+      "grad_norm": 0.1543455831571161,
+      "kl": 0.478515625,
+      "learning_rate": 9.020979020979021e-05,
+      "loss": 0.0025,
+      "step": 258
+    },
+    {
+      "clip_ratio": 0.0012050624936819077,
+      "completion_length": 129.375,
+      "epoch": 0.3622377622377622,
+      "grad_norm": 0.16107873823673136,
+      "kl": 1.6640625,
+      "learning_rate": 9.055944055944057e-05,
+      "loss": 0.0054,
+      "num_tokens": 2641815.0,
+      "reward": 2.4805288314819336,
+      "reward_std": 0.3510296940803528,
+      "rewards/check_gptzero_func": 0.8392857313156128,
+      "rewards/check_perplexity_diff_func": 0.8214285969734192,
+      "rewards/check_winston_local_func": 0.8198142647743225,
+      "step": 259
+    },
+    {
+      "clip_ratio": 0.006208478473126888,
+      "epoch": 0.36363636363636365,
+      "grad_norm": 0.10546346547579268,
+      "kl": 0.8046875,
+      "learning_rate": 9.090909090909092e-05,
+      "loss": 0.0019,
+      "step": 260
+    },
+    {
+      "clip_ratio": 0.0025593352038413286,
+      "completion_length": 128.57144165039062,
+      "epoch": 0.365034965034965,
+      "grad_norm": 0.1986264685447053,
+      "kl": 0.6171875,
+      "learning_rate": 9.125874125874126e-05,
+      "loss": -0.0025,
+      "num_tokens": 2664343.0,
+      "reward": 2.261420249938965,
+      "reward_std": 0.5704914927482605,
+      "rewards/check_gptzero_func": 0.7857142686843872,
+      "rewards/check_perplexity_diff_func": 0.6428571343421936,
+      "rewards/check_winston_local_func": 0.8328486680984497,
+      "step": 261
+    },
+    {
+      "clip_ratio": 0.002767725382000208,
+      "epoch": 0.36643356643356645,
+      "grad_norm": 0.14267481856078115,
+      "kl": 0.66015625,
+      "learning_rate": 9.160839160839161e-05,
+      "loss": -0.0071,
+      "step": 262
+    },
+    {
+      "clip_ratio": 0.0036019759718328714,
+      "completion_length": 108.30357360839844,
+      "epoch": 0.3678321678321678,
+      "grad_norm": 0.25429899311591847,
+      "kl": 0.61328125,
+      "learning_rate": 9.195804195804196e-05,
+      "loss": 0.0114,
+      "num_tokens": 2684460.0,
+      "reward": 2.1853349208831787,
+      "reward_std": 0.5545978546142578,
+      "rewards/check_gptzero_func": 0.7142857313156128,
+      "rewards/check_perplexity_diff_func": 0.6428571343421936,
+      "rewards/check_winston_local_func": 0.828191876411438,
+      "step": 263
+    },
+    {
+      "clip_ratio": 0.002995865885168314,
+      "epoch": 0.36923076923076925,
+      "grad_norm": 0.17454926732818832,
+      "kl": 0.62109375,
+      "learning_rate": 9.230769230769232e-05,
+      "loss": 0.0028,
+      "step": 264
+    },
+    {
+      "clip_ratio": 0.002558046253398061,
+      "completion_length": 102.80357360839844,
+      "epoch": 0.3706293706293706,
+      "grad_norm": 0.18701691067416304,
+      "kl": 0.734375,
+      "learning_rate": 9.265734265734266e-05,
+      "loss": 0.0401,
+      "num_tokens": 2704225.0,
+      "reward": 2.1770143508911133,
+      "reward_std": 0.573693037033081,
+      "rewards/check_gptzero_func": 0.8035714030265808,
+      "rewards/check_perplexity_diff_func": 0.5357142686843872,
+      "rewards/check_winston_local_func": 0.8377286195755005,
+      "step": 265
+    },
+    {
+      "clip_ratio": 0.006838452070951462,
+      "epoch": 0.37202797202797205,
+      "grad_norm": 0.13942096443359253,
+      "kl": 0.71875,
+      "learning_rate": 9.300699300699301e-05,
+      "loss": 0.0354,
+      "step": 266
+    },
+    {
+      "clip_ratio": 0.00375689216889441,
+      "completion_length": 106.37500762939453,
+      "epoch": 0.3734265734265734,
+      "grad_norm": 0.25266992310982356,
+      "kl": 0.7421875,
+      "learning_rate": 9.335664335664336e-05,
+      "loss": -0.0156,
+      "num_tokens": 2724530.0,
+      "reward": 2.11887526512146,
+      "reward_std": 0.5366143584251404,
+      "rewards/check_gptzero_func": 0.875,
+      "rewards/check_perplexity_diff_func": 0.4642857015132904,
+      "rewards/check_winston_local_func": 0.7795897126197815,
+      "step": 267
+    },
+    {
+      "clip_ratio": 0.006593361962586641,
+      "epoch": 0.3748251748251748,
+      "grad_norm": 0.18497903526945206,
+      "kl": 0.7265625,
+      "learning_rate": 9.370629370629372e-05,
+      "loss": -0.0243,
+      "step": 268
+    },
+    {
+      "clip_ratio": 0.004569682292640209,
+      "completion_length": 124.39286041259766,
+      "epoch": 0.37622377622377623,
+      "grad_norm": 0.21331332062489702,
+      "kl": 0.953125,
+      "learning_rate": 9.405594405594406e-05,
+      "loss": 0.0346,
+      "num_tokens": 2746246.0,
+      "reward": 2.403602123260498,
+      "reward_std": 0.5621734857559204,
+      "rewards/check_gptzero_func": 0.7857142686843872,
+      "rewards/check_perplexity_diff_func": 0.75,
+      "rewards/check_winston_local_func": 0.8678879141807556,
+      "step": 269
+    },
+    {
+      "clip_ratio": 0.006006812676787376,
+      "epoch": 0.3776223776223776,
+      "grad_norm": 0.15420322253654764,
+      "kl": 0.8671875,
+      "learning_rate": 9.440559440559441e-05,
+      "loss": 0.0283,
+      "step": 270
+    },
+    {
+      "clip_ratio": 0.002397142816334963,
+      "completion_length": 102.41072082519531,
+      "epoch": 0.37902097902097903,
+      "grad_norm": 0.22567114655173642,
+      "kl": 0.609375,
+      "learning_rate": 9.475524475524477e-05,
+      "loss": 0.0231,
+      "num_tokens": 2765203.0,
+      "reward": 2.649076223373413,
+      "reward_std": 0.47521141171455383,
+      "rewards/check_gptzero_func": 0.875,
+      "rewards/check_perplexity_diff_func": 1.0,
+      "rewards/check_winston_local_func": 0.7740762829780579,
+      "step": 271
+    },
+    {
+      "clip_ratio": 0.007639365270733833,
+      "epoch": 0.3804195804195804,
+      "grad_norm": 0.18662930190408594,
+      "kl": 0.578125,
+      "learning_rate": 9.510489510489511e-05,
+      "loss": 0.0165,
+      "step": 272
+    },
+    {
+      "clip_ratio": 0.0010056205792352557,
+      "completion_length": 115.28572082519531,
+      "epoch": 0.38181818181818183,
+      "grad_norm": 0.14954423121719349,
+      "kl": 0.625,
+      "learning_rate": 9.545454545454546e-05,
+      "loss": 0.0237,
+      "num_tokens": 2785985.0,
+      "reward": 2.3541975021362305,
+      "reward_std": 0.5445494651794434,
+      "rewards/check_gptzero_func": 0.8392857313156128,
+      "rewards/check_perplexity_diff_func": 0.75,
+      "rewards/check_winston_local_func": 0.7649118304252625,
+      "step": 273
+    },
+    {
+      "clip_ratio": 0.003297280054539442,
+      "epoch": 0.3832167832167832,
+      "grad_norm": 0.11713806873969078,
+      "kl": 0.640625,
+      "learning_rate": 9.580419580419581e-05,
+      "loss": 0.0206,
+      "step": 274
+    },
+    {
+      "clip_ratio": 0.0016172031173482537,
+      "completion_length": 128.73214721679688,
+      "epoch": 0.38461538461538464,
+      "grad_norm": 0.155009547701029,
+      "kl": 0.51171875,
+      "learning_rate": 9.615384615384617e-05,
+      "loss": 0.0191,
+      "num_tokens": 2808414.0,
+      "reward": 2.376694679260254,
+      "reward_std": 0.33215656876564026,
+      "rewards/check_gptzero_func": 0.8392857313156128,
+      "rewards/check_perplexity_diff_func": 0.8571428656578064,
+      "rewards/check_winston_local_func": 0.6802659630775452,
+      "step": 275
+    },
+    {
+      "clip_ratio": 0.004294655751436949,
+      "epoch": 0.386013986013986,
+      "grad_norm": 0.11409067239204185,
+      "kl": 0.515625,
+      "learning_rate": 9.65034965034965e-05,
+      "loss": 0.0155,
+      "step": 276
+    },
+    {
+      "clip_ratio": 0.002925862092524767,
+      "completion_length": 116.41072082519531,
+      "epoch": 0.38741258741258744,
+      "grad_norm": 0.2764503095198425,
+      "kl": 0.578125,
+      "learning_rate": 9.685314685314686e-05,
+      "loss": 0.01,
+      "num_tokens": 2829285.0,
+      "reward": 2.3352210521698,
+      "reward_std": 0.7207943201065063,
+      "rewards/check_gptzero_func": 0.6607142686843872,
+      "rewards/check_perplexity_diff_func": 0.9642857313156128,
+      "rewards/check_winston_local_func": 0.7102211713790894,
+      "step": 277
+    },
+    {
+      "clip_ratio": 0.005557883996516466,
+      "epoch": 0.3888111888111888,
+      "grad_norm": 0.2140511595370052,
+      "kl": 0.5859375,
+      "learning_rate": 9.72027972027972e-05,
+      "loss": -0.0004,
+      "step": 278
+    },
+    {
+      "clip_ratio": 0.0029403052758425474,
+      "completion_length": 114.25000762939453,
+      "epoch": 0.3902097902097902,
+      "grad_norm": 0.21363596109825803,
+      "kl": 0.63671875,
+      "learning_rate": 9.755244755244755e-05,
+      "loss": 0.0259,
+      "num_tokens": 2849989.0,
+      "reward": 2.6570937633514404,
+      "reward_std": 0.5007120370864868,
+      "rewards/check_gptzero_func": 0.75,
+      "rewards/check_perplexity_diff_func": 1.2142857313156128,
+      "rewards/check_winston_local_func": 0.6928080320358276,
+      "step": 279
+    },
+    {
+      "clip_ratio": 0.00384224159643054,
+      "epoch": 0.3916083916083916,
+      "grad_norm": 0.15102706915256958,
+      "kl": 0.6875,
+      "learning_rate": 9.790209790209791e-05,
+      "loss": 0.0185,
+      "step": 280
+    },
+    {
+      "clip_ratio": 0.0027283646631985903,
+      "completion_length": 90.25000762939453,
+      "epoch": 0.393006993006993,
+      "grad_norm": 0.3585134615134873,
+      "kl": 0.73046875,
+      "learning_rate": 9.825174825174826e-05,
+      "loss": -0.006,
+      "num_tokens": 2867567.0,
+      "reward": 2.4528698921203613,
+      "reward_std": 0.6383817791938782,
+      "rewards/check_gptzero_func": 0.6071428656578064,
+      "rewards/check_perplexity_diff_func": 1.1428571939468384,
+      "rewards/check_winston_local_func": 0.7028695940971375,
+      "step": 281
+    },
+    {
+      "clip_ratio": 0.016295742243528366,
+      "epoch": 0.3944055944055944,
+      "grad_norm": 0.2358766583499307,
+      "kl": 0.73046875,
+      "learning_rate": 9.86013986013986e-05,
+      "loss": -0.0151,
+      "step": 282
+    },
+    {
+      "clip_ratio": 0.0025735762901604176,
+      "completion_length": 132.125,
+      "epoch": 0.3958041958041958,
+      "grad_norm": 0.12427534341566092,
+      "kl": 0.515625,
+      "learning_rate": 9.895104895104895e-05,
+      "loss": 0.006,
+      "num_tokens": 2890954.0,
+      "reward": 2.172079086303711,
+      "reward_std": 0.31105130910873413,
+      "rewards/check_gptzero_func": 0.6607142686843872,
+      "rewards/check_perplexity_diff_func": 0.7857142686843872,
+      "rewards/check_winston_local_func": 0.7256504893302917,
+      "step": 283
+    },
+    {
+      "clip_ratio": 0.008568070828914642,
+      "epoch": 0.3972027972027972,
+      "grad_norm": 0.10751149742872082,
+      "kl": 0.5234375,
+      "learning_rate": 9.930069930069931e-05,
+      "loss": 0.0038,
+      "step": 284
+    },
+    {
+      "clip_ratio": 0.0013303400482982397,
+      "completion_length": 115.50000762939453,
+      "epoch": 0.3986013986013986,
+      "grad_norm": 0.11155145645226237,
+      "kl": 0.427734375,
+      "learning_rate": 9.965034965034964e-05,
+      "loss": 0.0136,
+      "num_tokens": 2912320.0,
+      "reward": 2.499699354171753,
+      "reward_std": 0.22870054841041565,
+      "rewards/check_gptzero_func": 0.6071428656578064,
+      "rewards/check_perplexity_diff_func": 1.3214285373687744,
+      "rewards/check_winston_local_func": 0.5711276531219482,
+      "step": 285
+    },
+    {
+      "clip_ratio": 0.004690885543823242,
+      "epoch": 0.4,
+      "grad_norm": 0.08246302090980155,
+      "kl": 0.443359375,
+      "learning_rate": 0.0001,
+      "loss": 0.0117,
+      "step": 286
+    },
+    {
+      "clip_ratio": 0.0013457606546580791,
+      "completion_length": 153.60714721679688,
+      "epoch": 0.4013986013986014,
+      "grad_norm": 0.0959414180998777,
+      "kl": 0.458984375,
+      "learning_rate": 9.999996275889018e-05,
+      "loss": -0.0037,
+      "num_tokens": 2937286.0,
+      "reward": 2.60290789604187,
+      "reward_std": 0.2643657624721527,
+      "rewards/check_gptzero_func": 0.625,
+      "rewards/check_perplexity_diff_func": 1.3214285373687744,
+      "rewards/check_winston_local_func": 0.6564791798591614,
+      "step": 287
+    },
+    {
+      "clip_ratio": 0.0025680752005428076,
+      "epoch": 0.4027972027972028,
+      "grad_norm": 0.08295696905412306,
+      "kl": 0.453125,
+      "learning_rate": 9.999985103561615e-05,
+      "loss": -0.0053,
+      "step": 288
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 106.39286041259766,
+      "epoch": 0.4041958041958042,
+      "grad_norm": 0.12524695486629228,
+      "kl": 0.3671875,
+      "learning_rate": 9.999966483034437e-05,
+      "loss": -0.0031,
+      "num_tokens": 2957572.0,
+      "reward": 2.607825994491577,
+      "reward_std": 0.2113000899553299,
+      "rewards/check_gptzero_func": 0.5535714030265808,
+      "rewards/check_perplexity_diff_func": 1.4642857313156128,
+      "rewards/check_winston_local_func": 0.5899689793586731,
+      "step": 289
+    },
+    {
+      "clip_ratio": 0.0015743181575089693,
+      "epoch": 0.40559440559440557,
+      "grad_norm": 0.08958540244405325,
+      "kl": 0.36328125,
+      "learning_rate": 9.999940414335222e-05,
+      "loss": -0.0046,
+      "step": 290
+    },
+    {
+      "clip_ratio": 0.0017556955572217703,
+      "completion_length": 96.0714340209961,
+      "epoch": 0.406993006993007,
+      "grad_norm": 0.19185238369967877,
+      "kl": 2.109375,
+      "learning_rate": 9.999906897502803e-05,
+      "loss": 0.008,
+      "num_tokens": 2976076.0,
+      "reward": 2.6148557662963867,
+      "reward_std": 0.24822747707366943,
+      "rewards/check_gptzero_func": 0.5178571343421936,
+      "rewards/check_perplexity_diff_func": 1.4285714626312256,
+      "rewards/check_winston_local_func": 0.6684269309043884,
+      "step": 291
+    },
+    {
+      "clip_ratio": 0.00337741756811738,
+      "epoch": 0.4083916083916084,
+      "grad_norm": 0.08769186584743054,
+      "kl": 0.671875,
+      "learning_rate": 9.999865932587107e-05,
+      "loss": 0.0046,
+      "step": 292
+    },
+    {
+      "clip_ratio": 0.0012849332997575402,
+      "completion_length": 91.87500762939453,
+      "epoch": 0.4097902097902098,
+      "grad_norm": 0.13036265265723118,
+      "kl": 0.4453125,
+      "learning_rate": 9.999817519649158e-05,
+      "loss": 0.0127,
+      "num_tokens": 2994773.0,
+      "reward": 2.552300214767456,
+      "reward_std": 0.14500018954277039,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 1.5,
+      "rewards/check_winston_local_func": 0.6237286329269409,
+      "step": 293
+    },
+    {
+      "clip_ratio": 0.0027297178748995066,
+      "epoch": 0.4111888111888112,
+      "grad_norm": 0.0976002240930887,
+      "kl": 0.439453125,
+      "learning_rate": 9.99976165876107e-05,
+      "loss": 0.0101,
+      "step": 294
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 113.35714721679688,
+      "epoch": 0.4125874125874126,
+      "grad_norm": 0.02609480806956132,
+      "kl": 0.1591796875,
+      "learning_rate": 9.999698350006063e-05,
+      "loss": 0.0012,
+      "num_tokens": 3015993.0,
+      "reward": 2.874926805496216,
+      "reward_std": 0.0874238833785057,
+      "rewards/check_gptzero_func": 0.4821428656578064,
+      "rewards/check_perplexity_diff_func": 1.8928571939468384,
+      "rewards/check_winston_local_func": 0.4999266266822815,
+      "step": 295
+    },
+    {
+      "clip_ratio": 9.294544724980369e-05,
+      "epoch": 0.413986013986014,
+      "grad_norm": 0.02050573548257188,
+      "kl": 0.16015625,
+      "learning_rate": 9.999627593478442e-05,
+      "loss": 0.0011,
+      "step": 296
+    },
+    {
+      "clip_ratio": 0.00012664640962611884,
+      "completion_length": 98.5714340209961,
+      "epoch": 0.4153846153846154,
+      "grad_norm": 0.01680753751461254,
+      "kl": 0.201171875,
+      "learning_rate": 9.999549389283606e-05,
+      "loss": -0.0003,
+      "num_tokens": 3035155.0,
+      "reward": 3.0224545001983643,
+      "reward_std": 0.016296973451972008,
+      "rewards/check_gptzero_func": 0.4107142984867096,
+      "rewards/check_perplexity_diff_func": 1.9642857313156128,
+      "rewards/check_winston_local_func": 0.647454559803009,
+      "step": 297
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.4167832167832168,
+      "grad_norm": 0.012884367061070623,
+      "kl": 0.203125,
+      "learning_rate": 9.999463737538053e-05,
+      "loss": -0.0003,
+      "step": 298
+    },
+    {
+      "clip_ratio": 0.0003541912592481822,
+      "completion_length": 95.76786041259766,
+      "epoch": 0.41818181818181815,
+      "grad_norm": 0.04526732622569909,
+      "kl": 0.291015625,
+      "learning_rate": 9.999370638369377e-05,
+      "loss": 0.0002,
+      "num_tokens": 3054174.0,
+      "reward": 2.3303959369659424,
+      "reward_std": 0.08971218019723892,
+      "rewards/check_gptzero_func": 0.3035714328289032,
+      "rewards/check_perplexity_diff_func": 1.6071428060531616,
+      "rewards/check_winston_local_func": 0.4196813404560089,
+      "step": 299
+    },
+    {
+      "clip_ratio": 0.00011806376278400421,
+      "epoch": 0.4195804195804196,
+      "grad_norm": 0.03862986955837794,
+      "kl": 0.296875,
+      "learning_rate": 9.999270091916257e-05,
+      "loss": -0.0002,
+      "step": 300
+    },
+    {
+      "clip_ratio": 0.0012040403671562672,
+      "completion_length": 101.73214721679688,
+      "epoch": 0.42097902097902096,
+      "grad_norm": 0.12616717395678975,
+      "kl": 0.35546875,
+      "learning_rate": 9.999162098328474e-05,
+      "loss": 0.0033,
+      "num_tokens": 3073401.0,
+      "reward": 2.8153316974639893,
+      "reward_std": 0.18918544054031372,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 1.7142857313156128,
+      "rewards/check_winston_local_func": 0.6724745631217957,
+      "step": 301
+    },
+    {
+      "clip_ratio": 0.0013847867958247662,
+      "epoch": 0.4223776223776224,
+      "grad_norm": 0.06929991057149441,
+      "kl": 0.412109375,
+      "learning_rate": 9.999046657766903e-05,
+      "loss": 0.0012,
+      "step": 302
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 81.3214340209961,
+      "epoch": 0.42377622377622376,
+      "grad_norm": 0.06730614630696258,
+      "kl": 0.2099609375,
+      "learning_rate": 9.998923770403505e-05,
+      "loss": -0.0019,
+      "num_tokens": 3090625.0,
+      "reward": 2.802220106124878,
+      "reward_std": 0.03781326860189438,
+      "rewards/check_gptzero_func": 0.375,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.42722010612487793,
+      "step": 303
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.4251748251748252,
+      "grad_norm": 0.022175315902565728,
+      "kl": 0.2109375,
+      "learning_rate": 9.99879343642134e-05,
+      "loss": -0.0024,
+      "step": 304
+    },
+    {
+      "clip_ratio": 0.00013683634460903704,
+      "completion_length": 105.53572082519531,
+      "epoch": 0.42657342657342656,
+      "grad_norm": 0.04867704636152166,
+      "kl": 0.173828125,
+      "learning_rate": 9.998655656014561e-05,
+      "loss": 0.0117,
+      "num_tokens": 3110877.0,
+      "reward": 2.6083197593688965,
+      "reward_std": 0.13502858579158783,
+      "rewards/check_gptzero_func": 0.3571428656578064,
+      "rewards/check_perplexity_diff_func": 1.8928571939468384,
+      "rewards/check_winston_local_func": 0.35831958055496216,
+      "step": 305
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.427972027972028,
+      "grad_norm": 0.043987709695075865,
+      "kl": 0.1728515625,
+      "learning_rate": 9.99851042938841e-05,
+      "loss": 0.0113,
+      "step": 306
+    },
+    {
+      "clip_ratio": 0.0004155792994424701,
+      "completion_length": 124.03572082519531,
+      "epoch": 0.42937062937062936,
+      "grad_norm": 0.03684760048760383,
+      "kl": 0.2080078125,
+      "learning_rate": 9.998357756759222e-05,
+      "loss": -0.0258,
+      "num_tokens": 3134097.0,
+      "reward": 2.7439539432525635,
+      "reward_std": 0.11075940728187561,
+      "rewards/check_gptzero_func": 0.4642857015132904,
+      "rewards/check_perplexity_diff_func": 1.7857142686843872,
+      "rewards/check_winston_local_func": 0.4939536452293396,
+      "step": 307
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.4307692307692308,
+      "grad_norm": 0.013648247572273129,
+      "kl": 0.20703125,
+      "learning_rate": 9.998197638354428e-05,
+      "loss": -0.026,
+      "step": 308
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 89.12500762939453,
+      "epoch": 0.43216783216783217,
+      "grad_norm": 0.02465210474674629,
+      "kl": 0.443359375,
+      "learning_rate": 9.998030074412545e-05,
+      "loss": 0.0014,
+      "num_tokens": 3152358.0,
+      "reward": 2.420307159423828,
+      "reward_std": 0.11704181134700775,
+      "rewards/check_gptzero_func": 0.2321428507566452,
+      "rewards/check_perplexity_diff_func": 1.75,
+      "rewards/check_winston_local_func": 0.4381641745567322,
+      "step": 309
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.43356643356643354,
+      "grad_norm": 0.021184461362474653,
+      "kl": 0.48828125,
+      "learning_rate": 9.997855065183184e-05,
+      "loss": 0.0014,
+      "step": 310
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 115.73214721679688,
+      "epoch": 0.43496503496503497,
+      "grad_norm": 0.0049134080725686995,
+      "kl": 0.162109375,
+      "learning_rate": 9.997672610927047e-05,
+      "loss": -0.0047,
+      "num_tokens": 3173997.0,
+      "reward": 2.61786150932312,
+      "reward_std": 0.013112460263073444,
+      "rewards/check_gptzero_func": 0.375,
+      "rewards/check_perplexity_diff_func": 1.8214285373687744,
+      "rewards/check_winston_local_func": 0.4214327931404114,
+      "step": 311
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.43636363636363634,
+      "grad_norm": 0.00595537475288129,
+      "kl": 0.173828125,
+      "learning_rate": 9.997482711915927e-05,
+      "loss": -0.0047,
+      "step": 312
+    },
+    {
+      "clip_ratio": 0.0001360544265480712,
+      "completion_length": 90.42857360839844,
+      "epoch": 0.43776223776223777,
+      "grad_norm": 0.007409687643468917,
+      "kl": 0.2275390625,
+      "learning_rate": 9.997285368432703e-05,
+      "loss": 0.0006,
+      "num_tokens": 3192417.0,
+      "reward": 2.682650566101074,
+      "reward_std": 0.02417343109846115,
+      "rewards/check_gptzero_func": 0.375,
+      "rewards/check_perplexity_diff_func": 1.8214285373687744,
+      "rewards/check_winston_local_func": 0.4862218499183655,
+      "step": 313
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.43916083916083914,
+      "grad_norm": 0.007129921917131849,
+      "kl": 0.2275390625,
+      "learning_rate": 9.997080580771349e-05,
+      "loss": 0.0006,
+      "step": 314
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 83.14286041259766,
+      "epoch": 0.4405594405594406,
+      "grad_norm": 1.3606081670938062e-05,
+      "kl": 0.2255859375,
+      "learning_rate": 9.996868349236927e-05,
+      "loss": 0.0002,
+      "num_tokens": 3209875.0,
+      "reward": 2.671295404434204,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3571428656578064,
+      "rewards/check_perplexity_diff_func": 1.8571428060531616,
+      "rewards/check_winston_local_func": 0.45700937509536743,
+      "step": 315
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.44195804195804195,
+      "grad_norm": 1.3537787639105168e-05,
+      "kl": 0.2255859375,
+      "learning_rate": 9.996648674145583e-05,
+      "loss": 0.0002,
+      "step": 316
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 126.21429443359375,
+      "epoch": 0.4433566433566434,
+      "grad_norm": 0.00991208286904361,
+      "kl": 0.15625,
+      "learning_rate": 9.996421555824558e-05,
+      "loss": 0.0103,
+      "num_tokens": 3232963.0,
+      "reward": 2.6654393672943115,
+      "reward_std": 0.0725974440574646,
+      "rewards/check_gptzero_func": 0.5357142686843872,
+      "rewards/check_perplexity_diff_func": 1.6071428060531616,
+      "rewards/check_winston_local_func": 0.5225821733474731,
+      "step": 317
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.44475524475524475,
+      "grad_norm": 0.010677817125212348,
+      "kl": 0.154296875,
+      "learning_rate": 9.996186994612176e-05,
+      "loss": 0.0102,
+      "step": 318
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 103.4464340209961,
+      "epoch": 0.4461538461538462,
+      "grad_norm": 5.275581365835224e-05,
+      "kl": 0.16796875,
+      "learning_rate": 9.995944990857849e-05,
+      "loss": 0.0002,
+      "num_tokens": 3252898.0,
+      "reward": 3.0420854091644287,
+      "reward_std": 0.00039901022682897747,
+      "rewards/check_gptzero_func": 0.5714285969734192,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.5420854687690735,
+      "step": 319
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.44755244755244755,
+      "grad_norm": 6.10762506510971e-05,
+      "kl": 0.16796875,
+      "learning_rate": 9.995695544922077e-05,
+      "loss": 0.0002,
+      "step": 320
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 102.28572082519531,
+      "epoch": 0.4489510489510489,
+      "grad_norm": 0.010696195320132886,
+      "kl": 0.1669921875,
+      "learning_rate": 9.995438657176448e-05,
+      "loss": -0.0177,
+      "num_tokens": 3272574.0,
+      "reward": 2.7592146396636963,
+      "reward_std": 0.03917063772678375,
+      "rewards/check_gptzero_func": 0.3571428656578064,
+      "rewards/check_perplexity_diff_func": 1.9642857313156128,
+      "rewards/check_winston_local_func": 0.437785804271698,
+      "step": 321
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.45034965034965035,
+      "grad_norm": 0.01083841197384593,
+      "kl": 0.1669921875,
+      "learning_rate": 9.995174328003631e-05,
+      "loss": -0.0177,
+      "step": 322
+    },
+    {
+      "clip_ratio": 8.473140769638121e-05,
+      "completion_length": 117.41072082519531,
+      "epoch": 0.45174825174825173,
+      "grad_norm": 0.004522641361347185,
+      "kl": 0.1640625,
+      "learning_rate": 9.994902557797382e-05,
+      "loss": -0.0002,
+      "num_tokens": 3294025.0,
+      "reward": 2.92221736907959,
+      "reward_std": 0.025547320023179054,
+      "rewards/check_gptzero_func": 0.5535714030265808,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.36864587664604187,
+      "step": 323
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.45314685314685316,
+      "grad_norm": 0.004737684573095642,
+      "kl": 0.1650390625,
+      "learning_rate": 9.994623346962544e-05,
+      "loss": -0.0002,
+      "step": 324
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 95.26786041259766,
+      "epoch": 0.45454545454545453,
+      "grad_norm": 3.372088632491828e-06,
+      "kl": 0.1796875,
+      "learning_rate": 9.99433669591504e-05,
+      "loss": 0.0002,
+      "num_tokens": 3312902.0,
+      "reward": 2.595499038696289,
+      "reward_std": 4.388691013446078e-05,
+      "rewards/check_gptzero_func": 0.25,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.34549909830093384,
+      "step": 325
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.45594405594405596,
+      "grad_norm": 3.349405120445447e-06,
+      "kl": 0.1796875,
+      "learning_rate": 9.994042605081879e-05,
+      "loss": 0.0002,
+      "step": 326
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 97.80357360839844,
+      "epoch": 0.45734265734265733,
+      "grad_norm": 0.010175912326793437,
+      "kl": 0.177734375,
+      "learning_rate": 9.993741074901153e-05,
+      "loss": 0.0003,
+      "num_tokens": 3331815.0,
+      "reward": 2.496483325958252,
+      "reward_std": 0.04960284009575844,
+      "rewards/check_gptzero_func": 0.2142857164144516,
+      "rewards/check_perplexity_diff_func": 1.9642857313156128,
+      "rewards/check_winston_local_func": 0.31791168451309204,
+      "step": 327
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.45874125874125876,
+      "grad_norm": 0.01017493586702021,
+      "kl": 0.177734375,
+      "learning_rate": 9.993432105822034e-05,
+      "loss": 0.0003,
+      "step": 328
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 96.30357360839844,
+      "epoch": 0.46013986013986014,
+      "grad_norm": 0.0165654292436505,
+      "kl": 0.1572265625,
+      "learning_rate": 9.993115698304774e-05,
+      "loss": 0.0003,
+      "num_tokens": 3351210.0,
+      "reward": 2.8074374198913574,
+      "reward_std": 0.07505974173545837,
+      "rewards/check_gptzero_func": 0.5178571343421936,
+      "rewards/check_perplexity_diff_func": 1.8571428060531616,
+      "rewards/check_winston_local_func": 0.4324372708797455,
+      "step": 329
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.46153846153846156,
+      "grad_norm": 0.014211382809754432,
+      "kl": 0.16015625,
+      "learning_rate": 9.992791852820709e-05,
+      "loss": 0.0003,
+      "step": 330
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 86.1964340209961,
+      "epoch": 0.46293706293706294,
+      "grad_norm": 0.0580909978469744,
+      "kl": 0.2119140625,
+      "learning_rate": 9.992460569852256e-05,
+      "loss": 0.0013,
+      "num_tokens": 3369025.0,
+      "reward": 2.986760377883911,
+      "reward_std": 0.10425443947315216,
+      "rewards/check_gptzero_func": 0.4642857015132904,
+      "rewards/check_perplexity_diff_func": 1.8571428060531616,
+      "rewards/check_winston_local_func": 0.6653318405151367,
+      "step": 331
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.4643356643356643,
+      "grad_norm": 0.0012843504186470114,
+      "kl": 0.2109375,
+      "learning_rate": 9.992121849892904e-05,
+      "loss": 0.0012,
+      "step": 332
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 113.42857360839844,
+      "epoch": 0.46573426573426574,
+      "grad_norm": 0.03054414776989089,
+      "kl": 0.1494140625,
+      "learning_rate": 9.99177569344723e-05,
+      "loss": -0.0,
+      "num_tokens": 3390093.0,
+      "reward": 2.5890934467315674,
+      "reward_std": 0.1023683100938797,
+      "rewards/check_gptzero_func": 0.25,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.41052165627479553,
+      "step": 333
+    },
+    {
+      "clip_ratio": 0.00022784233442507684,
+      "epoch": 0.4671328671328671,
+      "grad_norm": 0.006141110067038759,
+      "kl": 0.150390625,
+      "learning_rate": 9.99142210103088e-05,
+      "loss": -0.0001,
+      "step": 334
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 109.00000762939453,
+      "epoch": 0.46853146853146854,
+      "grad_norm": 0.011702000814979486,
+      "kl": 0.158203125,
+      "learning_rate": 9.991061073170585e-05,
+      "loss": 0.0001,
+      "num_tokens": 3410511.0,
+      "reward": 2.601181745529175,
+      "reward_std": 0.04859553650021553,
+      "rewards/check_gptzero_func": 0.2857142984867096,
+      "rewards/check_perplexity_diff_func": 1.8928571939468384,
+      "rewards/check_winston_local_func": 0.42261001467704773,
+      "step": 335
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.4699300699300699,
+      "grad_norm": 0.004380253375033637,
+      "kl": 0.158203125,
+      "learning_rate": 9.990692610404145e-05,
+      "loss": 0.0001,
+      "step": 336
+    },
+    {
+      "clip_ratio": 0.00015908369095996022,
+      "completion_length": 124.37500762939453,
+      "epoch": 0.47132867132867134,
+      "grad_norm": 0.011635533606345248,
+      "kl": 0.1318359375,
+      "learning_rate": 9.99031671328044e-05,
+      "loss": 0.0002,
+      "num_tokens": 3433250.0,
+      "reward": 2.719287395477295,
+      "reward_std": 0.05448899790644646,
+      "rewards/check_gptzero_func": 0.5,
+      "rewards/check_perplexity_diff_func": 1.75,
+      "rewards/check_winston_local_func": 0.4692873954772949,
+      "step": 337
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.4727272727272727,
+      "grad_norm": 0.009393665715625162,
+      "kl": 0.1318359375,
+      "learning_rate": 9.989933382359422e-05,
+      "loss": 0.0001,
+      "step": 338
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 110.8214340209961,
+      "epoch": 0.47412587412587415,
+      "grad_norm": 6.954019260271828e-07,
+      "kl": 0.15625,
+      "learning_rate": 9.98954261821212e-05,
+      "loss": 0.0002,
+      "num_tokens": 3454202.0,
+      "reward": 2.5842461585998535,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 1.7857142686843872,
+      "rewards/check_winston_local_func": 0.40567442774772644,
+      "step": 339
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.4755244755244755,
+      "grad_norm": 6.417345751201852e-07,
+      "kl": 0.15625,
+      "learning_rate": 9.98914442142063e-05,
+      "loss": 0.0002,
+      "step": 340
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 76.55357360839844,
+      "epoch": 0.47692307692307695,
+      "grad_norm": 0.008322229771779063,
+      "kl": 0.263671875,
+      "learning_rate": 9.988738792578126e-05,
+      "loss": 0.0005,
+      "num_tokens": 3471041.0,
+      "reward": 2.916919231414795,
+      "reward_std": 0.05185602605342865,
+      "rewards/check_gptzero_func": 0.3214285671710968,
+      "rewards/check_perplexity_diff_func": 1.9642857313156128,
+      "rewards/check_winston_local_func": 0.631205141544342,
+      "step": 341
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.4783216783216783,
+      "grad_norm": 0.010116392524351476,
+      "kl": 0.263671875,
+      "learning_rate": 9.988325732288851e-05,
+      "loss": 0.0004,
+      "step": 342
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 122.21429443359375,
+      "epoch": 0.4797202797202797,
+      "grad_norm": 1.4137707959903356e-05,
+      "kl": 0.134765625,
+      "learning_rate": 9.987905241168117e-05,
+      "loss": 0.0001,
+      "num_tokens": 3493483.0,
+      "reward": 2.7157416343688965,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 1.8571428060531616,
+      "rewards/check_winston_local_func": 0.46574172377586365,
+      "step": 343
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.4811188811188811,
+      "grad_norm": 1.3729789968238485e-05,
+      "kl": 0.134765625,
+      "learning_rate": 9.987477319842307e-05,
+      "loss": 0.0001,
+      "step": 344
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 93.53572082519531,
+      "epoch": 0.4825174825174825,
+      "grad_norm": 7.468121696306164e-07,
+      "kl": 0.1845703125,
+      "learning_rate": 9.987041968948869e-05,
+      "loss": 0.0002,
+      "num_tokens": 3512009.0,
+      "reward": 2.5505218505859375,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.1785714328289032,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.44337886571884155,
+      "step": 345
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.48391608391608393,
+      "grad_norm": 7.137709602226676e-07,
+      "kl": 0.1845703125,
+      "learning_rate": 9.986599189136325e-05,
+      "loss": 0.0002,
+      "step": 346
+    },
+    {
+      "clip_ratio": 0.00031959093757905066,
+      "completion_length": 102.58928680419922,
+      "epoch": 0.4853146853146853,
+      "grad_norm": 0.04707770778424934,
+      "kl": 0.1982421875,
+      "learning_rate": 9.986148981064258e-05,
+      "loss": 0.0004,
+      "num_tokens": 3531574.0,
+      "reward": 2.7341363430023193,
+      "reward_std": 0.0231277234852314,
+      "rewards/check_gptzero_func": 0.4107142984867096,
+      "rewards/check_perplexity_diff_func": 1.8928571939468384,
+      "rewards/check_winston_local_func": 0.430564820766449,
+      "step": 347
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.48671328671328673,
+      "grad_norm": 0.020919799034125892,
+      "kl": 0.2060546875,
+      "learning_rate": 9.985691345403316e-05,
+      "loss": 0.0002,
+      "step": 348
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 127.67857360839844,
+      "epoch": 0.4881118881118881,
+      "grad_norm": 6.517786151476589e-05,
+      "kl": 0.1328125,
+      "learning_rate": 9.985226282835216e-05,
+      "loss": 0.0001,
+      "num_tokens": 3554440.0,
+      "reward": 2.926541566848755,
+      "reward_std": 0.0005768488626927137,
+      "rewards/check_gptzero_func": 0.5357142686843872,
+      "rewards/check_perplexity_diff_func": 1.8571428060531616,
+      "rewards/check_winston_local_func": 0.5336844325065613,
+      "step": 349
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.48951048951048953,
+      "grad_norm": 6.370254129427888e-05,
+      "kl": 0.1328125,
+      "learning_rate": 9.984753794052735e-05,
+      "loss": 0.0001,
+      "step": 350
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 98.75000762939453,
+      "epoch": 0.4909090909090909,
+      "grad_norm": 2.085540464425244e-05,
+      "kl": 0.205078125,
+      "learning_rate": 9.984273879759713e-05,
+      "loss": 0.0002,
+      "num_tokens": 3573976.0,
+      "reward": 2.8204286098480225,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.4989997446537018,
+      "step": 351
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.49230769230769234,
+      "grad_norm": 2.009578048518353e-05,
+      "kl": 0.205078125,
+      "learning_rate": 9.983786540671051e-05,
+      "loss": 0.0002,
+      "step": 352
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 85.21428680419922,
+      "epoch": 0.4937062937062937,
+      "grad_norm": 2.1499404490472988e-06,
+      "kl": 0.1826171875,
+      "learning_rate": 9.983291777512711e-05,
+      "loss": 0.0002,
+      "num_tokens": 3591644.0,
+      "reward": 2.8764142990112305,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.44784265756607056,
+      "step": 353
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.4951048951048951,
+      "grad_norm": 2.160124410750654e-06,
+      "kl": 0.1826171875,
+      "learning_rate": 9.982789591021715e-05,
+      "loss": 0.0002,
+      "step": 354
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 112.00000762939453,
+      "epoch": 0.4965034965034965,
+      "grad_norm": 0.00014814082463704463,
+      "kl": 0.1669921875,
+      "learning_rate": 9.982279981946143e-05,
+      "loss": 0.0002,
+      "num_tokens": 3612596.0,
+      "reward": 2.8398191928863525,
+      "reward_std": 0.0006443771999329329,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.5183902382850647,
+      "step": 355
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.4979020979020979,
+      "grad_norm": 0.00014606763289714295,
+      "kl": 0.1669921875,
+      "learning_rate": 9.98176295104513e-05,
+      "loss": 0.0002,
+      "step": 356
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 88.50000762939453,
+      "epoch": 0.4993006993006993,
+      "grad_norm": 3.33550016470454e-07,
+      "kl": 0.1669921875,
+      "learning_rate": 9.98123849908887e-05,
+      "loss": 0.0002,
+      "num_tokens": 3630862.0,
+      "reward": 2.655492067337036,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3214285671710968,
+      "rewards/check_perplexity_diff_func": 1.8571428060531616,
+      "rewards/check_winston_local_func": 0.47692039608955383,
+      "step": 357
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5006993006993007,
+      "grad_norm": 3.3462192763241657e-07,
+      "kl": 0.1669921875,
+      "learning_rate": 9.980706626858607e-05,
+      "loss": 0.0002,
+      "step": 358
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 126.39286041259766,
+      "epoch": 0.5020979020979021,
+      "grad_norm": 3.4860786526471938e-06,
+      "kl": 0.1416015625,
+      "learning_rate": 9.980167335146642e-05,
+      "loss": 0.0001,
+      "num_tokens": 3653738.0,
+      "reward": 2.765826463699341,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.5,
+      "rewards/check_perplexity_diff_func": 1.8571428060531616,
+      "rewards/check_winston_local_func": 0.4086834490299225,
+      "step": 359
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5034965034965035,
+      "grad_norm": 3.71619859471226e-06,
+      "kl": 0.1416015625,
+      "learning_rate": 9.979620624756329e-05,
+      "loss": 0.0001,
+      "step": 360
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 129.92857360839844,
+      "epoch": 0.5048951048951049,
+      "grad_norm": 8.195642286342138e-07,
+      "kl": 0.1376953125,
+      "learning_rate": 9.979066496502074e-05,
+      "loss": 0.0001,
+      "num_tokens": 3677274.0,
+      "reward": 2.8696835041046143,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.5357142686843872,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.4053979218006134,
+      "step": 361
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5062937062937063,
+      "grad_norm": 8.760444459307852e-07,
+      "kl": 0.1376953125,
+      "learning_rate": 9.978504951209327e-05,
+      "loss": 0.0001,
+      "step": 362
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 118.0714340209961,
+      "epoch": 0.5076923076923077,
+      "grad_norm": 1.595860947799347e-06,
+      "kl": 0.162109375,
+      "learning_rate": 9.977935989714595e-05,
+      "loss": 0.0002,
+      "num_tokens": 3698896.0,
+      "reward": 2.6664552688598633,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.2857142984867096,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.45216941833496094,
+      "step": 363
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.509090909090909,
+      "grad_norm": 1.6260194368178069e-06,
+      "kl": 0.162109375,
+      "learning_rate": 9.977359612865423e-05,
+      "loss": 0.0002,
+      "step": 364
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 85.96428680419922,
+      "epoch": 0.5104895104895105,
+      "grad_norm": 4.357301744161908e-06,
+      "kl": 0.2138671875,
+      "learning_rate": 9.976775821520412e-05,
+      "loss": 0.0002,
+      "num_tokens": 3716536.0,
+      "reward": 2.8877956867218018,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.566366970539093,
+      "step": 365
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5118881118881119,
+      "grad_norm": 4.353902612581284e-06,
+      "kl": 0.2138671875,
+      "learning_rate": 9.976184616549203e-05,
+      "loss": 0.0002,
+      "step": 366
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 116.96429443359375,
+      "epoch": 0.5132867132867133,
+      "grad_norm": 1.5148419096170459e-06,
+      "kl": 0.146484375,
+      "learning_rate": 9.97558599883248e-05,
+      "loss": 0.0001,
+      "num_tokens": 3737950.0,
+      "reward": 2.687312602996826,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.36588403582572937,
+      "step": 367
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5146853146853146,
+      "grad_norm": 1.4470246412763422e-06,
+      "kl": 0.146484375,
+      "learning_rate": 9.97497996926197e-05,
+      "loss": 0.0001,
+      "step": 368
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 94.83928680419922,
+      "epoch": 0.5160839160839161,
+      "grad_norm": 0.0004647591690002515,
+      "kl": 0.2138671875,
+      "learning_rate": 9.974366528740441e-05,
+      "loss": 0.0002,
+      "num_tokens": 3756935.0,
+      "reward": 2.634523391723633,
+      "reward_std": 0.0005499019753187895,
+      "rewards/check_gptzero_func": 0.2857142984867096,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.4202377200126648,
+      "step": 369
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5174825174825175,
+      "grad_norm": 0.0004885464229234562,
+      "kl": 0.2138671875,
+      "learning_rate": 9.973745678181705e-05,
+      "loss": 0.0002,
+      "step": 370
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 100.17857360839844,
+      "epoch": 0.5188811188811189,
+      "grad_norm": 3.0418176986251483e-06,
+      "kl": 0.1640625,
+      "learning_rate": 9.973117418510605e-05,
+      "loss": 0.0002,
+      "num_tokens": 3776087.0,
+      "reward": 2.5216457843780518,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.2857142984867096,
+      "rewards/check_perplexity_diff_func": 1.8571428060531616,
+      "rewards/check_winston_local_func": 0.3787885308265686,
+      "step": 371
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5202797202797202,
+      "grad_norm": 3.0723635480561103e-06,
+      "kl": 0.1640625,
+      "learning_rate": 9.972481750663026e-05,
+      "loss": 0.0002,
+      "step": 372
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 89.66072082519531,
+      "epoch": 0.5216783216783217,
+      "grad_norm": 0.008590656277289149,
+      "kl": 0.40234375,
+      "learning_rate": 9.971838675585888e-05,
+      "loss": 0.0007,
+      "num_tokens": 3794550.0,
+      "reward": 2.5267839431762695,
+      "reward_std": 0.05061079189181328,
+      "rewards/check_gptzero_func": 0.3214285671710968,
+      "rewards/check_perplexity_diff_func": 1.8214285373687744,
+      "rewards/check_winston_local_func": 0.383926659822464,
+      "step": 373
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5230769230769231,
+      "grad_norm": 0.003770576536622864,
+      "kl": 0.4296875,
+      "learning_rate": 9.97118819423714e-05,
+      "loss": 0.0007,
+      "step": 374
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 80.17857360839844,
+      "epoch": 0.5244755244755245,
+      "grad_norm": 3.560813902520289e-07,
+      "kl": 0.2216796875,
+      "learning_rate": 9.970530307585773e-05,
+      "loss": 0.0002,
+      "num_tokens": 3811756.0,
+      "reward": 2.883481740951538,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.5263389348983765,
+      "step": 375
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5258741258741259,
+      "grad_norm": 3.5463275242856317e-07,
+      "kl": 0.2216796875,
+      "learning_rate": 9.9698650166118e-05,
+      "loss": 0.0002,
+      "step": 376
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 86.75000762939453,
+      "epoch": 0.5272727272727272,
+      "grad_norm": 4.776682547251814e-07,
+      "kl": 0.205078125,
+      "learning_rate": 9.969192322306271e-05,
+      "loss": 0.0002,
+      "num_tokens": 3829616.0,
+      "reward": 2.9529268741607666,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4642857015132904,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.5600695610046387,
+      "step": 377
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5286713286713287,
+      "grad_norm": 4.804658823710114e-07,
+      "kl": 0.205078125,
+      "learning_rate": 9.96851222567126e-05,
+      "loss": 0.0002,
+      "step": 378
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 117.89286041259766,
+      "epoch": 0.5300699300699301,
+      "grad_norm": 9.143102926453055e-06,
+      "kl": 0.158203125,
+      "learning_rate": 9.96782472771987e-05,
+      "loss": 0.0002,
+      "num_tokens": 3851170.0,
+      "reward": 2.7617037296295166,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.40456071496009827,
+      "step": 379
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5314685314685315,
+      "grad_norm": 9.41817239216292e-06,
+      "kl": 0.158203125,
+      "learning_rate": 9.967129829476228e-05,
+      "loss": 0.0002,
+      "step": 380
+    },
+    {
+      "clip_ratio": 0.000286861730273813,
+      "completion_length": 80.8214340209961,
+      "epoch": 0.5328671328671328,
+      "grad_norm": 0.03337399461775748,
+      "kl": 0.224609375,
+      "learning_rate": 9.966427531975483e-05,
+      "loss": -0.0402,
+      "num_tokens": 3868508.0,
+      "reward": 2.820319652557373,
+      "reward_std": 0.0946396142244339,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 1.8571428060531616,
+      "rewards/check_winston_local_func": 0.5346053838729858,
+      "step": 381
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5342657342657343,
+      "grad_norm": 0.00034351439093039966,
+      "kl": 0.224609375,
+      "learning_rate": 9.965717836263812e-05,
+      "loss": -0.0403,
+      "step": 382
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 95.03572082519531,
+      "epoch": 0.5356643356643357,
+      "grad_norm": 3.3945579906577137e-07,
+      "kl": 0.216796875,
+      "learning_rate": 9.965000743398408e-05,
+      "loss": 0.0002,
+      "num_tokens": 3887044.0,
+      "reward": 2.8529021739959717,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.4243304431438446,
+      "step": 383
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5370629370629371,
+      "grad_norm": 2.8390305863820515e-07,
+      "kl": 0.216796875,
+      "learning_rate": 9.964276254447484e-05,
+      "loss": 0.0002,
+      "step": 384
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 140.1428680419922,
+      "epoch": 0.5384615384615384,
+      "grad_norm": 8.521451122880038e-07,
+      "kl": 0.134765625,
+      "learning_rate": 9.96354437049027e-05,
+      "loss": 0.0001,
+      "num_tokens": 3911264.0,
+      "reward": 2.4770619869232178,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.2857142984867096,
+      "rewards/check_perplexity_diff_func": 1.8571428060531616,
+      "rewards/check_winston_local_func": 0.33420488238334656,
+      "step": 385
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5398601398601398,
+      "grad_norm": 6.753306754213034e-07,
+      "kl": 0.134765625,
+      "learning_rate": 9.962805092617016e-05,
+      "loss": 0.0001,
+      "step": 386
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 97.64286041259766,
+      "epoch": 0.5412587412587413,
+      "grad_norm": 0.08784199050747857,
+      "kl": 2.28125,
+      "learning_rate": 9.962058421928979e-05,
+      "loss": 0.0023,
+      "num_tokens": 3930478.0,
+      "reward": 2.897810459136963,
+      "reward_std": 0.002231778111308813,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.4692386984825134,
+      "step": 387
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5426573426573427,
+      "grad_norm": 0.03620083942117924,
+      "kl": 0.98828125,
+      "learning_rate": 9.961304359538437e-05,
+      "loss": 0.001,
+      "step": 388
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 102.21428680419922,
+      "epoch": 0.544055944055944,
+      "grad_norm": 0.025447754609912356,
+      "kl": 0.1806640625,
+      "learning_rate": 9.96054290656867e-05,
+      "loss": 0.0002,
+      "num_tokens": 3950362.0,
+      "reward": 2.8110532760620117,
+      "reward_std": 0.04895726963877678,
+      "rewards/check_gptzero_func": 0.5357142686843872,
+      "rewards/check_perplexity_diff_func": 1.8214285373687744,
+      "rewards/check_winston_local_func": 0.45391014218330383,
+      "step": 389
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5454545454545454,
+      "grad_norm": 5.1964811610508515e-05,
+      "kl": 0.1806640625,
+      "learning_rate": 9.959774064153977e-05,
+      "loss": 0.0001,
+      "step": 390
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 103.5714340209961,
+      "epoch": 0.5468531468531469,
+      "grad_norm": 3.816406536402076e-07,
+      "kl": 0.173828125,
+      "learning_rate": 9.958997833439657e-05,
+      "loss": 0.0002,
+      "num_tokens": 3970164.0,
+      "reward": 2.5441431999206543,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3214285671710968,
+      "rewards/check_perplexity_diff_func": 1.8571428060531616,
+      "rewards/check_winston_local_func": 0.3655718266963959,
+      "step": 391
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5482517482517483,
+      "grad_norm": 4.199010025491445e-07,
+      "kl": 0.173828125,
+      "learning_rate": 9.958214215582018e-05,
+      "loss": 0.0002,
+      "step": 392
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 126.8214340209961,
+      "epoch": 0.5496503496503496,
+      "grad_norm": 6.309179959293235e-06,
+      "kl": 0.1533203125,
+      "learning_rate": 9.957423211748374e-05,
+      "loss": 0.0002,
+      "num_tokens": 3992994.0,
+      "reward": 2.7432734966278076,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.42184481024742126,
+      "step": 393
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.551048951048951,
+      "grad_norm": 5.547540566387994e-06,
+      "kl": 0.1533203125,
+      "learning_rate": 9.956624823117036e-05,
+      "loss": 0.0002,
+      "step": 394
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 125.62500762939453,
+      "epoch": 0.5524475524475524,
+      "grad_norm": 0.007987942179452279,
+      "kl": 0.142578125,
+      "learning_rate": 9.955819050877321e-05,
+      "loss": 0.0001,
+      "num_tokens": 4015669.0,
+      "reward": 3.091567039489746,
+      "reward_std": 0.018998777493834496,
+      "rewards/check_gptzero_func": 0.5535714030265808,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.6094242334365845,
+      "step": 395
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5538461538461539,
+      "grad_norm": 0.0034758785486018436,
+      "kl": 0.142578125,
+      "learning_rate": 9.955005896229543e-05,
+      "loss": 0.0,
+      "step": 396
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 96.75000762939453,
+      "epoch": 0.5552447552447553,
+      "grad_norm": 0.011495641299277456,
+      "kl": 0.283203125,
+      "learning_rate": 9.954185360385013e-05,
+      "loss": 0.0006,
+      "num_tokens": 4035041.0,
+      "reward": 2.950004816055298,
+      "reward_std": 0.01496803853660822,
+      "rewards/check_gptzero_func": 0.4821428656578064,
+      "rewards/check_perplexity_diff_func": 1.8928571939468384,
+      "rewards/check_winston_local_func": 0.5750047564506531,
+      "step": 397
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5566433566433566,
+      "grad_norm": 0.010082500068390739,
+      "kl": 0.283203125,
+      "learning_rate": 9.953357444566039e-05,
+      "loss": 0.0006,
+      "step": 398
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 110.8214340209961,
+      "epoch": 0.558041958041958,
+      "grad_norm": 8.489376613416e-06,
+      "kl": 0.1787109375,
+      "learning_rate": 9.952522150005919e-05,
+      "loss": 0.0002,
+      "num_tokens": 4055855.0,
+      "reward": 2.913240909576416,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4642857015132904,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.5203836560249329,
+      "step": 399
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5594405594405595,
+      "grad_norm": 8.766485288799225e-06,
+      "kl": 0.1787109375,
+      "learning_rate": 9.951679477948947e-05,
+      "loss": 0.0002,
+      "step": 400
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 105.46428680419922,
+      "epoch": 0.5608391608391609,
+      "grad_norm": 8.652849000461684e-06,
+      "kl": 0.150390625,
+      "learning_rate": 9.9508294296504e-05,
+      "loss": 0.0002,
+      "num_tokens": 4076051.0,
+      "reward": 2.871354818344116,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.44278329610824585,
+      "step": 401
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5622377622377622,
+      "grad_norm": 8.18246821524729e-06,
+      "kl": 0.150390625,
+      "learning_rate": 9.949972006376556e-05,
+      "loss": 0.0002,
+      "step": 402
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 75.25,
+      "epoch": 0.5636363636363636,
+      "grad_norm": 7.351342876079361e-07,
+      "kl": 0.1953125,
+      "learning_rate": 9.949107209404665e-05,
+      "loss": 0.0002,
+      "num_tokens": 4092527.0,
+      "reward": 2.9322519302368164,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3571428656578064,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.5751089453697205,
+      "step": 403
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5650349650349651,
+      "grad_norm": 7.136344464618312e-07,
+      "kl": 0.1953125,
+      "learning_rate": 9.948235040022966e-05,
+      "loss": 0.0002,
+      "step": 404
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 108.08928680419922,
+      "epoch": 0.5664335664335665,
+      "grad_norm": 0.007956878335048736,
+      "kl": 0.169921875,
+      "learning_rate": 9.947355499530683e-05,
+      "loss": 0.0049,
+      "num_tokens": 4113282.0,
+      "reward": 3.1798436641693115,
+      "reward_std": 0.06495096534490585,
+      "rewards/check_gptzero_func": 0.6785714030265808,
+      "rewards/check_perplexity_diff_func": 1.9642857313156128,
+      "rewards/check_winston_local_func": 0.5369864702224731,
+      "step": 405
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5678321678321678,
+      "grad_norm": 0.006640329404323574,
+      "kl": 0.1728515625,
+      "learning_rate": 9.946468589238021e-05,
+      "loss": 0.0049,
+      "step": 406
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 121.39286041259766,
+      "epoch": 0.5692307692307692,
+      "grad_norm": 1.5620797981026132e-06,
+      "kl": 0.1396484375,
+      "learning_rate": 9.945574310466159e-05,
+      "loss": 0.0001,
+      "num_tokens": 4135506.0,
+      "reward": 2.573127269744873,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.2857142984867096,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.3588416278362274,
+      "step": 407
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5706293706293706,
+      "grad_norm": 1.51593967204227e-06,
+      "kl": 0.1396484375,
+      "learning_rate": 9.944672664547252e-05,
+      "loss": 0.0001,
+      "step": 408
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 107.8214340209961,
+      "epoch": 0.5720279720279721,
+      "grad_norm": 0.0007775272826913423,
+      "kl": 0.17578125,
+      "learning_rate": 9.943763652824436e-05,
+      "loss": 0.0002,
+      "num_tokens": 4156060.0,
+      "reward": 2.55350661277771,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3571428656578064,
+      "rewards/check_perplexity_diff_func": 1.7142857313156128,
+      "rewards/check_winston_local_func": 0.48207801580429077,
+      "step": 409
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5734265734265734,
+      "grad_norm": 0.0008914625567854122,
+      "kl": 0.1767578125,
+      "learning_rate": 9.942847276651811e-05,
+      "loss": 0.0002,
+      "step": 410
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 101.62500762939453,
+      "epoch": 0.5748251748251748,
+      "grad_norm": 4.9539161428254635e-05,
+      "kl": 0.1474609375,
+      "learning_rate": 9.941923537394456e-05,
+      "loss": 0.0001,
+      "num_tokens": 4176019.0,
+      "reward": 2.8906424045562744,
+      "reward_std": 0.00012528452498372644,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.5334992408752441,
+      "step": 411
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5762237762237762,
+      "grad_norm": 5.844422779033556e-05,
+      "kl": 0.1474609375,
+      "learning_rate": 9.940992436428409e-05,
+      "loss": 0.0001,
+      "step": 412
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 105.28572082519531,
+      "epoch": 0.5776223776223777,
+      "grad_norm": 0.0006008878616346907,
+      "kl": 0.2216796875,
+      "learning_rate": 9.940053975140684e-05,
+      "loss": 0.0002,
+      "num_tokens": 4196667.0,
+      "reward": 2.943253755569458,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.5357142686843872,
+      "rewards/check_perplexity_diff_func": 1.8571428060531616,
+      "rewards/check_winston_local_func": 0.5503963828086853,
+      "step": 413
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.579020979020979,
+      "grad_norm": 0.0003365492175069851,
+      "kl": 0.220703125,
+      "learning_rate": 9.939108154929252e-05,
+      "loss": 0.0002,
+      "step": 414
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 114.60714721679688,
+      "epoch": 0.5804195804195804,
+      "grad_norm": 0.008680207230787525,
+      "kl": 0.171875,
+      "learning_rate": 9.938154977203049e-05,
+      "loss": -0.0005,
+      "num_tokens": 4217727.0,
+      "reward": 2.7773444652557373,
+      "reward_std": 0.051937274634838104,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 1.8214285373687744,
+      "rewards/check_winston_local_func": 0.5273441076278687,
+      "step": 415
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5818181818181818,
+      "grad_norm": 0.00033864647249595475,
+      "kl": 0.171875,
+      "learning_rate": 9.937194443381972e-05,
+      "loss": -0.0005,
+      "step": 416
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 100.60714721679688,
+      "epoch": 0.5832167832167832,
+      "grad_norm": 4.017405087628847e-06,
+      "kl": 0.193359375,
+      "learning_rate": 9.936226554896875e-05,
+      "loss": 0.0002,
+      "num_tokens": 4237189.0,
+      "reward": 2.7684807777404785,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.44705215096473694,
+      "step": 417
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5846153846153846,
+      "grad_norm": 3.880139616189675e-06,
+      "kl": 0.193359375,
+      "learning_rate": 9.935251313189564e-05,
+      "loss": 0.0002,
+      "step": 418
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 107.75000762939453,
+      "epoch": 0.586013986013986,
+      "grad_norm": 0.00032680437267053027,
+      "kl": 0.1826171875,
+      "learning_rate": 9.934268719712807e-05,
+      "loss": 0.0002,
+      "num_tokens": 4258213.0,
+      "reward": 2.577641487121582,
+      "reward_std": 0.0015505586052313447,
+      "rewards/check_gptzero_func": 0.2857142984867096,
+      "rewards/check_perplexity_diff_func": 1.8571428060531616,
+      "rewards/check_winston_local_func": 0.4347843825817108,
+      "step": 419
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5874125874125874,
+      "grad_norm": 0.0003238465648274268,
+      "kl": 0.1826171875,
+      "learning_rate": 9.933278775930317e-05,
+      "loss": 0.0002,
+      "step": 420
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 108.67857360839844,
+      "epoch": 0.5888111888111888,
+      "grad_norm": 0.015591092439371165,
+      "kl": 0.16015625,
+      "learning_rate": 9.932281483316758e-05,
+      "loss": 0.0004,
+      "num_tokens": 4278847.0,
+      "reward": 2.615447998046875,
+      "reward_std": 0.05428782477974892,
+      "rewards/check_gptzero_func": 0.3214285671710968,
+      "rewards/check_perplexity_diff_func": 1.8214285373687744,
+      "rewards/check_winston_local_func": 0.47259071469306946,
+      "step": 421
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5902097902097903,
+      "grad_norm": 0.00016911180587449477,
+      "kl": 0.16015625,
+      "learning_rate": 9.931276843357742e-05,
+      "loss": 0.0003,
+      "step": 422
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 111.16072082519531,
+      "epoch": 0.5916083916083916,
+      "grad_norm": 1.6847767080478394e-05,
+      "kl": 0.1630859375,
+      "learning_rate": 9.930264857549825e-05,
+      "loss": 0.0002,
+      "num_tokens": 4300096.0,
+      "reward": 2.857067823410034,
+      "reward_std": 0.00101040443405509,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 1.8571428060531616,
+      "rewards/check_winston_local_func": 0.6070676445960999,
+      "step": 423
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.593006993006993,
+      "grad_norm": 2.453025078669727e-05,
+      "kl": 0.1630859375,
+      "learning_rate": 9.929245527400503e-05,
+      "loss": 0.0002,
+      "step": 424
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 102.50000762939453,
+      "epoch": 0.5944055944055944,
+      "grad_norm": 0.0015183266942611472,
+      "kl": 0.173828125,
+      "learning_rate": 9.928218854428221e-05,
+      "loss": 0.0002,
+      "num_tokens": 4320064.0,
+      "reward": 2.6430578231811523,
+      "reward_std": 5.020291791879572e-05,
+      "rewards/check_gptzero_func": 0.3571428656578064,
+      "rewards/check_perplexity_diff_func": 1.7857142686843872,
+      "rewards/check_winston_local_func": 0.5002006888389587,
+      "step": 425
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5958041958041959,
+      "grad_norm": 0.0013596061771866538,
+      "kl": 0.1708984375,
+      "learning_rate": 9.927184840162354e-05,
+      "loss": 0.0002,
+      "step": 426
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 110.66072082519531,
+      "epoch": 0.5972027972027972,
+      "grad_norm": 0.12076121626694686,
+      "kl": 0.189453125,
+      "learning_rate": 9.926143486143214e-05,
+      "loss": -0.001,
+      "num_tokens": 4341145.0,
+      "reward": 2.6579177379608154,
+      "reward_std": 0.05154726281762123,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 1.8214285373687744,
+      "rewards/check_winston_local_func": 0.4079175889492035,
+      "step": 427
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5986013986013986,
+      "grad_norm": 0.0008289218843548021,
+      "kl": 0.19140625,
+      "learning_rate": 9.92509479392205e-05,
+      "loss": -0.0013,
+      "step": 428
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 89.67857360839844,
+      "epoch": 0.6,
+      "grad_norm": 0.005379412774279031,
+      "kl": 0.1787109375,
+      "learning_rate": 9.924038765061042e-05,
+      "loss": 0.0001,
+      "num_tokens": 4359277.0,
+      "reward": 2.633777141571045,
+      "reward_std": 0.05039219558238983,
+      "rewards/check_gptzero_func": 0.2857142984867096,
+      "rewards/check_perplexity_diff_func": 1.9642857313156128,
+      "rewards/check_winston_local_func": 0.38377735018730164,
+      "step": 429
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6013986013986014,
+      "grad_norm": 0.010439668945298157,
+      "kl": 0.1787109375,
+      "learning_rate": 9.922975401133293e-05,
+      "loss": 0.0002,
+      "step": 430
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 102.67857360839844,
+      "epoch": 0.6027972027972028,
+      "grad_norm": 0.06317955454404899,
+      "kl": 0.201171875,
+      "learning_rate": 9.92190470372284e-05,
+      "loss": 0.0068,
+      "num_tokens": 4379109.0,
+      "reward": 2.6650938987731934,
+      "reward_std": 0.05930115655064583,
+      "rewards/check_gptzero_func": 0.25,
+      "rewards/check_perplexity_diff_func": 1.9642857313156128,
+      "rewards/check_winston_local_func": 0.4508081078529358,
+      "step": 431
+    },
+    {
+      "clip_ratio": 0.0002587991766631603,
+      "epoch": 0.6041958041958042,
+      "grad_norm": 0.05122842630547611,
+      "kl": 0.1865234375,
+      "learning_rate": 9.920826674424642e-05,
+      "loss": 0.0063,
+      "step": 432
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 80.35714721679688,
+      "epoch": 0.6055944055944056,
+      "grad_norm": 7.162658137250354e-07,
+      "kl": 0.1875,
+      "learning_rate": 9.919741314844577e-05,
+      "loss": 0.0002,
+      "num_tokens": 4396693.0,
+      "reward": 3.0637807846069336,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.5357142686843872,
+      "rewards/check_perplexity_diff_func": 1.8571428060531616,
+      "rewards/check_winston_local_func": 0.6709235310554504,
+      "step": 433
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.606993006993007,
+      "grad_norm": 8.451796450955314e-07,
+      "kl": 0.1875,
+      "learning_rate": 9.918648626599447e-05,
+      "loss": 0.0002,
+      "step": 434
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 124.98214721679688,
+      "epoch": 0.6083916083916084,
+      "grad_norm": 0.016099687268680437,
+      "kl": 0.1455078125,
+      "learning_rate": 9.91754861131697e-05,
+      "loss": -0.0001,
+      "num_tokens": 4418950.0,
+      "reward": 2.94807505607605,
+      "reward_std": 0.050821539014577866,
+      "rewards/check_gptzero_func": 0.5357142686843872,
+      "rewards/check_perplexity_diff_func": 1.8928571939468384,
+      "rewards/check_winston_local_func": 0.5195035338401794,
+      "step": 435
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6097902097902098,
+      "grad_norm": 0.014390989758817311,
+      "kl": 0.1455078125,
+      "learning_rate": 9.916441270635772e-05,
+      "loss": -0.0001,
+      "step": 436
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 106.64286041259766,
+      "epoch": 0.6111888111888112,
+      "grad_norm": 1.968332582426962e-06,
+      "kl": 0.14453125,
+      "learning_rate": 9.915326606205404e-05,
+      "loss": 0.0001,
+      "num_tokens": 4439018.0,
+      "reward": 2.77024507522583,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.4131018817424774,
+      "step": 437
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6125874125874126,
+      "grad_norm": 1.8961430119120163e-06,
+      "kl": 0.14453125,
+      "learning_rate": 9.914204619686314e-05,
+      "loss": 0.0001,
+      "step": 438
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 102.21428680419922,
+      "epoch": 0.6139860139860139,
+      "grad_norm": 0.000668351376631818,
+      "kl": 0.1728515625,
+      "learning_rate": 9.913075312749866e-05,
+      "loss": 0.0002,
+      "num_tokens": 4458806.0,
+      "reward": 2.8182172775268555,
+      "reward_std": 0.0025242711417376995,
+      "rewards/check_gptzero_func": 0.5357142686843872,
+      "rewards/check_perplexity_diff_func": 1.8571428060531616,
+      "rewards/check_winston_local_func": 0.42535990476608276,
+      "step": 439
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6153846153846154,
+      "grad_norm": 0.0006751756209535994,
+      "kl": 0.1728515625,
+      "learning_rate": 9.911938687078324e-05,
+      "loss": 0.0002,
+      "step": 440
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 81.25,
+      "epoch": 0.6167832167832168,
+      "grad_norm": 0.007904133813646444,
+      "kl": 0.240234375,
+      "learning_rate": 9.910794744364857e-05,
+      "loss": -0.0001,
+      "num_tokens": 4475982.0,
+      "reward": 2.8269970417022705,
+      "reward_std": 0.05014092102646828,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 1.8928571939468384,
+      "rewards/check_winston_local_func": 0.5412828326225281,
+      "step": 441
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6181818181818182,
+      "grad_norm": 0.00827597712943806,
+      "kl": 0.240234375,
+      "learning_rate": 9.909643486313533e-05,
+      "loss": -0.0001,
+      "step": 442
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 113.00000762939453,
+      "epoch": 0.6195804195804195,
+      "grad_norm": 6.016032313066646e-06,
+      "kl": 0.16796875,
+      "learning_rate": 9.908484914639318e-05,
+      "loss": 0.0002,
+      "num_tokens": 4497200.0,
+      "reward": 2.9243876934051514,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.5315303206443787,
+      "step": 443
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.620979020979021,
+      "grad_norm": 7.198804696892823e-06,
+      "kl": 0.16796875,
+      "learning_rate": 9.90731903106807e-05,
+      "loss": 0.0002,
+      "step": 444
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 129.375,
+      "epoch": 0.6223776223776224,
+      "grad_norm": 0.02075011817403209,
+      "kl": 0.158203125,
+      "learning_rate": 9.90614583733654e-05,
+      "loss": -0.0001,
+      "num_tokens": 4519807.0,
+      "reward": 2.866011619567871,
+      "reward_std": 0.10101933032274246,
+      "rewards/check_gptzero_func": 0.4642857015132904,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.47315436601638794,
+      "step": 445
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6237762237762238,
+      "grad_norm": 0.0034619333956659956,
+      "kl": 0.1591796875,
+      "learning_rate": 9.904965335192373e-05,
+      "loss": -0.0002,
+      "step": 446
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 98.92857360839844,
+      "epoch": 0.6251748251748251,
+      "grad_norm": 0.010057352537761423,
+      "kl": 0.212890625,
+      "learning_rate": 9.903777526394094e-05,
+      "loss": -0.0086,
+      "num_tokens": 4539039.0,
+      "reward": 2.980358362197876,
+      "reward_std": 0.05100167542695999,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 1.9642857313156128,
+      "rewards/check_winston_local_func": 0.587501049041748,
+      "step": 447
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6265734265734266,
+      "grad_norm": 0.006958390301115274,
+      "kl": 0.21875,
+      "learning_rate": 9.90258241271112e-05,
+      "loss": -0.0087,
+      "step": 448
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 111.71429443359375,
+      "epoch": 0.627972027972028,
+      "grad_norm": 1.6536770511152556e-06,
+      "kl": 0.142578125,
+      "learning_rate": 9.901379995923738e-05,
+      "loss": 0.0001,
+      "num_tokens": 4559675.0,
+      "reward": 2.681791067123413,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.2857142984867096,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.4675052762031555,
+      "step": 449
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6293706293706294,
+      "grad_norm": 1.6773170531853214e-06,
+      "kl": 0.142578125,
+      "learning_rate": 9.900170277823129e-05,
+      "loss": 0.0001,
+      "step": 450
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 107.35714721679688,
+      "epoch": 0.6307692307692307,
+      "grad_norm": 2.0958023033337787e-06,
+      "kl": 0.2109375,
+      "learning_rate": 9.898953260211338e-05,
+      "loss": 0.0002,
+      "num_tokens": 4579995.0,
+      "reward": 2.7400293350219727,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.4186006188392639,
+      "step": 451
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6321678321678321,
+      "grad_norm": 1.982971590769392e-06,
+      "kl": 0.2109375,
+      "learning_rate": 9.897728944901292e-05,
+      "loss": 0.0002,
+      "step": 452
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 118.26786041259766,
+      "epoch": 0.6335664335664336,
+      "grad_norm": 0.009450375112303724,
+      "kl": 0.1416015625,
+      "learning_rate": 9.896497333716783e-05,
+      "loss": -0.0051,
+      "num_tokens": 4602048.0,
+      "reward": 2.640944004058838,
+      "reward_std": 0.043768420815467834,
+      "rewards/check_gptzero_func": 0.5357142686843872,
+      "rewards/check_perplexity_diff_func": 1.5357142686843872,
+      "rewards/check_winston_local_func": 0.5695151090621948,
+      "step": 453
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.634965034965035,
+      "grad_norm": 0.0017713963386990425,
+      "kl": 0.1416015625,
+      "learning_rate": 9.895258428492475e-05,
+      "loss": -0.0051,
+      "step": 454
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 98.42857360839844,
+      "epoch": 0.6363636363636364,
+      "grad_norm": 0.014431904681184264,
+      "kl": 0.1943359375,
+      "learning_rate": 9.894012231073894e-05,
+      "loss": 0.0002,
+      "num_tokens": 4621654.0,
+      "reward": 2.7700040340423584,
+      "reward_std": 0.03728308901190758,
+      "rewards/check_gptzero_func": 0.4107142984867096,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.43071839213371277,
+      "step": 455
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6377622377622377,
+      "grad_norm": 0.008239966094546339,
+      "kl": 0.1943359375,
+      "learning_rate": 9.892758743317434e-05,
+      "loss": 0.0002,
+      "step": 456
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 90.50000762939453,
+      "epoch": 0.6391608391608392,
+      "grad_norm": 2.579298856684113e-06,
+      "kl": 0.1923828125,
+      "learning_rate": 9.891497967090344e-05,
+      "loss": 0.0002,
+      "num_tokens": 4640140.0,
+      "reward": 2.6383018493652344,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.25,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.4597306251525879,
+      "step": 457
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6405594405594406,
+      "grad_norm": 2.6625863420088633e-06,
+      "kl": 0.1923828125,
+      "learning_rate": 9.890229904270731e-05,
+      "loss": 0.0002,
+      "step": 458
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 113.03572082519531,
+      "epoch": 0.641958041958042,
+      "grad_norm": 1.7932858894376123e-05,
+      "kl": 0.1748046875,
+      "learning_rate": 9.888954556747563e-05,
+      "loss": 0.0002,
+      "num_tokens": 4661276.0,
+      "reward": 2.74210524559021,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3571428656578064,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.3849623203277588,
+      "step": 459
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6433566433566433,
+      "grad_norm": 1.7905922488331998e-05,
+      "kl": 0.1748046875,
+      "learning_rate": 9.887671926420648e-05,
+      "loss": 0.0002,
+      "step": 460
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 112.42857360839844,
+      "epoch": 0.6447552447552447,
+      "grad_norm": 0.014077582219597583,
+      "kl": 0.173828125,
+      "learning_rate": 9.886382015200652e-05,
+      "loss": 0.0002,
+      "num_tokens": 4682450.0,
+      "reward": 2.8680295944213867,
+      "reward_std": 0.05049533396959305,
+      "rewards/check_gptzero_func": 0.4642857015132904,
+      "rewards/check_perplexity_diff_func": 1.9642857313156128,
+      "rewards/check_winston_local_func": 0.43945807218551636,
+      "step": 461
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6461538461538462,
+      "grad_norm": 0.010796297977350395,
+      "kl": 0.173828125,
+      "learning_rate": 9.885084825009086e-05,
+      "loss": 0.0001,
+      "step": 462
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 105.14286041259766,
+      "epoch": 0.6475524475524476,
+      "grad_norm": 0.02303930816984521,
+      "kl": 0.17578125,
+      "learning_rate": 9.883780357778299e-05,
+      "loss": 0.0003,
+      "num_tokens": 4702710.0,
+      "reward": 2.885434150695801,
+      "reward_std": 0.06019994616508484,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 1.75,
+      "rewards/check_winston_local_func": 0.7068625092506409,
+      "step": 463
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6489510489510489,
+      "grad_norm": 0.0015050223064125838,
+      "kl": 0.1767578125,
+      "learning_rate": 9.882468615451484e-05,
+      "loss": 0.0002,
+      "step": 464
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 127.5714340209961,
+      "epoch": 0.6503496503496503,
+      "grad_norm": 3.0460982652821775e-06,
+      "kl": 0.1650390625,
+      "learning_rate": 9.881149599982671e-05,
+      "loss": 0.0002,
+      "num_tokens": 4725654.0,
+      "reward": 2.407341957092285,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.2857142984867096,
+      "rewards/check_perplexity_diff_func": 1.8571428060531616,
+      "rewards/check_winston_local_func": 0.2644847333431244,
+      "step": 465
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6517482517482518,
+      "grad_norm": 2.8166666383288673e-06,
+      "kl": 0.1650390625,
+      "learning_rate": 9.879823313336722e-05,
+      "loss": 0.0002,
+      "step": 466
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 122.16072082519531,
+      "epoch": 0.6531468531468532,
+      "grad_norm": 0.0001021700279542065,
+      "kl": 0.140625,
+      "learning_rate": 9.878489757489337e-05,
+      "loss": 0.0001,
+      "num_tokens": 4748375.0,
+      "reward": 2.6119117736816406,
+      "reward_std": 2.7804879209725186e-05,
+      "rewards/check_gptzero_func": 0.5357142686843872,
+      "rewards/check_perplexity_diff_func": 1.5714285373687744,
+      "rewards/check_winston_local_func": 0.5047687888145447,
+      "step": 467
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6545454545454545,
+      "grad_norm": 9.33433928584176e-05,
+      "kl": 0.140625,
+      "learning_rate": 9.877148934427037e-05,
+      "loss": 0.0001,
+      "step": 468
+    },
+    {
+      "clip_ratio": 0.0006711409660056233,
+      "completion_length": 95.64286041259766,
+      "epoch": 0.6559440559440559,
+      "grad_norm": 0.0004684591425632982,
+      "kl": 0.171875,
+      "learning_rate": 9.87580084614717e-05,
+      "loss": 0.0002,
+      "num_tokens": 4767439.0,
+      "reward": 2.3873257637023926,
+      "reward_std": 0.0017833748133853078,
+      "rewards/check_gptzero_func": 0.25,
+      "rewards/check_perplexity_diff_func": 1.7142857313156128,
+      "rewards/check_winston_local_func": 0.42303988337516785,
+      "step": 469
+    },
+    {
+      "clip_ratio": 0.0004793864209204912,
+      "epoch": 0.6573426573426573,
+      "grad_norm": 0.00047799981287926325,
+      "kl": 0.171875,
+      "learning_rate": 9.874445494657911e-05,
+      "loss": 0.0002,
+      "step": 470
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 79.39286041259766,
+      "epoch": 0.6587412587412588,
+      "grad_norm": 0.011441113949782547,
+      "kl": 0.216796875,
+      "learning_rate": 9.873082881978251e-05,
+      "loss": -0.0064,
+      "num_tokens": 4784715.0,
+      "reward": 3.037661075592041,
+      "reward_std": 0.023328183218836784,
+      "rewards/check_gptzero_func": 0.4821428656578064,
+      "rewards/check_perplexity_diff_func": 1.9642857313156128,
+      "rewards/check_winston_local_func": 0.5912323594093323,
+      "step": 471
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6601398601398601,
+      "grad_norm": 0.008433986682608557,
+      "kl": 0.216796875,
+      "learning_rate": 9.871713010137997e-05,
+      "loss": -0.0064,
+      "step": 472
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 119.42857360839844,
+      "epoch": 0.6615384615384615,
+      "grad_norm": 7.638914800437301e-06,
+      "kl": 0.1435546875,
+      "learning_rate": 9.870335881177774e-05,
+      "loss": 0.0001,
+      "num_tokens": 4806207.0,
+      "reward": 2.9706830978393555,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4642857015132904,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.5063972473144531,
+      "step": 473
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6629370629370629,
+      "grad_norm": 7.505737589947172e-06,
+      "kl": 0.1435546875,
+      "learning_rate": 9.868951497149011e-05,
+      "loss": 0.0001,
+      "step": 474
+    },
+    {
+      "clip_ratio": 0.00019073051225859672,
+      "completion_length": 89.87500762939453,
+      "epoch": 0.6643356643356644,
+      "grad_norm": 0.03662860311910791,
+      "kl": 0.2060546875,
+      "learning_rate": 9.86755986011395e-05,
+      "loss": 0.0065,
+      "num_tokens": 4824332.0,
+      "reward": 2.643022060394287,
+      "reward_std": 0.04776443541049957,
+      "rewards/check_gptzero_func": 0.3214285671710968,
+      "rewards/check_perplexity_diff_func": 1.8928571939468384,
+      "rewards/check_winston_local_func": 0.42873620986938477,
+      "step": 475
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6657342657342658,
+      "grad_norm": 0.03126231173733144,
+      "kl": 0.2060546875,
+      "learning_rate": 9.866160972145634e-05,
+      "loss": 0.0062,
+      "step": 476
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 123.10714721679688,
+      "epoch": 0.6671328671328671,
+      "grad_norm": 0.0002559996362709361,
+      "kl": 0.1357421875,
+      "learning_rate": 9.864754835327909e-05,
+      "loss": 0.0001,
+      "num_tokens": 4847094.0,
+      "reward": 2.767104387283325,
+      "reward_std": 0.0007265785825438797,
+      "rewards/check_gptzero_func": 0.5,
+      "rewards/check_perplexity_diff_func": 1.7142857313156128,
+      "rewards/check_winston_local_func": 0.5528185963630676,
+      "step": 477
+    },
+    {
+      "clip_ratio": 8.97343925316818e-05,
+      "epoch": 0.6685314685314685,
+      "grad_norm": 0.00042633622080268514,
+      "kl": 0.1357421875,
+      "learning_rate": 9.86334145175542e-05,
+      "loss": 0.0001,
+      "step": 478
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 96.67857360839844,
+      "epoch": 0.66993006993007,
+      "grad_norm": 0.0008693689003033054,
+      "kl": 0.16015625,
+      "learning_rate": 9.861920823533606e-05,
+      "loss": 0.0002,
+      "num_tokens": 4866394.0,
+      "reward": 2.5761330127716064,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3571428656578064,
+      "rewards/check_perplexity_diff_func": 1.7142857313156128,
+      "rewards/check_winston_local_func": 0.504704475402832,
+      "step": 479
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6713286713286714,
+      "grad_norm": 0.0008435837049153919,
+      "kl": 0.16015625,
+      "learning_rate": 9.860492952778696e-05,
+      "loss": 0.0002,
+      "step": 480
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 94.39286041259766,
+      "epoch": 0.6727272727272727,
+      "grad_norm": 3.349433978828943e-06,
+      "kl": 0.177734375,
+      "learning_rate": 9.859057841617709e-05,
+      "loss": 0.0002,
+      "num_tokens": 4885318.0,
+      "reward": 2.8612334728240967,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3571428656578064,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.5755191445350647,
+      "step": 481
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6741258741258741,
+      "grad_norm": 3.5095106255593373e-06,
+      "kl": 0.177734375,
+      "learning_rate": 9.857615492188452e-05,
+      "loss": 0.0002,
+      "step": 482
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 87.98214721679688,
+      "epoch": 0.6755244755244755,
+      "grad_norm": 0.00016582191324414645,
+      "kl": 0.208984375,
+      "learning_rate": 9.856165906639513e-05,
+      "loss": 0.0002,
+      "num_tokens": 4903401.0,
+      "reward": 2.797102928161621,
+      "reward_std": 0.0008326892857439816,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.4756740629673004,
+      "step": 483
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.676923076923077,
+      "grad_norm": 0.00017592290286556547,
+      "kl": 0.208984375,
+      "learning_rate": 9.85470908713026e-05,
+      "loss": 0.0002,
+      "step": 484
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 122.14286041259766,
+      "epoch": 0.6783216783216783,
+      "grad_norm": 3.840262302649335e-05,
+      "kl": 0.140625,
+      "learning_rate": 9.853245035830834e-05,
+      "loss": 0.0001,
+      "num_tokens": 4925777.0,
+      "reward": 2.80380916595459,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4642857015132904,
+      "rewards/check_perplexity_diff_func": 1.8571428060531616,
+      "rewards/check_winston_local_func": 0.4823804199695587,
+      "step": 485
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6797202797202797,
+      "grad_norm": 3.855750270963149e-05,
+      "kl": 0.140625,
+      "learning_rate": 9.851773754922152e-05,
+      "loss": 0.0001,
+      "step": 486
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 110.5714340209961,
+      "epoch": 0.6811188811188811,
+      "grad_norm": 1.0000521495087345e-06,
+      "kl": 0.1572265625,
+      "learning_rate": 9.850295246595898e-05,
+      "loss": 0.0002,
+      "num_tokens": 4946515.0,
+      "reward": 2.9591448307037354,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.5,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.5305731296539307,
+      "step": 487
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6825174825174826,
+      "grad_norm": 9.579988148024946e-07,
+      "kl": 0.1572265625,
+      "learning_rate": 9.848809513054523e-05,
+      "loss": 0.0002,
+      "step": 488
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 104.67857360839844,
+      "epoch": 0.6839160839160839,
+      "grad_norm": 4.4290499511336513e-07,
+      "kl": 0.1728515625,
+      "learning_rate": 9.847316556511245e-05,
+      "loss": 0.0002,
+      "num_tokens": 4966537.0,
+      "reward": 3.0352413654327393,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4642857015132904,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.5709553956985474,
+      "step": 489
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6853146853146853,
+      "grad_norm": 4.228282290104698e-07,
+      "kl": 0.1728515625,
+      "learning_rate": 9.845816379190036e-05,
+      "loss": 0.0002,
+      "step": 490
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 106.92857360839844,
+      "epoch": 0.6867132867132867,
+      "grad_norm": 7.161264654013381e-07,
+      "kl": 0.1865234375,
+      "learning_rate": 9.844308983325625e-05,
+      "loss": 0.0002,
+      "num_tokens": 4986837.0,
+      "reward": 2.8011527061462402,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4642857015132904,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.40829533338546753,
+      "step": 491
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6881118881118881,
+      "grad_norm": 6.606409864369399e-07,
+      "kl": 0.1865234375,
+      "learning_rate": 9.842794371163501e-05,
+      "loss": 0.0002,
+      "step": 492
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 127.3214340209961,
+      "epoch": 0.6895104895104895,
+      "grad_norm": 3.5056088482557826e-05,
+      "kl": 0.11767578125,
+      "learning_rate": 9.841272544959892e-05,
+      "loss": 0.0001,
+      "num_tokens": 5009783.0,
+      "reward": 2.6788315773010254,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.5,
+      "rewards/check_perplexity_diff_func": 1.7857142686843872,
+      "rewards/check_winston_local_func": 0.393117219209671,
+      "step": 493
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6909090909090909,
+      "grad_norm": 3.5363245519599165e-05,
+      "kl": 0.11767578125,
+      "learning_rate": 9.839743506981782e-05,
+      "loss": 0.0001,
+      "step": 494
+    },
+    {
+      "clip_ratio": 0.0004093328316230327,
+      "completion_length": 99.08928680419922,
+      "epoch": 0.6923076923076923,
+      "grad_norm": 0.0028759704706919223,
+      "kl": 0.1650390625,
+      "learning_rate": 9.838207259506891e-05,
+      "loss": 0.0002,
+      "num_tokens": 5029238.0,
+      "reward": 2.5216493606567383,
+      "reward_std": 0.0011310166446492076,
+      "rewards/check_gptzero_func": 0.3571428656578064,
+      "rewards/check_perplexity_diff_func": 1.7142857313156128,
+      "rewards/check_winston_local_func": 0.45022064447402954,
+      "step": 495
+    },
+    {
+      "clip_ratio": 0.0004093328316230327,
+      "epoch": 0.6937062937062937,
+      "grad_norm": 0.0035050811312865673,
+      "kl": 0.1650390625,
+      "learning_rate": 9.836663804823683e-05,
+      "loss": 0.0002,
+      "step": 496
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 124.85714721679688,
+      "epoch": 0.6951048951048951,
+      "grad_norm": 3.8191875557546916e-05,
+      "kl": 0.158203125,
+      "learning_rate": 9.835113145231356e-05,
+      "loss": 0.0002,
+      "num_tokens": 5051408.0,
+      "reward": 2.8089945316314697,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4642857015132904,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.4161372184753418,
+      "step": 497
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6965034965034965,
+      "grad_norm": 3.1854181850026e-05,
+      "kl": 0.158203125,
+      "learning_rate": 9.833555283039842e-05,
+      "loss": 0.0002,
+      "step": 498
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 115.17857360839844,
+      "epoch": 0.6979020979020979,
+      "grad_norm": 3.735234931957505e-07,
+      "kl": 0.15234375,
+      "learning_rate": 9.831990220569801e-05,
+      "loss": 0.0002,
+      "num_tokens": 5072486.0,
+      "reward": 2.7535111904144287,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3571428656578064,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.46779707074165344,
+      "step": 499
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6993006993006993,
+      "grad_norm": 3.749561048042854e-07,
+      "kl": 0.15234375,
+      "learning_rate": 9.83041796015262e-05,
+      "loss": 0.0002,
+      "step": 500
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 137.85714721679688,
+      "epoch": 0.7006993006993008,
+      "grad_norm": 0.004581651252064326,
+      "kl": 0.1357421875,
+      "learning_rate": 9.828838504130406e-05,
+      "loss": 0.0001,
+      "num_tokens": 24128.0,
+      "reward": 2.5751242637634277,
+      "reward_std": 0.050465863198041916,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 1.8214285373687744,
+      "rewards/check_winston_local_func": 0.36083847284317017,
+      "step": 501
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7020979020979021,
+      "grad_norm": 0.004623783518082634,
+      "kl": 0.1357421875,
+      "learning_rate": 9.827251854855991e-05,
+      "loss": 0.0001,
+      "step": 502
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 84.0714340209961,
+      "epoch": 0.7034965034965035,
+      "grad_norm": 5.4224474596128455e-06,
+      "kl": 0.1962890625,
+      "learning_rate": 9.825658014692914e-05,
+      "loss": 0.0002,
+      "num_tokens": 41660.0,
+      "reward": 2.881941556930542,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.48908427357673645,
+      "step": 503
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7048951048951049,
+      "grad_norm": 5.774418529053615e-06,
+      "kl": 0.1962890625,
+      "learning_rate": 9.824056986015433e-05,
+      "loss": 0.0002,
+      "step": 504
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 125.76786041259766,
+      "epoch": 0.7062937062937062,
+      "grad_norm": 0.012190639094338367,
+      "kl": 0.158203125,
+      "learning_rate": 9.82244877120851e-05,
+      "loss": -0.0,
+      "num_tokens": 63809.0,
+      "reward": 2.9153220653533936,
+      "reward_std": 0.05151599273085594,
+      "rewards/check_gptzero_func": 0.5,
+      "rewards/check_perplexity_diff_func": 1.9642857313156128,
+      "rewards/check_winston_local_func": 0.45103612542152405,
+      "step": 505
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7076923076923077,
+      "grad_norm": 0.0008706312902749691,
+      "kl": 0.1572265625,
+      "learning_rate": 9.820833372667812e-05,
+      "loss": -0.0001,
+      "step": 506
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 106.67857360839844,
+      "epoch": 0.7090909090909091,
+      "grad_norm": 7.4245877265908374e-06,
+      "kl": 0.171875,
+      "learning_rate": 9.819210792799712e-05,
+      "loss": 0.0002,
+      "num_tokens": 83877.0,
+      "reward": 2.742607831954956,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3571428656578064,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.3854646682739258,
+      "step": 507
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7104895104895105,
+      "grad_norm": 5.9863315776137925e-06,
+      "kl": 0.171875,
+      "learning_rate": 9.817581034021272e-05,
+      "loss": 0.0002,
+      "step": 508
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 112.21429443359375,
+      "epoch": 0.7118881118881119,
+      "grad_norm": 1.5219106062929897e-05,
+      "kl": 0.154296875,
+      "learning_rate": 9.815944098760257e-05,
+      "loss": 0.0002,
+      "num_tokens": 105225.0,
+      "reward": 2.886364698410034,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.5649359226226807,
+      "step": 509
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7132867132867133,
+      "grad_norm": 3.826854189322478e-05,
+      "kl": 0.154296875,
+      "learning_rate": 9.814299989455117e-05,
+      "loss": 0.0002,
+      "step": 510
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 93.50000762939453,
+      "epoch": 0.7146853146853147,
+      "grad_norm": 3.6634275067756628e-06,
+      "kl": 0.18359375,
+      "learning_rate": 9.81264870855499e-05,
+      "loss": 0.0002,
+      "num_tokens": 124237.0,
+      "reward": 2.9750092029571533,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.5,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.5464377403259277,
+      "step": 511
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7160839160839161,
+      "grad_norm": 3.3120330925968346e-06,
+      "kl": 0.18359375,
+      "learning_rate": 9.810990258519699e-05,
+      "loss": 0.0002,
+      "step": 512
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 114.5714340209961,
+      "epoch": 0.7174825174825175,
+      "grad_norm": 1.5887563588226216e-07,
+      "kl": 0.1474609375,
+      "learning_rate": 9.809324641819741e-05,
+      "loss": 0.0001,
+      "num_tokens": 145557.0,
+      "reward": 2.767850637435913,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3571428656578064,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.4107076823711395,
+      "step": 513
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7188811188811188,
+      "grad_norm": 1.5359302478120885e-07,
+      "kl": 0.1474609375,
+      "learning_rate": 9.807651860936297e-05,
+      "loss": 0.0001,
+      "step": 514
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 110.3214340209961,
+      "epoch": 0.7202797202797203,
+      "grad_norm": 1.620447571811333e-07,
+      "kl": 0.1630859375,
+      "learning_rate": 9.805971918361214e-05,
+      "loss": 0.0002,
+      "num_tokens": 165997.0,
+      "reward": 2.9055941104888916,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.5127367377281189,
+      "step": 515
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7216783216783217,
+      "grad_norm": 1.472040658632873e-07,
+      "kl": 0.1630859375,
+      "learning_rate": 9.804284816597008e-05,
+      "loss": 0.0002,
+      "step": 516
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 91.67857360839844,
+      "epoch": 0.7230769230769231,
+      "grad_norm": 2.4623250489986615e-06,
+      "kl": 0.142578125,
+      "learning_rate": 9.802590558156862e-05,
+      "loss": 0.0001,
+      "num_tokens": 184617.0,
+      "reward": 2.853912830352783,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.46105554699897766,
+      "step": 517
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7244755244755244,
+      "grad_norm": 2.0559549189734124e-06,
+      "kl": 0.142578125,
+      "learning_rate": 9.800889145564617e-05,
+      "loss": 0.0001,
+      "step": 518
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 127.21429443359375,
+      "epoch": 0.7258741258741259,
+      "grad_norm": 2.3530034340292295e-07,
+      "kl": 0.1552734375,
+      "learning_rate": 9.799180581354774e-05,
+      "loss": 0.0002,
+      "num_tokens": 207223.0,
+      "reward": 2.7364511489868164,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3571428656578064,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.45073673129081726,
+      "step": 519
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7272727272727273,
+      "grad_norm": 2.2322343800833895e-07,
+      "kl": 0.1552734375,
+      "learning_rate": 9.797464868072488e-05,
+      "loss": 0.0002,
+      "step": 520
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 104.5714340209961,
+      "epoch": 0.7286713286713287,
+      "grad_norm": 2.968346809067501e-07,
+      "kl": 0.177734375,
+      "learning_rate": 9.795742008273558e-05,
+      "loss": 0.0002,
+      "num_tokens": 227059.0,
+      "reward": 2.6932995319366455,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3571428656578064,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.3361565172672272,
+      "step": 521
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.73006993006993,
+      "grad_norm": 2.852980195095417e-07,
+      "kl": 0.177734375,
+      "learning_rate": 9.794012004524434e-05,
+      "loss": 0.0002,
+      "step": 522
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 105.42857360839844,
+      "epoch": 0.7314685314685314,
+      "grad_norm": 3.3796985539550523e-06,
+      "kl": 0.1640625,
+      "learning_rate": 9.792274859402205e-05,
+      "loss": 0.0002,
+      "num_tokens": 246949.0,
+      "reward": 2.6483211517333984,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3214285671710968,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.3983212113380432,
+      "step": 523
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7328671328671329,
+      "grad_norm": 3.1476330793325015e-06,
+      "kl": 0.1640625,
+      "learning_rate": 9.790530575494603e-05,
+      "loss": 0.0002,
+      "step": 524
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 125.03572082519531,
+      "epoch": 0.7342657342657343,
+      "grad_norm": 0.0060337949589422705,
+      "kl": 0.16796875,
+      "learning_rate": 9.788779155399987e-05,
+      "loss": -0.0001,
+      "num_tokens": 269441.0,
+      "reward": 2.768662214279175,
+      "reward_std": 0.049707408994436264,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 1.9642857313156128,
+      "rewards/check_winston_local_func": 0.4115191400051117,
+      "step": 525
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7356643356643356,
+      "grad_norm": 0.0068588182961818464,
+      "kl": 0.16796875,
+      "learning_rate": 9.787020601727352e-05,
+      "loss": -0.0,
+      "step": 526
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 78.39286041259766,
+      "epoch": 0.737062937062937,
+      "grad_norm": 1.3153821115529096e-05,
+      "kl": 0.2353515625,
+      "learning_rate": 9.785254917096318e-05,
+      "loss": 0.0002,
+      "num_tokens": 286345.0,
+      "reward": 3.074557304382324,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.5,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.5745573043823242,
+      "step": 527
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7384615384615385,
+      "grad_norm": 1.256427796379324e-05,
+      "kl": 0.2353515625,
+      "learning_rate": 9.783482104137127e-05,
+      "loss": 0.0002,
+      "step": 528
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 124.39286041259766,
+      "epoch": 0.7398601398601399,
+      "grad_norm": 1.3517164643713064e-05,
+      "kl": 0.12353515625,
+      "learning_rate": 9.781702165490639e-05,
+      "loss": 0.0001,
+      "num_tokens": 309195.0,
+      "reward": 2.4804210662841797,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 1.6428571939468384,
+      "rewards/check_winston_local_func": 0.44470664858818054,
+      "step": 529
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7412587412587412,
+      "grad_norm": 1.3131407312669822e-05,
+      "kl": 0.12353515625,
+      "learning_rate": 9.779915103808328e-05,
+      "loss": 0.0001,
+      "step": 530
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 111.53572082519531,
+      "epoch": 0.7426573426573426,
+      "grad_norm": 7.637416736333039e-07,
+      "kl": 0.15625,
+      "learning_rate": 9.778120921752285e-05,
+      "loss": 0.0002,
+      "num_tokens": 330505.0,
+      "reward": 3.070021390914917,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.5357142686843872,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.5343068838119507,
+      "step": 531
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7440559440559441,
+      "grad_norm": 7.316834780823122e-07,
+      "kl": 0.15625,
+      "learning_rate": 9.776319621995201e-05,
+      "loss": 0.0002,
+      "step": 532
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 106.35714721679688,
+      "epoch": 0.7454545454545455,
+      "grad_norm": 3.663080215251399e-07,
+      "kl": 0.193359375,
+      "learning_rate": 9.77451120722037e-05,
+      "loss": 0.0002,
+      "num_tokens": 350813.0,
+      "reward": 2.63775634765625,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3214285671710968,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.3877563774585724,
+      "step": 533
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7468531468531469,
+      "grad_norm": 3.8678973306591443e-07,
+      "kl": 0.193359375,
+      "learning_rate": 9.77269568012169e-05,
+      "loss": 0.0002,
+      "step": 534
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 106.67857360839844,
+      "epoch": 0.7482517482517482,
+      "grad_norm": 6.123173544504598e-06,
+      "kl": 0.2216796875,
+      "learning_rate": 9.770873043403648e-05,
+      "loss": 0.0002,
+      "num_tokens": 371585.0,
+      "reward": 2.874614953994751,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.5174719095230103,
+      "step": 535
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7496503496503496,
+      "grad_norm": 6.141198966103098e-06,
+      "kl": 0.2216796875,
+      "learning_rate": 9.769043299781327e-05,
+      "loss": 0.0002,
+      "step": 536
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 104.71428680419922,
+      "epoch": 0.7510489510489511,
+      "grad_norm": 6.024782598667822e-07,
+      "kl": 0.212890625,
+      "learning_rate": 9.767206451980394e-05,
+      "loss": 0.0002,
+      "num_tokens": 391201.0,
+      "reward": 2.666369676589966,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.25,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.4163695275783539,
+      "step": 537
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7524475524475525,
+      "grad_norm": 5.651815000751003e-07,
+      "kl": 0.212890625,
+      "learning_rate": 9.765362502737097e-05,
+      "loss": 0.0002,
+      "step": 538
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 107.21428680419922,
+      "epoch": 0.7538461538461538,
+      "grad_norm": 4.5359076717402366e-07,
+      "kl": 0.19140625,
+      "learning_rate": 9.763511454798268e-05,
+      "loss": 0.0002,
+      "num_tokens": 412087.0,
+      "reward": 2.660142660140991,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.25,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.4815710484981537,
+      "step": 539
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7552447552447552,
+      "grad_norm": 7.72545386297213e-07,
+      "kl": 0.19140625,
+      "learning_rate": 9.761653310921307e-05,
+      "loss": 0.0002,
+      "step": 540
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 133.85714721679688,
+      "epoch": 0.7566433566433567,
+      "grad_norm": 8.430058973039463e-08,
+      "kl": 0.146484375,
+      "learning_rate": 9.759788073874189e-05,
+      "loss": 0.0001,
+      "num_tokens": 435535.0,
+      "reward": 2.720244884490967,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.3631021976470947,
+      "step": 541
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7580419580419581,
+      "grad_norm": 8.327469845321351e-08,
+      "kl": 0.146484375,
+      "learning_rate": 9.757915746435453e-05,
+      "loss": 0.0001,
+      "step": 542
+    },
+    {
+      "clip_ratio": 0.0002040816325461492,
+      "completion_length": 102.89286041259766,
+      "epoch": 0.7594405594405594,
+      "grad_norm": 0.000356312443076848,
+      "kl": 0.203125,
+      "learning_rate": 9.756036331394202e-05,
+      "loss": 0.0002,
+      "num_tokens": 455137.0,
+      "reward": 2.6731011867523193,
+      "reward_std": 1.646135569899343e-05,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.351672500371933,
+      "step": 543
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7608391608391608,
+      "grad_norm": 0.00031516713281171804,
+      "kl": 0.203125,
+      "learning_rate": 9.754149831550098e-05,
+      "loss": 0.0002,
+      "step": 544
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 89.35714721679688,
+      "epoch": 0.7622377622377622,
+      "grad_norm": 2.70311309473727e-07,
+      "kl": 0.2080078125,
+      "learning_rate": 9.752256249713351e-05,
+      "loss": 0.0002,
+      "num_tokens": 473297.0,
+      "reward": 2.869070291519165,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3571428656578064,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.5119272470474243,
+      "step": 545
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7636363636363637,
+      "grad_norm": 2.830814999627113e-07,
+      "kl": 0.2080078125,
+      "learning_rate": 9.750355588704727e-05,
+      "loss": 0.0002,
+      "step": 546
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 132.44644165039062,
+      "epoch": 0.765034965034965,
+      "grad_norm": 0.0009688545197921594,
+      "kl": 0.1806640625,
+      "learning_rate": 9.748447851355535e-05,
+      "loss": 0.0002,
+      "num_tokens": 496926.0,
+      "reward": 2.7388832569122314,
+      "reward_std": 0.0023702967446297407,
+      "rewards/check_gptzero_func": 0.5357142686843872,
+      "rewards/check_perplexity_diff_func": 1.7142857313156128,
+      "rewards/check_winston_local_func": 0.4888834059238434,
+      "step": 547
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7664335664335664,
+      "grad_norm": 0.0016734864168510978,
+      "kl": 0.1806640625,
+      "learning_rate": 9.746533040507624e-05,
+      "loss": 0.0002,
+      "step": 548
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 97.03572082519531,
+      "epoch": 0.7678321678321678,
+      "grad_norm": 1.9163803223436476e-06,
+      "kl": 0.177734375,
+      "learning_rate": 9.744611159013381e-05,
+      "loss": 0.0002,
+      "num_tokens": 516080.0,
+      "reward": 2.9889118671417236,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.5,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.48891177773475647,
+      "step": 549
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7692307692307693,
+      "grad_norm": 1.2761616262245509e-06,
+      "kl": 0.177734375,
+      "learning_rate": 9.742682209735727e-05,
+      "loss": 0.0002,
+      "step": 550
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 92.39286041259766,
+      "epoch": 0.7706293706293706,
+      "grad_norm": 1.896094459987994e-07,
+      "kl": 0.1923828125,
+      "learning_rate": 9.740746195548112e-05,
+      "loss": 0.0002,
+      "num_tokens": 534554.0,
+      "reward": 2.943678855895996,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.5151072144508362,
+      "step": 551
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.772027972027972,
+      "grad_norm": 2.1137044966091938e-07,
+      "kl": 0.1923828125,
+      "learning_rate": 9.738803119334506e-05,
+      "loss": 0.0002,
+      "step": 552
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 103.21428680419922,
+      "epoch": 0.7734265734265734,
+      "grad_norm": 4.705318347406964e-07,
+      "kl": 0.1826171875,
+      "learning_rate": 9.736852983989404e-05,
+      "loss": 0.0002,
+      "num_tokens": 554220.0,
+      "reward": 2.9958415031433105,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.5,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.4958415627479553,
+      "step": 553
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7748251748251749,
+      "grad_norm": 5.24917119837944e-07,
+      "kl": 0.1826171875,
+      "learning_rate": 9.734895792417811e-05,
+      "loss": 0.0002,
+      "step": 554
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 102.53572082519531,
+      "epoch": 0.7762237762237763,
+      "grad_norm": 0.0005595837692618245,
+      "kl": 0.1611328125,
+      "learning_rate": 9.73293154753525e-05,
+      "loss": -0.0001,
+      "num_tokens": 574520.0,
+      "reward": 2.660358190536499,
+      "reward_std": 0.05061452463269234,
+      "rewards/check_gptzero_func": 0.2857142984867096,
+      "rewards/check_perplexity_diff_func": 1.75,
+      "rewards/check_winston_local_func": 0.6246438026428223,
+      "step": 555
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7776223776223776,
+      "grad_norm": 0.0005476655329061088,
+      "kl": 0.1611328125,
+      "learning_rate": 9.730960252267743e-05,
+      "loss": -0.0001,
+      "step": 556
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 110.28572082519531,
+      "epoch": 0.779020979020979,
+      "grad_norm": 1.9994241055330475e-07,
+      "kl": 0.162109375,
+      "learning_rate": 9.728981909551824e-05,
+      "loss": 0.0002,
+      "num_tokens": 595056.0,
+      "reward": 2.7659287452697754,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3571428656578064,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.40878555178642273,
+      "step": 557
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7804195804195804,
+      "grad_norm": 2.102720872162732e-07,
+      "kl": 0.162109375,
+      "learning_rate": 9.726996522334516e-05,
+      "loss": 0.0002,
+      "step": 558
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 116.3214340209961,
+      "epoch": 0.7818181818181819,
+      "grad_norm": 1.4418577967265551e-06,
+      "kl": 0.197265625,
+      "learning_rate": 9.725004093573342e-05,
+      "loss": 0.0002,
+      "num_tokens": 616218.0,
+      "reward": 2.90523099899292,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.4766596257686615,
+      "step": 559
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7832167832167832,
+      "grad_norm": 1.4761501812795077e-06,
+      "kl": 0.197265625,
+      "learning_rate": 9.723004626236314e-05,
+      "loss": 0.0002,
+      "step": 560
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 92.28572082519531,
+      "epoch": 0.7846153846153846,
+      "grad_norm": 6.640556191873379e-07,
+      "kl": 0.1904296875,
+      "learning_rate": 9.720998123301923e-05,
+      "loss": 0.0002,
+      "num_tokens": 634450.0,
+      "reward": 2.910466432571411,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4642857015132904,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.4461804926395416,
+      "step": 561
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.786013986013986,
+      "grad_norm": 6.785626739636823e-07,
+      "kl": 0.1904296875,
+      "learning_rate": 9.718984587759148e-05,
+      "loss": 0.0002,
+      "step": 562
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 97.67857360839844,
+      "epoch": 0.7874125874125875,
+      "grad_norm": 0.0002364011375932314,
+      "kl": 0.1953125,
+      "learning_rate": 9.71696402260744e-05,
+      "loss": 0.0002,
+      "num_tokens": 653730.0,
+      "reward": 3.09153151512146,
+      "reward_std": 4.5663102355320007e-05,
+      "rewards/check_gptzero_func": 0.5714285969734192,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.5201030373573303,
+      "step": 563
+    },
+    {
+      "clip_ratio": 0.00014228800137061626,
+      "epoch": 0.7888111888111888,
+      "grad_norm": 0.00020388090420570156,
+      "kl": 0.1953125,
+      "learning_rate": 9.714936430856723e-05,
+      "loss": 0.0002,
+      "step": 564
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 120.66072082519531,
+      "epoch": 0.7902097902097902,
+      "grad_norm": 8.089432221384178e-05,
+      "kl": 0.1357421875,
+      "learning_rate": 9.712901815527386e-05,
+      "loss": 0.0001,
+      "num_tokens": 675657.0,
+      "reward": 2.674098253250122,
+      "reward_std": 0.0001108883589040488,
+      "rewards/check_gptzero_func": 0.3571428656578064,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.38838380575180054,
+      "step": 565
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7916083916083916,
+      "grad_norm": 8.889658368646956e-05,
+      "kl": 0.1357421875,
+      "learning_rate": 9.710860179650287e-05,
+      "loss": 0.0001,
+      "step": 566
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 107.3214340209961,
+      "epoch": 0.793006993006993,
+      "grad_norm": 0.0001037228042730343,
+      "kl": 0.17578125,
+      "learning_rate": 9.70881152626673e-05,
+      "loss": 0.0002,
+      "num_tokens": 695807.0,
+      "reward": 2.615325927734375,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.25,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.4367544949054718,
+      "step": 567
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7944055944055944,
+      "grad_norm": 9.62410190276379e-05,
+      "kl": 0.17578125,
+      "learning_rate": 9.706755858428486e-05,
+      "loss": 0.0002,
+      "step": 568
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 128.3928680419922,
+      "epoch": 0.7958041958041958,
+      "grad_norm": 3.2052372287941215e-05,
+      "kl": 0.12890625,
+      "learning_rate": 9.704693179197767e-05,
+      "loss": 0.0001,
+      "num_tokens": 719201.0,
+      "reward": 2.616921901702881,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 1.7857142686843872,
+      "rewards/check_winston_local_func": 0.40263620018959045,
+      "step": 569
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7972027972027972,
+      "grad_norm": 3.1589569913845654e-05,
+      "kl": 0.12890625,
+      "learning_rate": 9.702623491647233e-05,
+      "loss": 0.0001,
+      "step": 570
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 97.3214340209961,
+      "epoch": 0.7986013986013986,
+      "grad_norm": 2.966365002312992e-07,
+      "kl": 0.1572265625,
+      "learning_rate": 9.70054679885998e-05,
+      "loss": 0.0002,
+      "num_tokens": 738947.0,
+      "reward": 2.835333824157715,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.47819074988365173,
+      "step": 571
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8,
+      "grad_norm": 4.1573650424054196e-07,
+      "kl": 0.1572265625,
+      "learning_rate": 9.698463103929542e-05,
+      "loss": 0.0002,
+      "step": 572
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 105.78572082519531,
+      "epoch": 0.8013986013986014,
+      "grad_norm": 0.000588551803293507,
+      "kl": 0.1611328125,
+      "learning_rate": 9.696372409959886e-05,
+      "loss": 0.0002,
+      "num_tokens": 759359.0,
+      "reward": 2.967468738555908,
+      "reward_std": 0.0020603849552571774,
+      "rewards/check_gptzero_func": 0.5,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.5388973355293274,
+      "step": 573
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8027972027972028,
+      "grad_norm": 0.0006215594211925717,
+      "kl": 0.1611328125,
+      "learning_rate": 9.694274720065399e-05,
+      "loss": 0.0002,
+      "step": 574
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 97.14286041259766,
+      "epoch": 0.8041958041958042,
+      "grad_norm": 3.3967306751168834e-06,
+      "kl": 0.17578125,
+      "learning_rate": 9.692170037370898e-05,
+      "loss": 0.0002,
+      "num_tokens": 778637.0,
+      "reward": 2.9381115436553955,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.509539783000946,
+      "step": 575
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8055944055944056,
+      "grad_norm": 3.2036118908588705e-06,
+      "kl": 0.17578125,
+      "learning_rate": 9.690058365011607e-05,
+      "loss": 0.0002,
+      "step": 576
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 102.64286041259766,
+      "epoch": 0.806993006993007,
+      "grad_norm": 8.125129903430204e-06,
+      "kl": 0.1630859375,
+      "learning_rate": 9.687939706133168e-05,
+      "loss": 0.0002,
+      "num_tokens": 798473.0,
+      "reward": 2.781018018722534,
+      "reward_std": 0.025253813713788986,
+      "rewards/check_gptzero_func": 0.4107142984867096,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.4417320787906647,
+      "step": 577
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8083916083916084,
+      "grad_norm": 7.98774787404577e-06,
+      "kl": 0.1630859375,
+      "learning_rate": 9.685814063891631e-05,
+      "loss": 0.0002,
+      "step": 578
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 126.85714721679688,
+      "epoch": 0.8097902097902098,
+      "grad_norm": 3.679599498892351e-07,
+      "kl": 0.138671875,
+      "learning_rate": 9.683681441453445e-05,
+      "loss": 0.0001,
+      "num_tokens": 821081.0,
+      "reward": 2.8629186153411865,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.43434715270996094,
+      "step": 579
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8111888111888111,
+      "grad_norm": 3.93663211553172e-07,
+      "kl": 0.138671875,
+      "learning_rate": 9.681541841995461e-05,
+      "loss": 0.0001,
+      "step": 580
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 106.00000762939453,
+      "epoch": 0.8125874125874126,
+      "grad_norm": 8.022029708518222e-07,
+      "kl": 0.162109375,
+      "learning_rate": 9.67939526870492e-05,
+      "loss": 0.0002,
+      "num_tokens": 841529.0,
+      "reward": 2.8752760887145996,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.55384761095047,
+      "step": 581
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.813986013986014,
+      "grad_norm": 8.501682817147847e-07,
+      "kl": 0.162109375,
+      "learning_rate": 9.677241724779453e-05,
+      "loss": 0.0002,
+      "step": 582
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 109.85714721679688,
+      "epoch": 0.8153846153846154,
+      "grad_norm": 1.7930012533014129e-06,
+      "kl": 0.1689453125,
+      "learning_rate": 9.675081213427076e-05,
+      "loss": 0.0002,
+      "num_tokens": 862213.0,
+      "reward": 2.9139628410339355,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.48539119958877563,
+      "step": 583
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8167832167832167,
+      "grad_norm": 1.6519465345892093e-06,
+      "kl": 0.1689453125,
+      "learning_rate": 9.672913737866179e-05,
+      "loss": 0.0002,
+      "step": 584
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 98.66072082519531,
+      "epoch": 0.8181818181818182,
+      "grad_norm": 0.002787236207669909,
+      "kl": 0.25,
+      "learning_rate": 9.670739301325534e-05,
+      "loss": 0.0002,
+      "num_tokens": 881842.0,
+      "reward": 2.6230576038360596,
+      "reward_std": 0.006216124631464481,
+      "rewards/check_gptzero_func": 0.3571428656578064,
+      "rewards/check_perplexity_diff_func": 1.7857142686843872,
+      "rewards/check_winston_local_func": 0.48020049929618835,
+      "step": 585
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8195804195804196,
+      "grad_norm": 0.002885109231591189,
+      "kl": 0.2490234375,
+      "learning_rate": 9.668557907044276e-05,
+      "loss": 0.0002,
+      "step": 586
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 79.14286041259766,
+      "epoch": 0.820979020979021,
+      "grad_norm": 4.462250819797621e-06,
+      "kl": 0.1923828125,
+      "learning_rate": 9.666369558271909e-05,
+      "loss": 0.0002,
+      "num_tokens": 898716.0,
+      "reward": 2.900649309158325,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.5077921152114868,
+      "step": 587
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8223776223776224,
+      "grad_norm": 4.488075066816524e-06,
+      "kl": 0.1923828125,
+      "learning_rate": 9.66417425826829e-05,
+      "loss": 0.0002,
+      "step": 588
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 112.03572082519531,
+      "epoch": 0.8237762237762237,
+      "grad_norm": 1.3821264354883407e-07,
+      "kl": 0.14453125,
+      "learning_rate": 9.661972010303641e-05,
+      "loss": 0.0001,
+      "num_tokens": 919636.0,
+      "reward": 2.7668612003326416,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4642857015132904,
+      "rewards/check_perplexity_diff_func": 1.8571428060531616,
+      "rewards/check_winston_local_func": 0.4454323649406433,
+      "step": 589
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8251748251748252,
+      "grad_norm": 1.4174011116816794e-07,
+      "kl": 0.14453125,
+      "learning_rate": 9.659762817658524e-05,
+      "loss": 0.0001,
+      "step": 590
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 107.92857360839844,
+      "epoch": 0.8265734265734266,
+      "grad_norm": 8.773734381605369e-07,
+      "kl": 0.1572265625,
+      "learning_rate": 9.65754668362385e-05,
+      "loss": 0.0002,
+      "num_tokens": 940096.0,
+      "reward": 2.8289589881896973,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4642857015132904,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.4361015856266022,
+      "step": 591
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.827972027972028,
+      "grad_norm": 1.0023469469060766e-06,
+      "kl": 0.1572265625,
+      "learning_rate": 9.655323611500875e-05,
+      "loss": 0.0002,
+      "step": 592
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 113.85714721679688,
+      "epoch": 0.8293706293706293,
+      "grad_norm": 6.1541718025779055e-06,
+      "kl": 0.166015625,
+      "learning_rate": 9.653093604601183e-05,
+      "loss": 0.0002,
+      "num_tokens": 961340.0,
+      "reward": 2.9606730937957764,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.5357142686843872,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.4249587059020996,
+      "step": 593
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8307692307692308,
+      "grad_norm": 6.512813773208274e-06,
+      "kl": 0.166015625,
+      "learning_rate": 9.650856666246693e-05,
+      "loss": 0.0002,
+      "step": 594
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 91.60714721679688,
+      "epoch": 0.8321678321678322,
+      "grad_norm": 3.2190340652283214e-06,
+      "kl": 0.1767578125,
+      "learning_rate": 9.648612799769644e-05,
+      "loss": 0.0002,
+      "num_tokens": 979916.0,
+      "reward": 3.0096070766448975,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.5810357332229614,
+      "step": 595
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8335664335664336,
+      "grad_norm": 3.1137752081162388e-06,
+      "kl": 0.1767578125,
+      "learning_rate": 9.646362008512602e-05,
+      "loss": 0.0002,
+      "step": 596
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 106.53572082519531,
+      "epoch": 0.8349650349650349,
+      "grad_norm": 6.812051267845749e-06,
+      "kl": 0.185546875,
+      "learning_rate": 9.644104295828447e-05,
+      "loss": 0.0002,
+      "num_tokens": 1000300.0,
+      "reward": 2.7739078998565674,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.4167649447917938,
+      "step": 597
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8363636363636363,
+      "grad_norm": 2.398683493432653e-06,
+      "kl": 0.185546875,
+      "learning_rate": 9.641839665080363e-05,
+      "loss": 0.0002,
+      "step": 598
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 107.41072082519531,
+      "epoch": 0.8377622377622378,
+      "grad_norm": 0.0008785473557125962,
+      "kl": 0.1796875,
+      "learning_rate": 9.63956811964185e-05,
+      "loss": 0.0002,
+      "num_tokens": 1021105.0,
+      "reward": 2.91171932220459,
+      "reward_std": 0.00400555832311511,
+      "rewards/check_gptzero_func": 0.5357142686843872,
+      "rewards/check_perplexity_diff_func": 1.8571428060531616,
+      "rewards/check_winston_local_func": 0.518862247467041,
+      "step": 599
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8391608391608392,
+      "grad_norm": 0.000892784560673435,
+      "kl": 0.1796875,
+      "learning_rate": 9.6372896628967e-05,
+      "loss": 0.0002,
+      "step": 600
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 2860,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 4,
+  "save_steps": 100,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 0.0,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/checkpoint-600/training_args.bin b/checkpoint-600/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..d7c88a0374f4d875b15189b1d13e5e7ee4aa1983
--- /dev/null
+++ b/checkpoint-600/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c67632eb95048ac9b9e00b6b7f7052aa5e28ce695d584cdd89a4780d83119ce9
+size 7480
diff --git a/checkpoint-600/zero_to_fp32.py b/checkpoint-600/zero_to_fp32.py
new file mode 100644
index 0000000000000000000000000000000000000000..0e759146cadd92ddfefab3680146c2bd6a2b5c04
--- /dev/null
+++ b/checkpoint-600/zero_to_fp32.py
@@ -0,0 +1,760 @@
+#!/usr/bin/env python
+
+# Copyright (c) Microsoft Corporation.
+# SPDX-License-Identifier: Apache-2.0
+
+# DeepSpeed Team
+
+# This script extracts fp32 consolidated weights from a zero 1, 2 and 3 DeepSpeed checkpoints. It gets
+# copied into the top level checkpoint dir, so the user can easily do the conversion at any point in
+# the future. Once extracted, the weights don't require DeepSpeed and can be used in any
+# application.
+#
+# example:
+#   python zero_to_fp32.py . output_dir/
+#   or
+#   python zero_to_fp32.py . output_dir/ --safe_serialization
+
+import argparse
+import torch
+import glob
+import math
+import os
+import re
+import gc
+import json
+import numpy as np
+from tqdm import tqdm
+from collections import OrderedDict
+from dataclasses import dataclass
+
+# while this script doesn't use deepspeed to recover data, since the checkpoints are pickled with
+# DeepSpeed data structures it has to be available in the current python environment.
+from deepspeed.utils import logger
+from deepspeed.checkpoint.constants import (DS_VERSION, OPTIMIZER_STATE_DICT, SINGLE_PARTITION_OF_FP32_GROUPS,
+                                            FP32_FLAT_GROUPS, ZERO_STAGE, PARTITION_COUNT, PARAM_SHAPES, BUFFER_NAMES,
+                                            FROZEN_PARAM_SHAPES, FROZEN_PARAM_FRAGMENTS)
+
+
+@dataclass
+class zero_model_state:
+    buffers: dict()
+    param_shapes: dict()
+    shared_params: list
+    ds_version: int
+    frozen_param_shapes: dict()
+    frozen_param_fragments: dict()
+
+
+debug = 0
+
+# load to cpu
+device = torch.device('cpu')
+
+
+def atoi(text):
+    return int(text) if text.isdigit() else text
+
+
+def natural_keys(text):
+    '''
+    alist.sort(key=natural_keys) sorts in human order
+    http://nedbatchelder.com/blog/200712/human_sorting.html
+    (See Toothy's implementation in the comments)
+    '''
+    return [atoi(c) for c in re.split(r'(\d+)', text)]
+
+
+def get_model_state_file(checkpoint_dir, zero_stage):
+    if not os.path.isdir(checkpoint_dir):
+        raise FileNotFoundError(f"Directory '{checkpoint_dir}' doesn't exist")
+
+    # there should be only one file
+    if zero_stage <= 2:
+        file = os.path.join(checkpoint_dir, "mp_rank_00_model_states.pt")
+    elif zero_stage == 3:
+        file = os.path.join(checkpoint_dir, "zero_pp_rank_0_mp_rank_00_model_states.pt")
+
+    if not os.path.exists(file):
+        raise FileNotFoundError(f"can't find model states file at '{file}'")
+
+    return file
+
+
+def get_checkpoint_files(checkpoint_dir, glob_pattern):
+    # XXX: need to test that this simple glob rule works for multi-node setup too
+    ckpt_files = sorted(glob.glob(os.path.join(checkpoint_dir, glob_pattern)), key=natural_keys)
+
+    if len(ckpt_files) == 0:
+        raise FileNotFoundError(f"can't find {glob_pattern} files in directory '{checkpoint_dir}'")
+
+    return ckpt_files
+
+
+def get_optim_files(checkpoint_dir):
+    return get_checkpoint_files(checkpoint_dir, "*_optim_states.pt")
+
+
+def get_model_state_files(checkpoint_dir):
+    return get_checkpoint_files(checkpoint_dir, "*_model_states.pt")
+
+
+def parse_model_states(files):
+    zero_model_states = []
+    for file in files:
+        state_dict = torch.load(file, map_location=device, weights_only=False)
+
+        if BUFFER_NAMES not in state_dict:
+            raise ValueError(f"{file} is not a model state checkpoint")
+        buffer_names = state_dict[BUFFER_NAMES]
+        if debug:
+            print("Found buffers:", buffer_names)
+
+        # recover just the buffers while restoring them to fp32 if they were saved in fp16
+        buffers = {k: v.float() for k, v in state_dict["module"].items() if k in buffer_names}
+        param_shapes = state_dict[PARAM_SHAPES]
+
+        # collect parameters that are included in param_shapes
+        param_names = []
+        for s in param_shapes:
+            for name in s.keys():
+                param_names.append(name)
+
+        # update with frozen parameters
+        frozen_param_shapes = state_dict.get(FROZEN_PARAM_SHAPES, None)
+        if frozen_param_shapes is not None:
+            if debug:
+                print(f"Found frozen_param_shapes: {frozen_param_shapes}")
+            param_names += list(frozen_param_shapes.keys())
+
+        # handle shared params
+        shared_params = [[k, v] for k, v in state_dict["shared_params"].items()]
+
+        ds_version = state_dict.get(DS_VERSION, None)
+
+        frozen_param_fragments = state_dict.get(FROZEN_PARAM_FRAGMENTS, None)
+
+        z_model_state = zero_model_state(buffers=buffers,
+                                         param_shapes=param_shapes,
+                                         shared_params=shared_params,
+                                         ds_version=ds_version,
+                                         frozen_param_shapes=frozen_param_shapes,
+                                         frozen_param_fragments=frozen_param_fragments)
+        zero_model_states.append(z_model_state)
+
+    return zero_model_states
+
+
+def parse_optim_states(files, ds_checkpoint_dir):
+    total_files = len(files)
+    state_dicts = []
+    for f in tqdm(files, desc='Loading checkpoint shards'):
+        state_dict = torch.load(f, map_location=device, mmap=True, weights_only=False)
+        # immediately discard the potentially huge 2 optimizer states as we only care for fp32 master weights
+        # and also handle the case where it was already removed by another helper script
+        state_dict["optimizer_state_dict"].pop("optimizer_state_dict", None)
+        state_dicts.append(state_dict)
+
+    if not ZERO_STAGE in state_dicts[0][OPTIMIZER_STATE_DICT]:
+        raise ValueError(f"{files[0]} is not a zero checkpoint")
+    zero_stage = state_dicts[0][OPTIMIZER_STATE_DICT][ZERO_STAGE]
+    world_size = state_dicts[0][OPTIMIZER_STATE_DICT][PARTITION_COUNT]
+
+    # For ZeRO-2 each param group can have different partition_count as data parallelism for expert
+    # parameters can be different from data parallelism for non-expert parameters. So we can just
+    # use the max of the partition_count to get the dp world_size.
+
+    if type(world_size) is list:
+        world_size = max(world_size)
+
+    if world_size != total_files:
+        raise ValueError(
+            f"Expected {world_size} of '*_optim_states.pt' under '{ds_checkpoint_dir}' but found {total_files} files. "
+            "Possibly due to an overwrite of an old checkpoint, or a checkpoint didn't get saved by one or more processes."
+        )
+
+    # the groups are named differently in each stage
+    if zero_stage <= 2:
+        fp32_groups_key = SINGLE_PARTITION_OF_FP32_GROUPS
+    elif zero_stage == 3:
+        fp32_groups_key = FP32_FLAT_GROUPS
+    else:
+        raise ValueError(f"unknown zero stage {zero_stage}")
+
+    fp32_flat_groups = [state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key] for i in range(len(state_dicts))]
+    return zero_stage, world_size, fp32_flat_groups
+
+
+def _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters):
+    """
+    Returns fp32 state_dict reconstructed from ds checkpoint
+
+    Args:
+        - ``ds_checkpoint_dir``: path to the deepspeed checkpoint folder (where the optimizer files are)
+
+    """
+    print(f"Processing zero checkpoint '{ds_checkpoint_dir}'")
+
+    optim_files = get_optim_files(ds_checkpoint_dir)
+    zero_stage, world_size, fp32_flat_groups = parse_optim_states(optim_files, ds_checkpoint_dir)
+    print(f"Detected checkpoint of type zero stage {zero_stage}, world_size: {world_size}")
+
+    model_files = get_model_state_files(ds_checkpoint_dir)
+
+    zero_model_states = parse_model_states(model_files)
+    print(f'Parsing checkpoint created by deepspeed=={zero_model_states[0].ds_version}')
+
+    if zero_stage <= 2:
+        return _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states,
+                                                          exclude_frozen_parameters)
+    elif zero_stage == 3:
+        return _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states,
+                                                          exclude_frozen_parameters)
+
+
+def _zero2_merge_frozen_params(state_dict, zero_model_states):
+    if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0:
+        return
+
+    frozen_param_shapes = zero_model_states[0].frozen_param_shapes
+    frozen_param_fragments = zero_model_states[0].frozen_param_fragments
+
+    if debug:
+        num_elem = sum(s.numel() for s in frozen_param_shapes.values())
+        print(f'rank 0: {FROZEN_PARAM_SHAPES}.numel = {num_elem}')
+
+        wanted_params = len(frozen_param_shapes)
+        wanted_numel = sum(s.numel() for s in frozen_param_shapes.values())
+        avail_numel = sum([p.numel() for p in frozen_param_fragments.values()])
+        print(f'Frozen params: Have {avail_numel} numels to process.')
+        print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params')
+
+    total_params = 0
+    total_numel = 0
+    for name, shape in frozen_param_shapes.items():
+        total_params += 1
+        unpartitioned_numel = shape.numel()
+        total_numel += unpartitioned_numel
+
+        state_dict[name] = frozen_param_fragments[name]
+
+        if debug:
+            print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ")
+
+    print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements")
+
+
+def _has_callable(obj, fn):
+    attr = getattr(obj, fn, None)
+    return callable(attr)
+
+
+def _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states):
+    param_shapes = zero_model_states[0].param_shapes
+
+    # Reconstruction protocol:
+    #
+    # XXX: document this
+
+    if debug:
+        for i in range(world_size):
+            for j in range(len(fp32_flat_groups[0])):
+                print(f"{FP32_FLAT_GROUPS}[{i}][{j}].shape={fp32_flat_groups[i][j].shape}")
+
+    # XXX: memory usage doubles here (zero2)
+    num_param_groups = len(fp32_flat_groups[0])
+    merged_single_partition_of_fp32_groups = []
+    for i in range(num_param_groups):
+        merged_partitions = [sd[i] for sd in fp32_flat_groups]
+        full_single_fp32_vector = torch.cat(merged_partitions, 0)
+        merged_single_partition_of_fp32_groups.append(full_single_fp32_vector)
+    avail_numel = sum(
+        [full_single_fp32_vector.numel() for full_single_fp32_vector in merged_single_partition_of_fp32_groups])
+
+    if debug:
+        wanted_params = sum([len(shapes) for shapes in param_shapes])
+        wanted_numel = sum([sum(shape.numel() for shape in shapes.values()) for shapes in param_shapes])
+        # not asserting if there is a mismatch due to possible padding
+        print(f"Have {avail_numel} numels to process.")
+        print(f"Need {wanted_numel} numels in {wanted_params} params.")
+
+    # params
+    # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support
+    # out-of-core computing solution
+    total_numel = 0
+    total_params = 0
+    for shapes, full_single_fp32_vector in zip(param_shapes, merged_single_partition_of_fp32_groups):
+        offset = 0
+        avail_numel = full_single_fp32_vector.numel()
+        for name, shape in shapes.items():
+
+            unpartitioned_numel = shape.numel() if _has_callable(shape, 'numel') else math.prod(shape)
+            total_numel += unpartitioned_numel
+            total_params += 1
+
+            if debug:
+                print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ")
+            state_dict[name] = full_single_fp32_vector.narrow(0, offset, unpartitioned_numel).view(shape)
+            offset += unpartitioned_numel
+
+        # Z2 started to align to 2*world_size to improve nccl performance. Therefore both offset and
+        # avail_numel can differ by anywhere between 0..2*world_size. Due to two unrelated complex
+        # paddings performed in the code it's almost impossible to predict the exact numbers w/o the
+        # live optimizer object, so we are checking that the numbers are within the right range
+        align_to = 2 * world_size
+
+        def zero2_align(x):
+            return align_to * math.ceil(x / align_to)
+
+        if debug:
+            print(f"original offset={offset}, avail_numel={avail_numel}")
+
+        offset = zero2_align(offset)
+        avail_numel = zero2_align(avail_numel)
+
+        if debug:
+            print(f"aligned  offset={offset}, avail_numel={avail_numel}")
+
+        # Sanity check
+        if offset != avail_numel:
+            raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong")
+
+    print(f"Reconstructed fp32 state dict with {total_params} params {total_numel} elements")
+
+
+def _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states,
+                                               exclude_frozen_parameters):
+    state_dict = OrderedDict()
+
+    # buffers
+    buffers = zero_model_states[0].buffers
+    state_dict.update(buffers)
+    if debug:
+        print(f"added {len(buffers)} buffers")
+
+    if not exclude_frozen_parameters:
+        _zero2_merge_frozen_params(state_dict, zero_model_states)
+
+    _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states)
+
+    # recover shared parameters
+    for pair in zero_model_states[0].shared_params:
+        if pair[1] in state_dict:
+            state_dict[pair[0]] = state_dict[pair[1]]
+
+    return state_dict
+
+
+def zero3_partitioned_param_info(unpartitioned_numel, world_size):
+    remainder = unpartitioned_numel % world_size
+    padding_numel = (world_size - remainder) if remainder else 0
+    partitioned_numel = math.ceil(unpartitioned_numel / world_size)
+    return partitioned_numel, padding_numel
+
+
+def _zero3_merge_frozen_params(state_dict, world_size, zero_model_states):
+    if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0:
+        return
+
+    if debug:
+        for i in range(world_size):
+            num_elem = sum(s.numel() for s in zero_model_states[i].frozen_param_fragments.values())
+            print(f'rank {i}: {FROZEN_PARAM_SHAPES}.numel = {num_elem}')
+
+        frozen_param_shapes = zero_model_states[0].frozen_param_shapes
+        wanted_params = len(frozen_param_shapes)
+        wanted_numel = sum(s.numel() for s in frozen_param_shapes.values())
+        avail_numel = sum([p.numel() for p in zero_model_states[0].frozen_param_fragments.values()]) * world_size
+        print(f'Frozen params: Have {avail_numel} numels to process.')
+        print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params')
+
+    total_params = 0
+    total_numel = 0
+    for name, shape in zero_model_states[0].frozen_param_shapes.items():
+        total_params += 1
+        unpartitioned_numel = shape.numel()
+        total_numel += unpartitioned_numel
+
+        param_frags = tuple(model_state.frozen_param_fragments[name] for model_state in zero_model_states)
+        state_dict[name] = torch.cat(param_frags, 0).narrow(0, 0, unpartitioned_numel).view(shape)
+
+        partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size)
+
+        if debug:
+            print(
+                f"Frozen params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}"
+            )
+
+    print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements")
+
+
+class GatheredTensor:
+    """
+    A pseudo tensor that collects partitioned weights.
+    It is more memory efficient when there are multiple groups.
+    """
+
+    def __init__(self, flat_groups, flat_groups_offset, offset, partitioned_numel, shape):
+        self.flat_groups = flat_groups
+        self.flat_groups_offset = flat_groups_offset
+        self.offset = offset
+        self.partitioned_numel = partitioned_numel
+        self.shape = shape
+        self.dtype = self.flat_groups[0][0].dtype
+
+    def contiguous(self):
+        """
+        Merge partitioned weights from flat_groups into a single tensor.
+        """
+        end_idx = self.offset + self.partitioned_numel
+        world_size = len(self.flat_groups)
+        pad_flat_param_chunks = []
+
+        for rank_i in range(world_size):
+            # for each rank, we need to collect weights from related group/groups
+            flat_groups_at_rank_i = self.flat_groups[rank_i]
+            start_group_id = None
+            end_group_id = None
+            for group_id in range(len(self.flat_groups_offset)):
+                if self.flat_groups_offset[group_id] <= self.offset < self.flat_groups_offset[group_id + 1]:
+                    start_group_id = group_id
+                if self.flat_groups_offset[group_id] < end_idx <= self.flat_groups_offset[group_id + 1]:
+                    end_group_id = group_id
+                    break
+            # collect weights from related group/groups
+            for group_id in range(start_group_id, end_group_id + 1):
+                flat_tensor = flat_groups_at_rank_i[group_id]
+                start_offset = self.offset - self.flat_groups_offset[group_id]
+                end_offset = min(end_idx, self.flat_groups_offset[group_id + 1]) - self.flat_groups_offset[group_id]
+                pad_flat_param_chunks.append(flat_tensor[start_offset:end_offset])
+
+        # collect weights from all ranks
+        pad_flat_param = torch.cat(pad_flat_param_chunks, dim=0)
+        param = pad_flat_param[:self.shape.numel()].view(self.shape).contiguous()
+        return param
+
+
+def _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states):
+    param_shapes = zero_model_states[0].param_shapes
+    avail_numel = sum([flat_group.numel() for flat_group in fp32_flat_groups[0]]) * world_size
+
+    # Reconstruction protocol: For zero3 we need to zip the partitions together at boundary of each
+    # param, re-consolidating each param, while dealing with padding if any
+
+    # merge list of dicts, preserving order
+    param_shapes = {k: v for d in param_shapes for k, v in d.items()}
+
+    if debug:
+        for i in range(world_size):
+            print(f"{FP32_FLAT_GROUPS}[{i}].shape={fp32_flat_groups[i].shape}")
+
+        wanted_params = len(param_shapes)
+        wanted_numel = sum(shape.numel() for shape in param_shapes.values())
+        # not asserting if there is a mismatch due to possible padding
+        avail_numel = fp32_flat_groups[0].numel() * world_size
+        print(f"Trainable params: Have {avail_numel} numels to process.")
+        print(f"Trainable params: Need {wanted_numel} numels in {wanted_params} params.")
+
+    # params
+    # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support
+    # out-of-core computing solution
+    offset = 0
+    total_numel = 0
+    total_params = 0
+    flat_groups_offset = [0] + list(np.cumsum([flat_tensor.numel() for flat_tensor in fp32_flat_groups[0]]))
+    for name, shape in tqdm(param_shapes.items(), desc='Gathering sharded weights'):
+        unpartitioned_numel = shape.numel()
+        total_numel += unpartitioned_numel
+        total_params += 1
+        partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size)
+
+        if debug:
+            print(
+                f"Trainable params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}"
+            )
+
+        # memory efficient tensor
+        tensor = GatheredTensor(fp32_flat_groups, flat_groups_offset, offset, partitioned_numel, shape)
+        state_dict[name] = tensor
+        offset += partitioned_numel
+
+    offset *= world_size
+
+    # Sanity check
+    if offset != avail_numel:
+        raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong")
+
+    print(f"Reconstructed Trainable fp32 state dict with {total_params} params {total_numel} elements")
+
+
+def _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states,
+                                               exclude_frozen_parameters):
+    state_dict = OrderedDict()
+
+    # buffers
+    buffers = zero_model_states[0].buffers
+    state_dict.update(buffers)
+    if debug:
+        print(f"added {len(buffers)} buffers")
+
+    if not exclude_frozen_parameters:
+        _zero3_merge_frozen_params(state_dict, world_size, zero_model_states)
+
+    _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states)
+
+    # recover shared parameters
+    for pair in zero_model_states[0].shared_params:
+        if pair[1] in state_dict:
+            state_dict[pair[0]] = state_dict[pair[1]]
+
+    return state_dict
+
+
+def to_torch_tensor(state_dict, return_empty_tensor=False):
+    """
+    Convert state_dict of GatheredTensor to torch tensor
+    """
+    torch_state_dict = {}
+    converted_tensors = {}
+    for name, tensor in state_dict.items():
+        tensor_id = id(tensor)
+        if tensor_id in converted_tensors:  # shared tensors
+            shared_tensor = torch_state_dict[converted_tensors[tensor_id]]
+            torch_state_dict[name] = shared_tensor
+        else:
+            converted_tensors[tensor_id] = name
+            if return_empty_tensor:
+                torch_state_dict[name] = torch.empty(tensor.shape, dtype=tensor.dtype)
+            else:
+                torch_state_dict[name] = tensor.contiguous()
+    return torch_state_dict
+
+
+def get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir,
+                                             tag=None,
+                                             exclude_frozen_parameters=False,
+                                             lazy_mode=False):
+    """
+    Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated state_dict that can be loaded with
+    ``load_state_dict()`` and used for training without DeepSpeed or shared with others, for example
+    via a model hub.
+
+    Args:
+        - ``checkpoint_dir``: path to the desired checkpoint folder
+        - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in 'latest' file. e.g., ``global_step14``
+        - ``exclude_frozen_parameters``: exclude frozen parameters
+        - ``lazy_mode``: get state_dict in lazy mode. It returns a dict of pesduo tensor instead of torch tensor, which is more memory efficient.
+          Convert the pesduo tensor to torch tensor by ``.contiguous()``
+
+    Returns:
+        - pytorch ``state_dict``
+
+    A typical usage might be ::
+
+        from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint
+        # do the training and checkpoint saving
+        state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir) # already on cpu
+        model = model.cpu() # move to cpu
+        model.load_state_dict(state_dict)
+        # submit to model hub or save the model to share with others
+
+    In this example the ``model`` will no longer be usable in the deepspeed context of the same
+    application. i.e. you will need to re-initialize the deepspeed engine, since
+    ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it.
+
+    If you want it all done for you, use ``load_state_dict_from_zero_checkpoint`` instead.
+
+    Note: the above usage may not work if your application doesn't have sufficient free CPU memory.
+    You may need to use the offline approach using the ``zero_to_fp32.py`` script that is saved with
+    the checkpoint. Or you can load state_dict in lazy mode ::
+
+        from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint
+        state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, lazy_mode=True) # not on cpu
+        for name, lazy_tensor in state_dict.item():
+            tensor = lazy_tensor.contiguous()  # to cpu
+            print(name, tensor)
+            # del tensor to release memory if it no longer in use
+    """
+    if tag is None:
+        latest_path = os.path.join(checkpoint_dir, 'latest')
+        if os.path.isfile(latest_path):
+            with open(latest_path, 'r') as fd:
+                tag = fd.read().strip()
+        else:
+            raise ValueError(f"Unable to find 'latest' file at {latest_path}")
+
+    ds_checkpoint_dir = os.path.join(checkpoint_dir, tag)
+
+    if not os.path.isdir(ds_checkpoint_dir):
+        raise FileNotFoundError(f"Directory '{ds_checkpoint_dir}' doesn't exist")
+
+    state_dict = _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters)
+    if lazy_mode:
+        return state_dict
+    else:
+        return to_torch_tensor(state_dict)
+
+
+def convert_zero_checkpoint_to_fp32_state_dict(checkpoint_dir,
+                                               output_dir,
+                                               max_shard_size="5GB",
+                                               safe_serialization=False,
+                                               tag=None,
+                                               exclude_frozen_parameters=False):
+    """
+    Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` file that can be
+    loaded with ``torch.load(file)`` + ``load_state_dict()`` and used for training without DeepSpeed.
+
+    Args:
+        - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``)
+        - ``output_dir``: directory to the pytorch fp32 state_dict output files
+        - ``max_shard_size``: the maximum size for a checkpoint before being sharded, default value is 5GB
+        - ``safe_serialization``:  whether to save the model using `safetensors` or the traditional PyTorch way (that uses `pickle`).
+        - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14``
+        - ``exclude_frozen_parameters``: exclude frozen parameters
+    """
+
+    # Dependency pre-check
+    if safe_serialization:
+        try:
+            from safetensors.torch import save_file
+        except ImportError:
+            print('If you want to use `safe_serialization`, please `pip install safetensors`')
+            raise
+    if max_shard_size is not None:
+        try:
+            from huggingface_hub import split_torch_state_dict_into_shards
+        except ImportError:
+            print('If you want to use `max_shard_size`, please `pip install huggingface_hub`')
+            raise
+
+    # Convert zero checkpoint to state_dict
+    state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir,
+                                                          tag,
+                                                          exclude_frozen_parameters,
+                                                          lazy_mode=True)
+
+    # Shard the model if it is too big.
+    weights_name = "model.safetensors" if safe_serialization else "pytorch_model.bin"
+    if max_shard_size is not None:
+        filename_pattern = weights_name.replace(".bin", "{suffix}.bin").replace(".safetensors", "{suffix}.safetensors")
+        # an memory-efficient approach for sharding
+        empty_state_dict = to_torch_tensor(state_dict, return_empty_tensor=True)
+        state_dict_split = split_torch_state_dict_into_shards(empty_state_dict,
+                                                              filename_pattern=filename_pattern,
+                                                              max_shard_size=max_shard_size)
+    else:
+        from collections import namedtuple
+        StateDictSplit = namedtuple("StateDictSplit", ["is_sharded", "filename_to_tensors"])
+        state_dict_split = StateDictSplit(is_sharded=False,
+                                          filename_to_tensors={weights_name: list(state_dict.keys())})
+
+    # Save the model by shard
+    os.makedirs(output_dir, exist_ok=True)
+    filename_to_tensors = state_dict_split.filename_to_tensors.items()
+    for shard_file, tensors in tqdm(filename_to_tensors, desc="Saving checkpoint shards"):
+        shard_state_dict = {tensor_name: state_dict[tensor_name] for tensor_name in tensors}
+        shard_state_dict = to_torch_tensor(shard_state_dict)
+        output_path = os.path.join(output_dir, shard_file)
+        if safe_serialization:
+            save_file(shard_state_dict, output_path, metadata={"format": "pt"})
+        else:
+            torch.save(shard_state_dict, output_path)
+        # release the memory of current shard
+        for tensor_name in list(shard_state_dict.keys()):
+            del state_dict[tensor_name]
+            del shard_state_dict[tensor_name]
+        del shard_state_dict
+        gc.collect()
+
+    # Save index if sharded
+    if state_dict_split.is_sharded:
+        index = {
+            "metadata": state_dict_split.metadata,
+            "weight_map": state_dict_split.tensor_to_filename,
+        }
+        save_index_file = "model.safetensors.index.json" if safe_serialization else "pytorch_model.bin.index.json"
+        save_index_file = os.path.join(output_dir, save_index_file)
+        with open(save_index_file, "w", encoding="utf-8") as f:
+            content = json.dumps(index, indent=2, sort_keys=True) + "\n"
+            f.write(content)
+
+
+def load_state_dict_from_zero_checkpoint(model, checkpoint_dir, tag=None):
+    """
+    1. Put the provided model to cpu
+    2. Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict``
+    3. Load it into the provided model
+
+    Args:
+        - ``model``: the model object to update
+        - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``)
+        - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14``
+
+    Returns:
+        - ``model`: modified model
+
+    Make sure you have plenty of CPU memory available before you call this function. If you don't
+    have enough use the ``zero_to_fp32.py`` utility to do the conversion. You will find it
+    conveniently placed for you in the checkpoint folder.
+
+    A typical usage might be ::
+
+        from deepspeed.utils.zero_to_fp32 import load_state_dict_from_zero_checkpoint
+        model = load_state_dict_from_zero_checkpoint(trainer.model, checkpoint_dir)
+        # submit to model hub or save the model to share with others
+
+    Note, that once this was run, the ``model`` will no longer be usable in the deepspeed context
+    of the same application. i.e. you will need to re-initialize the deepspeed engine, since
+    ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it.
+
+    """
+    logger.info(f"Extracting fp32 weights")
+    state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag)
+
+    logger.info(f"Overwriting model with fp32 weights")
+    model = model.cpu()
+    model.load_state_dict(state_dict, strict=False)
+
+    return model
+
+
+if __name__ == "__main__":
+    parser = argparse.ArgumentParser()
+    parser.add_argument("checkpoint_dir",
+                        type=str,
+                        help="path to the desired checkpoint folder, e.g., path/checkpoint-12")
+    parser.add_argument("output_dir",
+                        type=str,
+                        help="directory to the pytorch fp32 state_dict output files"
+                        "(e.g. path/checkpoint-12-output/)")
+    parser.add_argument(
+        "--max_shard_size",
+        type=str,
+        default="5GB",
+        help="The maximum size for a checkpoint before being sharded. Checkpoints shard will then be each of size"
+        "lower than this size. If expressed as a string, needs to be digits followed by a unit (like `5MB`"
+        "We default it to 5GB in order for models to be able to run easily on free-tier google colab instances"
+        "without CPU OOM issues.")
+    parser.add_argument(
+        "--safe_serialization",
+        default=False,
+        action='store_true',
+        help="Whether to save the model using `safetensors` or the traditional PyTorch way (that uses `pickle`).")
+    parser.add_argument("-t",
+                        "--tag",
+                        type=str,
+                        default=None,
+                        help="checkpoint tag used as a unique identifier for checkpoint. e.g., global_step1")
+    parser.add_argument("--exclude_frozen_parameters", action='store_true', help="exclude frozen parameters")
+    parser.add_argument("-d", "--debug", action='store_true', help="enable debug")
+    args = parser.parse_args()
+
+    debug = args.debug
+
+    convert_zero_checkpoint_to_fp32_state_dict(args.checkpoint_dir,
+                                               args.output_dir,
+                                               max_shard_size=args.max_shard_size,
+                                               safe_serialization=args.safe_serialization,
+                                               tag=args.tag,
+                                               exclude_frozen_parameters=args.exclude_frozen_parameters)
diff --git a/checkpoint-700/README.md b/checkpoint-700/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..85f3785e3148e4fdfa69be9cd1528c3af1891702
--- /dev/null
+++ b/checkpoint-700/README.md
@@ -0,0 +1,202 @@
+---
+base_model: mistralai/Mistral-Nemo-Instruct-2407
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.14.0
\ No newline at end of file
diff --git a/checkpoint-700/adapter_config.json b/checkpoint-700/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..1a10cbd48cf122bc7508a7cd9c23007d932168d0
--- /dev/null
+++ b/checkpoint-700/adapter_config.json
@@ -0,0 +1,37 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "mistralai/Mistral-Nemo-Instruct-2407",
+  "bias": "none",
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 16,
+  "lora_bias": false,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 16,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "k_proj",
+    "down_proj",
+    "q_proj",
+    "v_proj",
+    "o_proj",
+    "up_proj",
+    "gate_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/checkpoint-700/adapter_model.safetensors b/checkpoint-700/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..ebecadbdee3eb3b3e79731d45710443f7583e04d
--- /dev/null
+++ b/checkpoint-700/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0718a3f61096c82635a0fa020f1da1b390495cbfad404f9a43ba28c220b4db37
+size 114107976
diff --git a/checkpoint-700/global_step700/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt b/checkpoint-700/global_step700/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..66b4ce614c55697fe87717d4185a7b9260b3c36f
--- /dev/null
+++ b/checkpoint-700/global_step700/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b561ddf7af02bb548cb83580c5a24ae7ed3a3e34ab3559719c6be954cab61ad1
+size 49133072
diff --git a/checkpoint-700/global_step700/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt b/checkpoint-700/global_step700/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..0c1cd0790968ee737e48062b0cd08db49ce921bd
--- /dev/null
+++ b/checkpoint-700/global_step700/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4e375f6dbff9a96f383b68fc88bba42213dc51510fe429a9250f7fe3effd4914
+size 49133072
diff --git a/checkpoint-700/global_step700/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt b/checkpoint-700/global_step700/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..8a626bbf688a66880eff35bb8fff67769b962d4e
--- /dev/null
+++ b/checkpoint-700/global_step700/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7bb8b05356d4966d773a27bc63b039977d70b27bd11e90cdf01cbe4b08b625b5
+size 49133072
diff --git a/checkpoint-700/global_step700/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt b/checkpoint-700/global_step700/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..16beb5c1a131bf6f7a5ba8f24c5dbecde062bbc5
--- /dev/null
+++ b/checkpoint-700/global_step700/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:842386851db080acac286b36de14e7826f86f1bf4d9fafa3d97b703897924cdc
+size 49133072
diff --git a/checkpoint-700/global_step700/bf16_zero_pp_rank_4_mp_rank_00_optim_states.pt b/checkpoint-700/global_step700/bf16_zero_pp_rank_4_mp_rank_00_optim_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..fac696ff8de9f5d830279095e9896267595addbc
--- /dev/null
+++ b/checkpoint-700/global_step700/bf16_zero_pp_rank_4_mp_rank_00_optim_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8dfd9ca9e8db08da1e3bb258ee9a127b71093c7940b76b0caac6e4e8732cac85
+size 49133072
diff --git a/checkpoint-700/global_step700/bf16_zero_pp_rank_5_mp_rank_00_optim_states.pt b/checkpoint-700/global_step700/bf16_zero_pp_rank_5_mp_rank_00_optim_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..1fe78309b7049c4d88e6e5381d7965664afa7a84
--- /dev/null
+++ b/checkpoint-700/global_step700/bf16_zero_pp_rank_5_mp_rank_00_optim_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3ddedfa1e723871696eee2fcd6916eed0e87d6063cd355a99b4dc10d77ff69c9
+size 49133072
diff --git a/checkpoint-700/global_step700/bf16_zero_pp_rank_6_mp_rank_00_optim_states.pt b/checkpoint-700/global_step700/bf16_zero_pp_rank_6_mp_rank_00_optim_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..f2b7e515336bae5356d4dbbf5f4902c81440d7f5
--- /dev/null
+++ b/checkpoint-700/global_step700/bf16_zero_pp_rank_6_mp_rank_00_optim_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9d27872a5bddd36c3d38e565721d77bcf1c98aa80b5e835990592b6dcb39ffa0
+size 49133072
diff --git a/checkpoint-700/global_step700/zero_pp_rank_0_mp_rank_00_model_states.pt b/checkpoint-700/global_step700/zero_pp_rank_0_mp_rank_00_model_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..8d1ac8f898c5e9e02d8c1be0547c0baaf773f31b
--- /dev/null
+++ b/checkpoint-700/global_step700/zero_pp_rank_0_mp_rank_00_model_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1cdf505ac8b5abde67ff1af1e615cc4d1e383d9ba9c5e5c91660b64aec417208
+size 553886
diff --git a/checkpoint-700/global_step700/zero_pp_rank_1_mp_rank_00_model_states.pt b/checkpoint-700/global_step700/zero_pp_rank_1_mp_rank_00_model_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..19883de8a399cbb69e8276021209474a534ef69c
--- /dev/null
+++ b/checkpoint-700/global_step700/zero_pp_rank_1_mp_rank_00_model_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bd08f132b64022e0b8e81d7f8dd73c1aaf1019ad5f3865470a00493ab3607ccc
+size 553886
diff --git a/checkpoint-700/global_step700/zero_pp_rank_2_mp_rank_00_model_states.pt b/checkpoint-700/global_step700/zero_pp_rank_2_mp_rank_00_model_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..13913fcc211ee7517971e6d27c53242bcec5a5d0
--- /dev/null
+++ b/checkpoint-700/global_step700/zero_pp_rank_2_mp_rank_00_model_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:516344bce5e1ff541918da7556129048b2965050de2a29539b0093b346d22ca8
+size 553886
diff --git a/checkpoint-700/global_step700/zero_pp_rank_3_mp_rank_00_model_states.pt b/checkpoint-700/global_step700/zero_pp_rank_3_mp_rank_00_model_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..027e82e5012faa584c91b5b77bfdaf1142ce91b4
--- /dev/null
+++ b/checkpoint-700/global_step700/zero_pp_rank_3_mp_rank_00_model_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:20acb93182ee7726c5e0f913aec37eb70cc9bab4c654fc96e977e83403270dd5
+size 553886
diff --git a/checkpoint-700/global_step700/zero_pp_rank_4_mp_rank_00_model_states.pt b/checkpoint-700/global_step700/zero_pp_rank_4_mp_rank_00_model_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..79ef1d08e525c4d39f1f6aae9cc6a56ade67b080
--- /dev/null
+++ b/checkpoint-700/global_step700/zero_pp_rank_4_mp_rank_00_model_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7de8bff4d424fd33c89ee9fe05cd650d774d04246d8ba9f5cbb7e53aad104f2b
+size 553886
diff --git a/checkpoint-700/global_step700/zero_pp_rank_5_mp_rank_00_model_states.pt b/checkpoint-700/global_step700/zero_pp_rank_5_mp_rank_00_model_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..dc0052b142e0c4f330650614bad822f8755e8aaf
--- /dev/null
+++ b/checkpoint-700/global_step700/zero_pp_rank_5_mp_rank_00_model_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1240f4121f63b7a5f53670a21509fedb312151d0058461b3e63a6c23614bd660
+size 553886
diff --git a/checkpoint-700/global_step700/zero_pp_rank_6_mp_rank_00_model_states.pt b/checkpoint-700/global_step700/zero_pp_rank_6_mp_rank_00_model_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..1d669c809ec1bf7266b51db6e6df19a67f93e049
--- /dev/null
+++ b/checkpoint-700/global_step700/zero_pp_rank_6_mp_rank_00_model_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:76d4a12b717d4d9094edc94162b6fb455b01c15da30896b9811d585bb6a83ac7
+size 553886
diff --git a/checkpoint-700/latest b/checkpoint-700/latest
new file mode 100644
index 0000000000000000000000000000000000000000..a0f3e526aa9af2ac647b278f006bb9616843c5d6
--- /dev/null
+++ b/checkpoint-700/latest
@@ -0,0 +1 @@
+global_step700
\ No newline at end of file
diff --git a/checkpoint-700/rng_state_0.pth b/checkpoint-700/rng_state_0.pth
new file mode 100644
index 0000000000000000000000000000000000000000..219d153dc0a5c87e6984b11c4631b9adbc95761e
--- /dev/null
+++ b/checkpoint-700/rng_state_0.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a8d371e063544c289252718ad0a095ded4f3d16c53694706dca3bc6e4f140e8f
+size 15728
diff --git a/checkpoint-700/rng_state_1.pth b/checkpoint-700/rng_state_1.pth
new file mode 100644
index 0000000000000000000000000000000000000000..967d18184ad070b3c7a3079a25fd6d21d9fbd221
--- /dev/null
+++ b/checkpoint-700/rng_state_1.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2446736f46adfa52316c84703a89a915adaaf66f2257e49862692948fbe3c088
+size 15728
diff --git a/checkpoint-700/rng_state_2.pth b/checkpoint-700/rng_state_2.pth
new file mode 100644
index 0000000000000000000000000000000000000000..0378c55a282175d0fcf12d19c55b4be3868289af
--- /dev/null
+++ b/checkpoint-700/rng_state_2.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cde238dd928eeb625dd6fd6ffb859b06013036d3775e20e040930c6ff7af0994
+size 15728
diff --git a/checkpoint-700/rng_state_3.pth b/checkpoint-700/rng_state_3.pth
new file mode 100644
index 0000000000000000000000000000000000000000..fb070001e3f31ecd00750c71976d54bfa4f1356a
--- /dev/null
+++ b/checkpoint-700/rng_state_3.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a407892725a08bf7de1e5174bec02aae102a684e0a161cad6aca0ccdc959be17
+size 15792
diff --git a/checkpoint-700/rng_state_4.pth b/checkpoint-700/rng_state_4.pth
new file mode 100644
index 0000000000000000000000000000000000000000..ed847b3aba901a664252a2117a38e918147e80b0
--- /dev/null
+++ b/checkpoint-700/rng_state_4.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fb5d6ed8699a40a7b96bb5f6884a2f70f6763b5173de2165c2a4f5cc0dbce3aa
+size 15728
diff --git a/checkpoint-700/rng_state_5.pth b/checkpoint-700/rng_state_5.pth
new file mode 100644
index 0000000000000000000000000000000000000000..4c864b98af38b2260e141c1a7bc7ae01ead8aa6f
--- /dev/null
+++ b/checkpoint-700/rng_state_5.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bae32c5d369ebe932180e64c28e0df21303f5991d259afe0687205ff65301ab0
+size 15728
diff --git a/checkpoint-700/rng_state_6.pth b/checkpoint-700/rng_state_6.pth
new file mode 100644
index 0000000000000000000000000000000000000000..57039a2b1637902c7b75391d65bf5353a6802e01
--- /dev/null
+++ b/checkpoint-700/rng_state_6.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:62e167c5ae3766e8aaa003f3bdc8e8066e88ccffa382638bc1dff5a4e96d8d19
+size 15728
diff --git a/checkpoint-700/scheduler.pt b/checkpoint-700/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..6d6356982cf2f7ab8d88b020e37e31244a4fe19f
--- /dev/null
+++ b/checkpoint-700/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f33e8163075eb17bfca34b8a693b635a42b9acfbda64f35b6be7a3d1d6e0ecbd
+size 1064
diff --git a/checkpoint-700/special_tokens_map.json b/checkpoint-700/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..72ecfeeb7e14d244c936169d2ed139eeae235ef1
--- /dev/null
+++ b/checkpoint-700/special_tokens_map.json
@@ -0,0 +1,24 @@
+{
+  "bos_token": {
+    "content": "<s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "</s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "</s>",
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}
diff --git a/checkpoint-700/tokenizer.json b/checkpoint-700/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..b35b1c0f0bd5007afc7ad5976eeaf44988bd2067
--- /dev/null
+++ b/checkpoint-700/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9a1c103d223ee5dc2dde8307635f7d12581b40855198a38efcfeb6db4e08da69
+size 17078445
diff --git a/checkpoint-700/tokenizer_config.json b/checkpoint-700/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..550f3f7fc9144301cd0f17eeb0ad8746a5fa5354
--- /dev/null
+++ b/checkpoint-700/tokenizer_config.json
@@ -0,0 +1,8020 @@
+{
+  "add_bos_token": true,
+  "add_eos_token": false,
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "[INST]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "4": {
+      "content": "[/INST]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "5": {
+      "content": "[AVAILABLE_TOOLS]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "6": {
+      "content": "[/AVAILABLE_TOOLS]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "7": {
+      "content": "[TOOL_RESULTS]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "8": {
+      "content": "[/TOOL_RESULTS]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "9": {
+      "content": "[TOOL_CALLS]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "10": {
+      "content": "<pad>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "11": {
+      "content": "[PREFIX]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "12": {
+      "content": "[MIDDLE]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "13": {
+      "content": "[SUFFIX]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "14": {
+      "content": "<SPECIAL_14>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "15": {
+      "content": "<SPECIAL_15>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "16": {
+      "content": "<SPECIAL_16>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "17": {
+      "content": "<SPECIAL_17>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "18": {
+      "content": "<SPECIAL_18>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "19": {
+      "content": "<SPECIAL_19>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "20": {
+      "content": "<SPECIAL_20>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "21": {
+      "content": "<SPECIAL_21>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "22": {
+      "content": "<SPECIAL_22>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "23": {
+      "content": "<SPECIAL_23>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "24": {
+      "content": "<SPECIAL_24>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "25": {
+      "content": "<SPECIAL_25>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "26": {
+      "content": "<SPECIAL_26>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "27": {
+      "content": "<SPECIAL_27>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "28": {
+      "content": "<SPECIAL_28>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "29": {
+      "content": "<SPECIAL_29>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "30": {
+      "content": "<SPECIAL_30>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "31": {
+      "content": "<SPECIAL_31>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32": {
+      "content": "<SPECIAL_32>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "33": {
+      "content": "<SPECIAL_33>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "34": {
+      "content": "<SPECIAL_34>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "35": {
+      "content": "<SPECIAL_35>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "36": {
+      "content": "<SPECIAL_36>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "37": {
+      "content": "<SPECIAL_37>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "38": {
+      "content": "<SPECIAL_38>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "39": {
+      "content": "<SPECIAL_39>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "40": {
+      "content": "<SPECIAL_40>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "41": {
+      "content": "<SPECIAL_41>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "42": {
+      "content": "<SPECIAL_42>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "43": {
+      "content": "<SPECIAL_43>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "44": {
+      "content": "<SPECIAL_44>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "45": {
+      "content": "<SPECIAL_45>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "46": {
+      "content": "<SPECIAL_46>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "47": {
+      "content": "<SPECIAL_47>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "48": {
+      "content": "<SPECIAL_48>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "49": {
+      "content": "<SPECIAL_49>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "50": {
+      "content": "<SPECIAL_50>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "51": {
+      "content": "<SPECIAL_51>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "52": {
+      "content": "<SPECIAL_52>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "53": {
+      "content": "<SPECIAL_53>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "54": {
+      "content": "<SPECIAL_54>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "55": {
+      "content": "<SPECIAL_55>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "56": {
+      "content": "<SPECIAL_56>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "57": {
+      "content": "<SPECIAL_57>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "58": {
+      "content": "<SPECIAL_58>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "59": {
+      "content": "<SPECIAL_59>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "60": {
+      "content": "<SPECIAL_60>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "61": {
+      "content": "<SPECIAL_61>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "62": {
+      "content": "<SPECIAL_62>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "63": {
+      "content": "<SPECIAL_63>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "64": {
+      "content": "<SPECIAL_64>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "65": {
+      "content": "<SPECIAL_65>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "66": {
+      "content": "<SPECIAL_66>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "67": {
+      "content": "<SPECIAL_67>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "68": {
+      "content": "<SPECIAL_68>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "69": {
+      "content": "<SPECIAL_69>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "70": {
+      "content": "<SPECIAL_70>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "71": {
+      "content": "<SPECIAL_71>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "72": {
+      "content": "<SPECIAL_72>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "73": {
+      "content": "<SPECIAL_73>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "74": {
+      "content": "<SPECIAL_74>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "75": {
+      "content": "<SPECIAL_75>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "76": {
+      "content": "<SPECIAL_76>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "77": {
+      "content": "<SPECIAL_77>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "78": {
+      "content": "<SPECIAL_78>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "79": {
+      "content": "<SPECIAL_79>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "80": {
+      "content": "<SPECIAL_80>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "81": {
+      "content": "<SPECIAL_81>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "82": {
+      "content": "<SPECIAL_82>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "83": {
+      "content": "<SPECIAL_83>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "84": {
+      "content": "<SPECIAL_84>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "85": {
+      "content": "<SPECIAL_85>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "86": {
+      "content": "<SPECIAL_86>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "87": {
+      "content": "<SPECIAL_87>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "88": {
+      "content": "<SPECIAL_88>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "89": {
+      "content": "<SPECIAL_89>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "90": {
+      "content": "<SPECIAL_90>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "91": {
+      "content": "<SPECIAL_91>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "92": {
+      "content": "<SPECIAL_92>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "93": {
+      "content": "<SPECIAL_93>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "94": {
+      "content": "<SPECIAL_94>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "95": {
+      "content": "<SPECIAL_95>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "96": {
+      "content": "<SPECIAL_96>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "97": {
+      "content": "<SPECIAL_97>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "98": {
+      "content": "<SPECIAL_98>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "99": {
+      "content": "<SPECIAL_99>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "100": {
+      "content": "<SPECIAL_100>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "101": {
+      "content": "<SPECIAL_101>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "102": {
+      "content": "<SPECIAL_102>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "103": {
+      "content": "<SPECIAL_103>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "104": {
+      "content": "<SPECIAL_104>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "105": {
+      "content": "<SPECIAL_105>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "106": {
+      "content": "<SPECIAL_106>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "107": {
+      "content": "<SPECIAL_107>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "108": {
+      "content": "<SPECIAL_108>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "109": {
+      "content": "<SPECIAL_109>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "110": {
+      "content": "<SPECIAL_110>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "111": {
+      "content": "<SPECIAL_111>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "112": {
+      "content": "<SPECIAL_112>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "113": {
+      "content": "<SPECIAL_113>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "114": {
+      "content": "<SPECIAL_114>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "115": {
+      "content": "<SPECIAL_115>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "116": {
+      "content": "<SPECIAL_116>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "117": {
+      "content": "<SPECIAL_117>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "118": {
+      "content": "<SPECIAL_118>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "119": {
+      "content": "<SPECIAL_119>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "120": {
+      "content": "<SPECIAL_120>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "121": {
+      "content": "<SPECIAL_121>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "122": {
+      "content": "<SPECIAL_122>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "123": {
+      "content": "<SPECIAL_123>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "124": {
+      "content": "<SPECIAL_124>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "125": {
+      "content": "<SPECIAL_125>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "126": {
+      "content": "<SPECIAL_126>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "127": {
+      "content": "<SPECIAL_127>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128": {
+      "content": "<SPECIAL_128>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "129": {
+      "content": "<SPECIAL_129>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "130": {
+      "content": "<SPECIAL_130>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "131": {
+      "content": "<SPECIAL_131>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "132": {
+      "content": "<SPECIAL_132>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "133": {
+      "content": "<SPECIAL_133>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "134": {
+      "content": "<SPECIAL_134>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "135": {
+      "content": "<SPECIAL_135>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "136": {
+      "content": "<SPECIAL_136>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "137": {
+      "content": "<SPECIAL_137>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "138": {
+      "content": "<SPECIAL_138>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "139": {
+      "content": "<SPECIAL_139>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "140": {
+      "content": "<SPECIAL_140>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "141": {
+      "content": "<SPECIAL_141>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "142": {
+      "content": "<SPECIAL_142>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "143": {
+      "content": "<SPECIAL_143>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "144": {
+      "content": "<SPECIAL_144>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "145": {
+      "content": "<SPECIAL_145>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "146": {
+      "content": "<SPECIAL_146>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "147": {
+      "content": "<SPECIAL_147>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "148": {
+      "content": "<SPECIAL_148>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "149": {
+      "content": "<SPECIAL_149>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "150": {
+      "content": "<SPECIAL_150>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151": {
+      "content": "<SPECIAL_151>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "152": {
+      "content": "<SPECIAL_152>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "153": {
+      "content": "<SPECIAL_153>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "154": {
+      "content": "<SPECIAL_154>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "155": {
+      "content": "<SPECIAL_155>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "156": {
+      "content": "<SPECIAL_156>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "157": {
+      "content": "<SPECIAL_157>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "158": {
+      "content": "<SPECIAL_158>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "159": {
+      "content": "<SPECIAL_159>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "160": {
+      "content": "<SPECIAL_160>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "161": {
+      "content": "<SPECIAL_161>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "162": {
+      "content": "<SPECIAL_162>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "163": {
+      "content": "<SPECIAL_163>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "164": {
+      "content": "<SPECIAL_164>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "165": {
+      "content": "<SPECIAL_165>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "166": {
+      "content": "<SPECIAL_166>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "167": {
+      "content": "<SPECIAL_167>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "168": {
+      "content": "<SPECIAL_168>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "169": {
+      "content": "<SPECIAL_169>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170": {
+      "content": "<SPECIAL_170>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "171": {
+      "content": "<SPECIAL_171>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "172": {
+      "content": "<SPECIAL_172>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "173": {
+      "content": "<SPECIAL_173>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "174": {
+      "content": "<SPECIAL_174>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "175": {
+      "content": "<SPECIAL_175>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "176": {
+      "content": "<SPECIAL_176>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "177": {
+      "content": "<SPECIAL_177>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "178": {
+      "content": "<SPECIAL_178>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "179": {
+      "content": "<SPECIAL_179>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "180": {
+      "content": "<SPECIAL_180>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "181": {
+      "content": "<SPECIAL_181>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "182": {
+      "content": "<SPECIAL_182>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "183": {
+      "content": "<SPECIAL_183>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "184": {
+      "content": "<SPECIAL_184>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "185": {
+      "content": "<SPECIAL_185>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "186": {
+      "content": "<SPECIAL_186>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "187": {
+      "content": "<SPECIAL_187>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "188": {
+      "content": "<SPECIAL_188>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "189": {
+      "content": "<SPECIAL_189>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "190": {
+      "content": "<SPECIAL_190>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "191": {
+      "content": "<SPECIAL_191>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "192": {
+      "content": "<SPECIAL_192>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "193": {
+      "content": "<SPECIAL_193>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "194": {
+      "content": "<SPECIAL_194>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "195": {
+      "content": "<SPECIAL_195>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "196": {
+      "content": "<SPECIAL_196>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "197": {
+      "content": "<SPECIAL_197>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "198": {
+      "content": "<SPECIAL_198>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "199": {
+      "content": "<SPECIAL_199>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "200": {
+      "content": "<SPECIAL_200>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "201": {
+      "content": "<SPECIAL_201>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "202": {
+      "content": "<SPECIAL_202>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "203": {
+      "content": "<SPECIAL_203>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "204": {
+      "content": "<SPECIAL_204>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "205": {
+      "content": "<SPECIAL_205>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "206": {
+      "content": "<SPECIAL_206>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "207": {
+      "content": "<SPECIAL_207>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "208": {
+      "content": "<SPECIAL_208>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "209": {
+      "content": "<SPECIAL_209>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "210": {
+      "content": "<SPECIAL_210>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "211": {
+      "content": "<SPECIAL_211>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "212": {
+      "content": "<SPECIAL_212>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "213": {
+      "content": "<SPECIAL_213>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "214": {
+      "content": "<SPECIAL_214>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "215": {
+      "content": "<SPECIAL_215>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "216": {
+      "content": "<SPECIAL_216>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "217": {
+      "content": "<SPECIAL_217>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "218": {
+      "content": "<SPECIAL_218>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "219": {
+      "content": "<SPECIAL_219>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "220": {
+      "content": "<SPECIAL_220>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "221": {
+      "content": "<SPECIAL_221>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "222": {
+      "content": "<SPECIAL_222>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "223": {
+      "content": "<SPECIAL_223>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "224": {
+      "content": "<SPECIAL_224>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "225": {
+      "content": "<SPECIAL_225>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "226": {
+      "content": "<SPECIAL_226>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "227": {
+      "content": "<SPECIAL_227>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "228": {
+      "content": "<SPECIAL_228>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "229": {
+      "content": "<SPECIAL_229>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "230": {
+      "content": "<SPECIAL_230>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "231": {
+      "content": "<SPECIAL_231>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "232": {
+      "content": "<SPECIAL_232>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "233": {
+      "content": "<SPECIAL_233>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "234": {
+      "content": "<SPECIAL_234>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "235": {
+      "content": "<SPECIAL_235>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "236": {
+      "content": "<SPECIAL_236>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "237": {
+      "content": "<SPECIAL_237>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "238": {
+      "content": "<SPECIAL_238>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "239": {
+      "content": "<SPECIAL_239>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "240": {
+      "content": "<SPECIAL_240>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "241": {
+      "content": "<SPECIAL_241>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "242": {
+      "content": "<SPECIAL_242>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "243": {
+      "content": "<SPECIAL_243>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "244": {
+      "content": "<SPECIAL_244>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "245": {
+      "content": "<SPECIAL_245>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "246": {
+      "content": "<SPECIAL_246>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "247": {
+      "content": "<SPECIAL_247>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "248": {
+      "content": "<SPECIAL_248>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "249": {
+      "content": "<SPECIAL_249>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250": {
+      "content": "<SPECIAL_250>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "251": {
+      "content": "<SPECIAL_251>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "252": {
+      "content": "<SPECIAL_252>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "253": {
+      "content": "<SPECIAL_253>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "254": {
+      "content": "<SPECIAL_254>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "255": {
+      "content": "<SPECIAL_255>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "256": {
+      "content": "<SPECIAL_256>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "257": {
+      "content": "<SPECIAL_257>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "258": {
+      "content": "<SPECIAL_258>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "259": {
+      "content": "<SPECIAL_259>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "260": {
+      "content": "<SPECIAL_260>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "261": {
+      "content": "<SPECIAL_261>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "262": {
+      "content": "<SPECIAL_262>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "263": {
+      "content": "<SPECIAL_263>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "264": {
+      "content": "<SPECIAL_264>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "265": {
+      "content": "<SPECIAL_265>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "266": {
+      "content": "<SPECIAL_266>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "267": {
+      "content": "<SPECIAL_267>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "268": {
+      "content": "<SPECIAL_268>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "269": {
+      "content": "<SPECIAL_269>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "270": {
+      "content": "<SPECIAL_270>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "271": {
+      "content": "<SPECIAL_271>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "272": {
+      "content": "<SPECIAL_272>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "273": {
+      "content": "<SPECIAL_273>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "274": {
+      "content": "<SPECIAL_274>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "275": {
+      "content": "<SPECIAL_275>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "276": {
+      "content": "<SPECIAL_276>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "277": {
+      "content": "<SPECIAL_277>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "278": {
+      "content": "<SPECIAL_278>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "279": {
+      "content": "<SPECIAL_279>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "280": {
+      "content": "<SPECIAL_280>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "281": {
+      "content": "<SPECIAL_281>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "282": {
+      "content": "<SPECIAL_282>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "283": {
+      "content": "<SPECIAL_283>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "284": {
+      "content": "<SPECIAL_284>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "285": {
+      "content": "<SPECIAL_285>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "286": {
+      "content": "<SPECIAL_286>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "287": {
+      "content": "<SPECIAL_287>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "288": {
+      "content": "<SPECIAL_288>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "289": {
+      "content": "<SPECIAL_289>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "290": {
+      "content": "<SPECIAL_290>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "291": {
+      "content": "<SPECIAL_291>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "292": {
+      "content": "<SPECIAL_292>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "293": {
+      "content": "<SPECIAL_293>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "294": {
+      "content": "<SPECIAL_294>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "295": {
+      "content": "<SPECIAL_295>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "296": {
+      "content": "<SPECIAL_296>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "297": {
+      "content": "<SPECIAL_297>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "298": {
+      "content": "<SPECIAL_298>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "299": {
+      "content": "<SPECIAL_299>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "300": {
+      "content": "<SPECIAL_300>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "301": {
+      "content": "<SPECIAL_301>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "302": {
+      "content": "<SPECIAL_302>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "303": {
+      "content": "<SPECIAL_303>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "304": {
+      "content": "<SPECIAL_304>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "305": {
+      "content": "<SPECIAL_305>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "306": {
+      "content": "<SPECIAL_306>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "307": {
+      "content": "<SPECIAL_307>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "308": {
+      "content": "<SPECIAL_308>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "309": {
+      "content": "<SPECIAL_309>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "310": {
+      "content": "<SPECIAL_310>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "311": {
+      "content": "<SPECIAL_311>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "312": {
+      "content": "<SPECIAL_312>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "313": {
+      "content": "<SPECIAL_313>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "314": {
+      "content": "<SPECIAL_314>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "315": {
+      "content": "<SPECIAL_315>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "316": {
+      "content": "<SPECIAL_316>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "317": {
+      "content": "<SPECIAL_317>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "318": {
+      "content": "<SPECIAL_318>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "319": {
+      "content": "<SPECIAL_319>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "320": {
+      "content": "<SPECIAL_320>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "321": {
+      "content": "<SPECIAL_321>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "322": {
+      "content": "<SPECIAL_322>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "323": {
+      "content": "<SPECIAL_323>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "324": {
+      "content": "<SPECIAL_324>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "325": {
+      "content": "<SPECIAL_325>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "326": {
+      "content": "<SPECIAL_326>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "327": {
+      "content": "<SPECIAL_327>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "328": {
+      "content": "<SPECIAL_328>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "329": {
+      "content": "<SPECIAL_329>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "330": {
+      "content": "<SPECIAL_330>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "331": {
+      "content": "<SPECIAL_331>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "332": {
+      "content": "<SPECIAL_332>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "333": {
+      "content": "<SPECIAL_333>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "334": {
+      "content": "<SPECIAL_334>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "335": {
+      "content": "<SPECIAL_335>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "336": {
+      "content": "<SPECIAL_336>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "337": {
+      "content": "<SPECIAL_337>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "338": {
+      "content": "<SPECIAL_338>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "339": {
+      "content": "<SPECIAL_339>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "340": {
+      "content": "<SPECIAL_340>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "341": {
+      "content": "<SPECIAL_341>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "342": {
+      "content": "<SPECIAL_342>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "343": {
+      "content": "<SPECIAL_343>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "344": {
+      "content": "<SPECIAL_344>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "345": {
+      "content": "<SPECIAL_345>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "346": {
+      "content": "<SPECIAL_346>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "347": {
+      "content": "<SPECIAL_347>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "348": {
+      "content": "<SPECIAL_348>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "349": {
+      "content": "<SPECIAL_349>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "350": {
+      "content": "<SPECIAL_350>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "351": {
+      "content": "<SPECIAL_351>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "352": {
+      "content": "<SPECIAL_352>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "353": {
+      "content": "<SPECIAL_353>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "354": {
+      "content": "<SPECIAL_354>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "355": {
+      "content": "<SPECIAL_355>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "356": {
+      "content": "<SPECIAL_356>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "357": {
+      "content": "<SPECIAL_357>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "358": {
+      "content": "<SPECIAL_358>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "359": {
+      "content": "<SPECIAL_359>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "360": {
+      "content": "<SPECIAL_360>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "361": {
+      "content": "<SPECIAL_361>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "362": {
+      "content": "<SPECIAL_362>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "363": {
+      "content": "<SPECIAL_363>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "364": {
+      "content": "<SPECIAL_364>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "365": {
+      "content": "<SPECIAL_365>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "366": {
+      "content": "<SPECIAL_366>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "367": {
+      "content": "<SPECIAL_367>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "368": {
+      "content": "<SPECIAL_368>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "369": {
+      "content": "<SPECIAL_369>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "370": {
+      "content": "<SPECIAL_370>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "371": {
+      "content": "<SPECIAL_371>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "372": {
+      "content": "<SPECIAL_372>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "373": {
+      "content": "<SPECIAL_373>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "374": {
+      "content": "<SPECIAL_374>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "375": {
+      "content": "<SPECIAL_375>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "376": {
+      "content": "<SPECIAL_376>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "377": {
+      "content": "<SPECIAL_377>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "378": {
+      "content": "<SPECIAL_378>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "379": {
+      "content": "<SPECIAL_379>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "380": {
+      "content": "<SPECIAL_380>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "381": {
+      "content": "<SPECIAL_381>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "382": {
+      "content": "<SPECIAL_382>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "383": {
+      "content": "<SPECIAL_383>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "384": {
+      "content": "<SPECIAL_384>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "385": {
+      "content": "<SPECIAL_385>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "386": {
+      "content": "<SPECIAL_386>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "387": {
+      "content": "<SPECIAL_387>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "388": {
+      "content": "<SPECIAL_388>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "389": {
+      "content": "<SPECIAL_389>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "390": {
+      "content": "<SPECIAL_390>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "391": {
+      "content": "<SPECIAL_391>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "392": {
+      "content": "<SPECIAL_392>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "393": {
+      "content": "<SPECIAL_393>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "394": {
+      "content": "<SPECIAL_394>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "395": {
+      "content": "<SPECIAL_395>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "396": {
+      "content": "<SPECIAL_396>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "397": {
+      "content": "<SPECIAL_397>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "398": {
+      "content": "<SPECIAL_398>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "399": {
+      "content": "<SPECIAL_399>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "400": {
+      "content": "<SPECIAL_400>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "401": {
+      "content": "<SPECIAL_401>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "402": {
+      "content": "<SPECIAL_402>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "403": {
+      "content": "<SPECIAL_403>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "404": {
+      "content": "<SPECIAL_404>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "405": {
+      "content": "<SPECIAL_405>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "406": {
+      "content": "<SPECIAL_406>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "407": {
+      "content": "<SPECIAL_407>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "408": {
+      "content": "<SPECIAL_408>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "409": {
+      "content": "<SPECIAL_409>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "410": {
+      "content": "<SPECIAL_410>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "411": {
+      "content": "<SPECIAL_411>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "412": {
+      "content": "<SPECIAL_412>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "413": {
+      "content": "<SPECIAL_413>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "414": {
+      "content": "<SPECIAL_414>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "415": {
+      "content": "<SPECIAL_415>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "416": {
+      "content": "<SPECIAL_416>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "417": {
+      "content": "<SPECIAL_417>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "418": {
+      "content": "<SPECIAL_418>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "419": {
+      "content": "<SPECIAL_419>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "420": {
+      "content": "<SPECIAL_420>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "421": {
+      "content": "<SPECIAL_421>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "422": {
+      "content": "<SPECIAL_422>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "423": {
+      "content": "<SPECIAL_423>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "424": {
+      "content": "<SPECIAL_424>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "425": {
+      "content": "<SPECIAL_425>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "426": {
+      "content": "<SPECIAL_426>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "427": {
+      "content": "<SPECIAL_427>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "428": {
+      "content": "<SPECIAL_428>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "429": {
+      "content": "<SPECIAL_429>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "430": {
+      "content": "<SPECIAL_430>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "431": {
+      "content": "<SPECIAL_431>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "432": {
+      "content": "<SPECIAL_432>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "433": {
+      "content": "<SPECIAL_433>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "434": {
+      "content": "<SPECIAL_434>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "435": {
+      "content": "<SPECIAL_435>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "436": {
+      "content": "<SPECIAL_436>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "437": {
+      "content": "<SPECIAL_437>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "438": {
+      "content": "<SPECIAL_438>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "439": {
+      "content": "<SPECIAL_439>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "440": {
+      "content": "<SPECIAL_440>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "441": {
+      "content": "<SPECIAL_441>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "442": {
+      "content": "<SPECIAL_442>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "443": {
+      "content": "<SPECIAL_443>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "444": {
+      "content": "<SPECIAL_444>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "445": {
+      "content": "<SPECIAL_445>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "446": {
+      "content": "<SPECIAL_446>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "447": {
+      "content": "<SPECIAL_447>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "448": {
+      "content": "<SPECIAL_448>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "449": {
+      "content": "<SPECIAL_449>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "450": {
+      "content": "<SPECIAL_450>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "451": {
+      "content": "<SPECIAL_451>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "452": {
+      "content": "<SPECIAL_452>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "453": {
+      "content": "<SPECIAL_453>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "454": {
+      "content": "<SPECIAL_454>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "455": {
+      "content": "<SPECIAL_455>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "456": {
+      "content": "<SPECIAL_456>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "457": {
+      "content": "<SPECIAL_457>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "458": {
+      "content": "<SPECIAL_458>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "459": {
+      "content": "<SPECIAL_459>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "460": {
+      "content": "<SPECIAL_460>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "461": {
+      "content": "<SPECIAL_461>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "462": {
+      "content": "<SPECIAL_462>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "463": {
+      "content": "<SPECIAL_463>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "464": {
+      "content": "<SPECIAL_464>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "465": {
+      "content": "<SPECIAL_465>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "466": {
+      "content": "<SPECIAL_466>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "467": {
+      "content": "<SPECIAL_467>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "468": {
+      "content": "<SPECIAL_468>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "469": {
+      "content": "<SPECIAL_469>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "470": {
+      "content": "<SPECIAL_470>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "471": {
+      "content": "<SPECIAL_471>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "472": {
+      "content": "<SPECIAL_472>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "473": {
+      "content": "<SPECIAL_473>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "474": {
+      "content": "<SPECIAL_474>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "475": {
+      "content": "<SPECIAL_475>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "476": {
+      "content": "<SPECIAL_476>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "477": {
+      "content": "<SPECIAL_477>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "478": {
+      "content": "<SPECIAL_478>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "479": {
+      "content": "<SPECIAL_479>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "480": {
+      "content": "<SPECIAL_480>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "481": {
+      "content": "<SPECIAL_481>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "482": {
+      "content": "<SPECIAL_482>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "483": {
+      "content": "<SPECIAL_483>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "484": {
+      "content": "<SPECIAL_484>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "485": {
+      "content": "<SPECIAL_485>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "486": {
+      "content": "<SPECIAL_486>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "487": {
+      "content": "<SPECIAL_487>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "488": {
+      "content": "<SPECIAL_488>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "489": {
+      "content": "<SPECIAL_489>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "490": {
+      "content": "<SPECIAL_490>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "491": {
+      "content": "<SPECIAL_491>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "492": {
+      "content": "<SPECIAL_492>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "493": {
+      "content": "<SPECIAL_493>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "494": {
+      "content": "<SPECIAL_494>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "495": {
+      "content": "<SPECIAL_495>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "496": {
+      "content": "<SPECIAL_496>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "497": {
+      "content": "<SPECIAL_497>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "498": {
+      "content": "<SPECIAL_498>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "499": {
+      "content": "<SPECIAL_499>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "500": {
+      "content": "<SPECIAL_500>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "501": {
+      "content": "<SPECIAL_501>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "502": {
+      "content": "<SPECIAL_502>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "503": {
+      "content": "<SPECIAL_503>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "504": {
+      "content": "<SPECIAL_504>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "505": {
+      "content": "<SPECIAL_505>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "506": {
+      "content": "<SPECIAL_506>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "507": {
+      "content": "<SPECIAL_507>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "508": {
+      "content": "<SPECIAL_508>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "509": {
+      "content": "<SPECIAL_509>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "510": {
+      "content": "<SPECIAL_510>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "511": {
+      "content": "<SPECIAL_511>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "512": {
+      "content": "<SPECIAL_512>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "513": {
+      "content": "<SPECIAL_513>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "514": {
+      "content": "<SPECIAL_514>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "515": {
+      "content": "<SPECIAL_515>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "516": {
+      "content": "<SPECIAL_516>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "517": {
+      "content": "<SPECIAL_517>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "518": {
+      "content": "<SPECIAL_518>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "519": {
+      "content": "<SPECIAL_519>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "520": {
+      "content": "<SPECIAL_520>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "521": {
+      "content": "<SPECIAL_521>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "522": {
+      "content": "<SPECIAL_522>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "523": {
+      "content": "<SPECIAL_523>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "524": {
+      "content": "<SPECIAL_524>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "525": {
+      "content": "<SPECIAL_525>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "526": {
+      "content": "<SPECIAL_526>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "527": {
+      "content": "<SPECIAL_527>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "528": {
+      "content": "<SPECIAL_528>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "529": {
+      "content": "<SPECIAL_529>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "530": {
+      "content": "<SPECIAL_530>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "531": {
+      "content": "<SPECIAL_531>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "532": {
+      "content": "<SPECIAL_532>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "533": {
+      "content": "<SPECIAL_533>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "534": {
+      "content": "<SPECIAL_534>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "535": {
+      "content": "<SPECIAL_535>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "536": {
+      "content": "<SPECIAL_536>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "537": {
+      "content": "<SPECIAL_537>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "538": {
+      "content": "<SPECIAL_538>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "539": {
+      "content": "<SPECIAL_539>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "540": {
+      "content": "<SPECIAL_540>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "541": {
+      "content": "<SPECIAL_541>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "542": {
+      "content": "<SPECIAL_542>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "543": {
+      "content": "<SPECIAL_543>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "544": {
+      "content": "<SPECIAL_544>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "545": {
+      "content": "<SPECIAL_545>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "546": {
+      "content": "<SPECIAL_546>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "547": {
+      "content": "<SPECIAL_547>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "548": {
+      "content": "<SPECIAL_548>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "549": {
+      "content": "<SPECIAL_549>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "550": {
+      "content": "<SPECIAL_550>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "551": {
+      "content": "<SPECIAL_551>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "552": {
+      "content": "<SPECIAL_552>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "553": {
+      "content": "<SPECIAL_553>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "554": {
+      "content": "<SPECIAL_554>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "555": {
+      "content": "<SPECIAL_555>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "556": {
+      "content": "<SPECIAL_556>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "557": {
+      "content": "<SPECIAL_557>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "558": {
+      "content": "<SPECIAL_558>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "559": {
+      "content": "<SPECIAL_559>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "560": {
+      "content": "<SPECIAL_560>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "561": {
+      "content": "<SPECIAL_561>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "562": {
+      "content": "<SPECIAL_562>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "563": {
+      "content": "<SPECIAL_563>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "564": {
+      "content": "<SPECIAL_564>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "565": {
+      "content": "<SPECIAL_565>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "566": {
+      "content": "<SPECIAL_566>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "567": {
+      "content": "<SPECIAL_567>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "568": {
+      "content": "<SPECIAL_568>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "569": {
+      "content": "<SPECIAL_569>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "570": {
+      "content": "<SPECIAL_570>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "571": {
+      "content": "<SPECIAL_571>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "572": {
+      "content": "<SPECIAL_572>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "573": {
+      "content": "<SPECIAL_573>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "574": {
+      "content": "<SPECIAL_574>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "575": {
+      "content": "<SPECIAL_575>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "576": {
+      "content": "<SPECIAL_576>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "577": {
+      "content": "<SPECIAL_577>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "578": {
+      "content": "<SPECIAL_578>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "579": {
+      "content": "<SPECIAL_579>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "580": {
+      "content": "<SPECIAL_580>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "581": {
+      "content": "<SPECIAL_581>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "582": {
+      "content": "<SPECIAL_582>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "583": {
+      "content": "<SPECIAL_583>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "584": {
+      "content": "<SPECIAL_584>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "585": {
+      "content": "<SPECIAL_585>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "586": {
+      "content": "<SPECIAL_586>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "587": {
+      "content": "<SPECIAL_587>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "588": {
+      "content": "<SPECIAL_588>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "589": {
+      "content": "<SPECIAL_589>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "590": {
+      "content": "<SPECIAL_590>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "591": {
+      "content": "<SPECIAL_591>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "592": {
+      "content": "<SPECIAL_592>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "593": {
+      "content": "<SPECIAL_593>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "594": {
+      "content": "<SPECIAL_594>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "595": {
+      "content": "<SPECIAL_595>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "596": {
+      "content": "<SPECIAL_596>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "597": {
+      "content": "<SPECIAL_597>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "598": {
+      "content": "<SPECIAL_598>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "599": {
+      "content": "<SPECIAL_599>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "600": {
+      "content": "<SPECIAL_600>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "601": {
+      "content": "<SPECIAL_601>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "602": {
+      "content": "<SPECIAL_602>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "603": {
+      "content": "<SPECIAL_603>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "604": {
+      "content": "<SPECIAL_604>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "605": {
+      "content": "<SPECIAL_605>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "606": {
+      "content": "<SPECIAL_606>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "607": {
+      "content": "<SPECIAL_607>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "608": {
+      "content": "<SPECIAL_608>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "609": {
+      "content": "<SPECIAL_609>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "610": {
+      "content": "<SPECIAL_610>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "611": {
+      "content": "<SPECIAL_611>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "612": {
+      "content": "<SPECIAL_612>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "613": {
+      "content": "<SPECIAL_613>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "614": {
+      "content": "<SPECIAL_614>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "615": {
+      "content": "<SPECIAL_615>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "616": {
+      "content": "<SPECIAL_616>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "617": {
+      "content": "<SPECIAL_617>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "618": {
+      "content": "<SPECIAL_618>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "619": {
+      "content": "<SPECIAL_619>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "620": {
+      "content": "<SPECIAL_620>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "621": {
+      "content": "<SPECIAL_621>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "622": {
+      "content": "<SPECIAL_622>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "623": {
+      "content": "<SPECIAL_623>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "624": {
+      "content": "<SPECIAL_624>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "625": {
+      "content": "<SPECIAL_625>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "626": {
+      "content": "<SPECIAL_626>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "627": {
+      "content": "<SPECIAL_627>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "628": {
+      "content": "<SPECIAL_628>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "629": {
+      "content": "<SPECIAL_629>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "630": {
+      "content": "<SPECIAL_630>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "631": {
+      "content": "<SPECIAL_631>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "632": {
+      "content": "<SPECIAL_632>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "633": {
+      "content": "<SPECIAL_633>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "634": {
+      "content": "<SPECIAL_634>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "635": {
+      "content": "<SPECIAL_635>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "636": {
+      "content": "<SPECIAL_636>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "637": {
+      "content": "<SPECIAL_637>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "638": {
+      "content": "<SPECIAL_638>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "639": {
+      "content": "<SPECIAL_639>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "640": {
+      "content": "<SPECIAL_640>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "641": {
+      "content": "<SPECIAL_641>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "642": {
+      "content": "<SPECIAL_642>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "643": {
+      "content": "<SPECIAL_643>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "644": {
+      "content": "<SPECIAL_644>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "645": {
+      "content": "<SPECIAL_645>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "646": {
+      "content": "<SPECIAL_646>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "647": {
+      "content": "<SPECIAL_647>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "648": {
+      "content": "<SPECIAL_648>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "649": {
+      "content": "<SPECIAL_649>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "650": {
+      "content": "<SPECIAL_650>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "651": {
+      "content": "<SPECIAL_651>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "652": {
+      "content": "<SPECIAL_652>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "653": {
+      "content": "<SPECIAL_653>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "654": {
+      "content": "<SPECIAL_654>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "655": {
+      "content": "<SPECIAL_655>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "656": {
+      "content": "<SPECIAL_656>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "657": {
+      "content": "<SPECIAL_657>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "658": {
+      "content": "<SPECIAL_658>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "659": {
+      "content": "<SPECIAL_659>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "660": {
+      "content": "<SPECIAL_660>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "661": {
+      "content": "<SPECIAL_661>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "662": {
+      "content": "<SPECIAL_662>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "663": {
+      "content": "<SPECIAL_663>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "664": {
+      "content": "<SPECIAL_664>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "665": {
+      "content": "<SPECIAL_665>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "666": {
+      "content": "<SPECIAL_666>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "667": {
+      "content": "<SPECIAL_667>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "668": {
+      "content": "<SPECIAL_668>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "669": {
+      "content": "<SPECIAL_669>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "670": {
+      "content": "<SPECIAL_670>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "671": {
+      "content": "<SPECIAL_671>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "672": {
+      "content": "<SPECIAL_672>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "673": {
+      "content": "<SPECIAL_673>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "674": {
+      "content": "<SPECIAL_674>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "675": {
+      "content": "<SPECIAL_675>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "676": {
+      "content": "<SPECIAL_676>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "677": {
+      "content": "<SPECIAL_677>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "678": {
+      "content": "<SPECIAL_678>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "679": {
+      "content": "<SPECIAL_679>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "680": {
+      "content": "<SPECIAL_680>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "681": {
+      "content": "<SPECIAL_681>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "682": {
+      "content": "<SPECIAL_682>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "683": {
+      "content": "<SPECIAL_683>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "684": {
+      "content": "<SPECIAL_684>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "685": {
+      "content": "<SPECIAL_685>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "686": {
+      "content": "<SPECIAL_686>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "687": {
+      "content": "<SPECIAL_687>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "688": {
+      "content": "<SPECIAL_688>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "689": {
+      "content": "<SPECIAL_689>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "690": {
+      "content": "<SPECIAL_690>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "691": {
+      "content": "<SPECIAL_691>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "692": {
+      "content": "<SPECIAL_692>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "693": {
+      "content": "<SPECIAL_693>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "694": {
+      "content": "<SPECIAL_694>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "695": {
+      "content": "<SPECIAL_695>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "696": {
+      "content": "<SPECIAL_696>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "697": {
+      "content": "<SPECIAL_697>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "698": {
+      "content": "<SPECIAL_698>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "699": {
+      "content": "<SPECIAL_699>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "700": {
+      "content": "<SPECIAL_700>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "701": {
+      "content": "<SPECIAL_701>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "702": {
+      "content": "<SPECIAL_702>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "703": {
+      "content": "<SPECIAL_703>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "704": {
+      "content": "<SPECIAL_704>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "705": {
+      "content": "<SPECIAL_705>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "706": {
+      "content": "<SPECIAL_706>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "707": {
+      "content": "<SPECIAL_707>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "708": {
+      "content": "<SPECIAL_708>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "709": {
+      "content": "<SPECIAL_709>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "710": {
+      "content": "<SPECIAL_710>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "711": {
+      "content": "<SPECIAL_711>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "712": {
+      "content": "<SPECIAL_712>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "713": {
+      "content": "<SPECIAL_713>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "714": {
+      "content": "<SPECIAL_714>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "715": {
+      "content": "<SPECIAL_715>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "716": {
+      "content": "<SPECIAL_716>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "717": {
+      "content": "<SPECIAL_717>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "718": {
+      "content": "<SPECIAL_718>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "719": {
+      "content": "<SPECIAL_719>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "720": {
+      "content": "<SPECIAL_720>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "721": {
+      "content": "<SPECIAL_721>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "722": {
+      "content": "<SPECIAL_722>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "723": {
+      "content": "<SPECIAL_723>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "724": {
+      "content": "<SPECIAL_724>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "725": {
+      "content": "<SPECIAL_725>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "726": {
+      "content": "<SPECIAL_726>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "727": {
+      "content": "<SPECIAL_727>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "728": {
+      "content": "<SPECIAL_728>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "729": {
+      "content": "<SPECIAL_729>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "730": {
+      "content": "<SPECIAL_730>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "731": {
+      "content": "<SPECIAL_731>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "732": {
+      "content": "<SPECIAL_732>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "733": {
+      "content": "<SPECIAL_733>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "734": {
+      "content": "<SPECIAL_734>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "735": {
+      "content": "<SPECIAL_735>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "736": {
+      "content": "<SPECIAL_736>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "737": {
+      "content": "<SPECIAL_737>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "738": {
+      "content": "<SPECIAL_738>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "739": {
+      "content": "<SPECIAL_739>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "740": {
+      "content": "<SPECIAL_740>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "741": {
+      "content": "<SPECIAL_741>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "742": {
+      "content": "<SPECIAL_742>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "743": {
+      "content": "<SPECIAL_743>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "744": {
+      "content": "<SPECIAL_744>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "745": {
+      "content": "<SPECIAL_745>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "746": {
+      "content": "<SPECIAL_746>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "747": {
+      "content": "<SPECIAL_747>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "748": {
+      "content": "<SPECIAL_748>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "749": {
+      "content": "<SPECIAL_749>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "750": {
+      "content": "<SPECIAL_750>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "751": {
+      "content": "<SPECIAL_751>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "752": {
+      "content": "<SPECIAL_752>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "753": {
+      "content": "<SPECIAL_753>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "754": {
+      "content": "<SPECIAL_754>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "755": {
+      "content": "<SPECIAL_755>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "756": {
+      "content": "<SPECIAL_756>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "757": {
+      "content": "<SPECIAL_757>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "758": {
+      "content": "<SPECIAL_758>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "759": {
+      "content": "<SPECIAL_759>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "760": {
+      "content": "<SPECIAL_760>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "761": {
+      "content": "<SPECIAL_761>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "762": {
+      "content": "<SPECIAL_762>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "763": {
+      "content": "<SPECIAL_763>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "764": {
+      "content": "<SPECIAL_764>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "765": {
+      "content": "<SPECIAL_765>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "766": {
+      "content": "<SPECIAL_766>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "767": {
+      "content": "<SPECIAL_767>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "768": {
+      "content": "<SPECIAL_768>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "769": {
+      "content": "<SPECIAL_769>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "770": {
+      "content": "<SPECIAL_770>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "771": {
+      "content": "<SPECIAL_771>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "772": {
+      "content": "<SPECIAL_772>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "773": {
+      "content": "<SPECIAL_773>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "774": {
+      "content": "<SPECIAL_774>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "775": {
+      "content": "<SPECIAL_775>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "776": {
+      "content": "<SPECIAL_776>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "777": {
+      "content": "<SPECIAL_777>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "778": {
+      "content": "<SPECIAL_778>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "779": {
+      "content": "<SPECIAL_779>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "780": {
+      "content": "<SPECIAL_780>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "781": {
+      "content": "<SPECIAL_781>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "782": {
+      "content": "<SPECIAL_782>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "783": {
+      "content": "<SPECIAL_783>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "784": {
+      "content": "<SPECIAL_784>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "785": {
+      "content": "<SPECIAL_785>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "786": {
+      "content": "<SPECIAL_786>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "787": {
+      "content": "<SPECIAL_787>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "788": {
+      "content": "<SPECIAL_788>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "789": {
+      "content": "<SPECIAL_789>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "790": {
+      "content": "<SPECIAL_790>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "791": {
+      "content": "<SPECIAL_791>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "792": {
+      "content": "<SPECIAL_792>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "793": {
+      "content": "<SPECIAL_793>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "794": {
+      "content": "<SPECIAL_794>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "795": {
+      "content": "<SPECIAL_795>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "796": {
+      "content": "<SPECIAL_796>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "797": {
+      "content": "<SPECIAL_797>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "798": {
+      "content": "<SPECIAL_798>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "799": {
+      "content": "<SPECIAL_799>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "800": {
+      "content": "<SPECIAL_800>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "801": {
+      "content": "<SPECIAL_801>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "802": {
+      "content": "<SPECIAL_802>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "803": {
+      "content": "<SPECIAL_803>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "804": {
+      "content": "<SPECIAL_804>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "805": {
+      "content": "<SPECIAL_805>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "806": {
+      "content": "<SPECIAL_806>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "807": {
+      "content": "<SPECIAL_807>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "808": {
+      "content": "<SPECIAL_808>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "809": {
+      "content": "<SPECIAL_809>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "810": {
+      "content": "<SPECIAL_810>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "811": {
+      "content": "<SPECIAL_811>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "812": {
+      "content": "<SPECIAL_812>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "813": {
+      "content": "<SPECIAL_813>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "814": {
+      "content": "<SPECIAL_814>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "815": {
+      "content": "<SPECIAL_815>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "816": {
+      "content": "<SPECIAL_816>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "817": {
+      "content": "<SPECIAL_817>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "818": {
+      "content": "<SPECIAL_818>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "819": {
+      "content": "<SPECIAL_819>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "820": {
+      "content": "<SPECIAL_820>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "821": {
+      "content": "<SPECIAL_821>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "822": {
+      "content": "<SPECIAL_822>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "823": {
+      "content": "<SPECIAL_823>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "824": {
+      "content": "<SPECIAL_824>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "825": {
+      "content": "<SPECIAL_825>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "826": {
+      "content": "<SPECIAL_826>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "827": {
+      "content": "<SPECIAL_827>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "828": {
+      "content": "<SPECIAL_828>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "829": {
+      "content": "<SPECIAL_829>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "830": {
+      "content": "<SPECIAL_830>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "831": {
+      "content": "<SPECIAL_831>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "832": {
+      "content": "<SPECIAL_832>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "833": {
+      "content": "<SPECIAL_833>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "834": {
+      "content": "<SPECIAL_834>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "835": {
+      "content": "<SPECIAL_835>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "836": {
+      "content": "<SPECIAL_836>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "837": {
+      "content": "<SPECIAL_837>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "838": {
+      "content": "<SPECIAL_838>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "839": {
+      "content": "<SPECIAL_839>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "840": {
+      "content": "<SPECIAL_840>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "841": {
+      "content": "<SPECIAL_841>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "842": {
+      "content": "<SPECIAL_842>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "843": {
+      "content": "<SPECIAL_843>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "844": {
+      "content": "<SPECIAL_844>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "845": {
+      "content": "<SPECIAL_845>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "846": {
+      "content": "<SPECIAL_846>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "847": {
+      "content": "<SPECIAL_847>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "848": {
+      "content": "<SPECIAL_848>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "849": {
+      "content": "<SPECIAL_849>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "850": {
+      "content": "<SPECIAL_850>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "851": {
+      "content": "<SPECIAL_851>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "852": {
+      "content": "<SPECIAL_852>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "853": {
+      "content": "<SPECIAL_853>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "854": {
+      "content": "<SPECIAL_854>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "855": {
+      "content": "<SPECIAL_855>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "856": {
+      "content": "<SPECIAL_856>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "857": {
+      "content": "<SPECIAL_857>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "858": {
+      "content": "<SPECIAL_858>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "859": {
+      "content": "<SPECIAL_859>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "860": {
+      "content": "<SPECIAL_860>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "861": {
+      "content": "<SPECIAL_861>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "862": {
+      "content": "<SPECIAL_862>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "863": {
+      "content": "<SPECIAL_863>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "864": {
+      "content": "<SPECIAL_864>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "865": {
+      "content": "<SPECIAL_865>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "866": {
+      "content": "<SPECIAL_866>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "867": {
+      "content": "<SPECIAL_867>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "868": {
+      "content": "<SPECIAL_868>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "869": {
+      "content": "<SPECIAL_869>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "870": {
+      "content": "<SPECIAL_870>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "871": {
+      "content": "<SPECIAL_871>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "872": {
+      "content": "<SPECIAL_872>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "873": {
+      "content": "<SPECIAL_873>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "874": {
+      "content": "<SPECIAL_874>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "875": {
+      "content": "<SPECIAL_875>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "876": {
+      "content": "<SPECIAL_876>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "877": {
+      "content": "<SPECIAL_877>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "878": {
+      "content": "<SPECIAL_878>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "879": {
+      "content": "<SPECIAL_879>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "880": {
+      "content": "<SPECIAL_880>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "881": {
+      "content": "<SPECIAL_881>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "882": {
+      "content": "<SPECIAL_882>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "883": {
+      "content": "<SPECIAL_883>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "884": {
+      "content": "<SPECIAL_884>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "885": {
+      "content": "<SPECIAL_885>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "886": {
+      "content": "<SPECIAL_886>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "887": {
+      "content": "<SPECIAL_887>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "888": {
+      "content": "<SPECIAL_888>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "889": {
+      "content": "<SPECIAL_889>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "890": {
+      "content": "<SPECIAL_890>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "891": {
+      "content": "<SPECIAL_891>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "892": {
+      "content": "<SPECIAL_892>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "893": {
+      "content": "<SPECIAL_893>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "894": {
+      "content": "<SPECIAL_894>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "895": {
+      "content": "<SPECIAL_895>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "896": {
+      "content": "<SPECIAL_896>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "897": {
+      "content": "<SPECIAL_897>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "898": {
+      "content": "<SPECIAL_898>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "899": {
+      "content": "<SPECIAL_899>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "900": {
+      "content": "<SPECIAL_900>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "901": {
+      "content": "<SPECIAL_901>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "902": {
+      "content": "<SPECIAL_902>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "903": {
+      "content": "<SPECIAL_903>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "904": {
+      "content": "<SPECIAL_904>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "905": {
+      "content": "<SPECIAL_905>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "906": {
+      "content": "<SPECIAL_906>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "907": {
+      "content": "<SPECIAL_907>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "908": {
+      "content": "<SPECIAL_908>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "909": {
+      "content": "<SPECIAL_909>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "910": {
+      "content": "<SPECIAL_910>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "911": {
+      "content": "<SPECIAL_911>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "912": {
+      "content": "<SPECIAL_912>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "913": {
+      "content": "<SPECIAL_913>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "914": {
+      "content": "<SPECIAL_914>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "915": {
+      "content": "<SPECIAL_915>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "916": {
+      "content": "<SPECIAL_916>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "917": {
+      "content": "<SPECIAL_917>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "918": {
+      "content": "<SPECIAL_918>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "919": {
+      "content": "<SPECIAL_919>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "920": {
+      "content": "<SPECIAL_920>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "921": {
+      "content": "<SPECIAL_921>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "922": {
+      "content": "<SPECIAL_922>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "923": {
+      "content": "<SPECIAL_923>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "924": {
+      "content": "<SPECIAL_924>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "925": {
+      "content": "<SPECIAL_925>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "926": {
+      "content": "<SPECIAL_926>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "927": {
+      "content": "<SPECIAL_927>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "928": {
+      "content": "<SPECIAL_928>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "929": {
+      "content": "<SPECIAL_929>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "930": {
+      "content": "<SPECIAL_930>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "931": {
+      "content": "<SPECIAL_931>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "932": {
+      "content": "<SPECIAL_932>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "933": {
+      "content": "<SPECIAL_933>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "934": {
+      "content": "<SPECIAL_934>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "935": {
+      "content": "<SPECIAL_935>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "936": {
+      "content": "<SPECIAL_936>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "937": {
+      "content": "<SPECIAL_937>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "938": {
+      "content": "<SPECIAL_938>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "939": {
+      "content": "<SPECIAL_939>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "940": {
+      "content": "<SPECIAL_940>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "941": {
+      "content": "<SPECIAL_941>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "942": {
+      "content": "<SPECIAL_942>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "943": {
+      "content": "<SPECIAL_943>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "944": {
+      "content": "<SPECIAL_944>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "945": {
+      "content": "<SPECIAL_945>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "946": {
+      "content": "<SPECIAL_946>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "947": {
+      "content": "<SPECIAL_947>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "948": {
+      "content": "<SPECIAL_948>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "949": {
+      "content": "<SPECIAL_949>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "950": {
+      "content": "<SPECIAL_950>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "951": {
+      "content": "<SPECIAL_951>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "952": {
+      "content": "<SPECIAL_952>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "953": {
+      "content": "<SPECIAL_953>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "954": {
+      "content": "<SPECIAL_954>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "955": {
+      "content": "<SPECIAL_955>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "956": {
+      "content": "<SPECIAL_956>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "957": {
+      "content": "<SPECIAL_957>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "958": {
+      "content": "<SPECIAL_958>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "959": {
+      "content": "<SPECIAL_959>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "960": {
+      "content": "<SPECIAL_960>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "961": {
+      "content": "<SPECIAL_961>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "962": {
+      "content": "<SPECIAL_962>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "963": {
+      "content": "<SPECIAL_963>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "964": {
+      "content": "<SPECIAL_964>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "965": {
+      "content": "<SPECIAL_965>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "966": {
+      "content": "<SPECIAL_966>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "967": {
+      "content": "<SPECIAL_967>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "968": {
+      "content": "<SPECIAL_968>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "969": {
+      "content": "<SPECIAL_969>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "970": {
+      "content": "<SPECIAL_970>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "971": {
+      "content": "<SPECIAL_971>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "972": {
+      "content": "<SPECIAL_972>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "973": {
+      "content": "<SPECIAL_973>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "974": {
+      "content": "<SPECIAL_974>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "975": {
+      "content": "<SPECIAL_975>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "976": {
+      "content": "<SPECIAL_976>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "977": {
+      "content": "<SPECIAL_977>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "978": {
+      "content": "<SPECIAL_978>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "979": {
+      "content": "<SPECIAL_979>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "980": {
+      "content": "<SPECIAL_980>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "981": {
+      "content": "<SPECIAL_981>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "982": {
+      "content": "<SPECIAL_982>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "983": {
+      "content": "<SPECIAL_983>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "984": {
+      "content": "<SPECIAL_984>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "985": {
+      "content": "<SPECIAL_985>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "986": {
+      "content": "<SPECIAL_986>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "987": {
+      "content": "<SPECIAL_987>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "988": {
+      "content": "<SPECIAL_988>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "989": {
+      "content": "<SPECIAL_989>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "990": {
+      "content": "<SPECIAL_990>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "991": {
+      "content": "<SPECIAL_991>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "992": {
+      "content": "<SPECIAL_992>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "993": {
+      "content": "<SPECIAL_993>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "994": {
+      "content": "<SPECIAL_994>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "995": {
+      "content": "<SPECIAL_995>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "996": {
+      "content": "<SPECIAL_996>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "997": {
+      "content": "<SPECIAL_997>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "998": {
+      "content": "<SPECIAL_998>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "999": {
+      "content": "<SPECIAL_999>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<s>",
+  "chat_template": "{%- if messages[0][\"role\"] == \"system\" %}\n    {%- set system_message = messages[0][\"content\"] %}\n    {%- set loop_messages = messages[1:] %}\n{%- else %}\n    {%- set loop_messages = messages %}\n{%- endif %}\n{%- if not tools is defined %}\n    {%- set tools = none %}\n{%- endif %}\n{%- set user_messages = loop_messages | selectattr(\"role\", \"equalto\", \"user\") | list %}\n\n{#- This block checks for alternating user/assistant messages, skipping tool calling messages #}\n{%- set ns = namespace() %}\n{%- set ns.index = 0 %}\n{%- for message in loop_messages %}\n    {%- if not (message.role == \"tool\" or message.role == \"tool_results\" or (message.tool_calls is defined and message.tool_calls is not none)) %}\n        {%- if (message[\"role\"] == \"user\") != (ns.index % 2 == 0) %}\n            {{- raise_exception(\"After the optional system message, conversation roles must alternate user/assistant/user/assistant/...\") }}\n        {%- endif %}\n        {%- set ns.index = ns.index + 1 %}\n    {%- endif %}\n{%- endfor %}\n\n{{- bos_token }}\n{%- for message in loop_messages %}\n    {%- if message[\"role\"] == \"user\" %}\n        {%- if tools is not none and (message == user_messages[-1]) %}\n            {{- \"[AVAILABLE_TOOLS][\" }}\n            {%- for tool in tools %}\n                {%- set tool = tool.function %}\n                {{- '{\"type\": \"function\", \"function\": {' }}\n                {%- for key, val in tool.items() if key != \"return\" %}\n                    {%- if val is string %}\n                        {{- '\"' + key + '\": \"' + val + '\"' }}\n                    {%- else %}\n                        {{- '\"' + key + '\": ' + val|tojson }}\n                    {%- endif %}\n                    {%- if not loop.last %}\n                        {{- \", \" }}\n                    {%- endif %}\n                {%- endfor %}\n                {{- \"}}\" }}\n                {%- if not loop.last %}\n                    {{- \", \" }}\n                {%- else %}\n                    {{- \"]\" }}\n                {%- endif %}\n            {%- endfor %}\n            {{- \"[/AVAILABLE_TOOLS]\" }}\n            {%- endif %}\n        {%- if loop.last and system_message is defined %}\n            {{- \"[INST]\" + system_message + \"\\n\\n\" + message[\"content\"] + \"[/INST]\" }}\n        {%- else %}\n            {{- \"[INST]\" + message[\"content\"] + \"[/INST]\" }}\n        {%- endif %}\n    {%- elif (message.tool_calls is defined and message.tool_calls is not none) %}\n        {{- \"[TOOL_CALLS][\" }}\n        {%- for tool_call in message.tool_calls %}\n            {%- set out = tool_call.function|tojson %}\n            {{- out[:-1] }}\n            {%- if not tool_call.id is defined or tool_call.id|length != 9 %}\n                {{- raise_exception(\"Tool call IDs should be alphanumeric strings with length 9!\") }}\n            {%- endif %}\n            {{- ', \"id\": \"' + tool_call.id + '\"}' }}\n            {%- if not loop.last %}\n                {{- \", \" }}\n            {%- else %}\n                {{- \"]\" + eos_token }}\n            {%- endif %}\n        {%- endfor %}\n    {%- elif message[\"role\"] == \"assistant\" %}\n        {{- message[\"content\"] + eos_token}}\n    {%- elif message[\"role\"] == \"tool_results\" or message[\"role\"] == \"tool\" %}\n        {%- if message.content is defined and message.content.content is defined %}\n            {%- set content = message.content.content %}\n        {%- else %}\n            {%- set content = message.content %}\n        {%- endif %}\n        {{- '[TOOL_RESULTS]{\"content\": ' + content|string + \", \" }}\n        {%- if not message.tool_call_id is defined or message.tool_call_id|length != 9 %}\n            {{- raise_exception(\"Tool call IDs should be alphanumeric strings with length 9!\") }}\n        {%- endif %}\n        {{- '\"call_id\": \"' + message.tool_call_id + '\"}[/TOOL_RESULTS]' }}\n    {%- else %}\n        {{- raise_exception(\"Only user and assistant roles are supported, with the exception of an initial optional system message!\") }}\n    {%- endif %}\n{%- endfor %}\n",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "</s>",
+  "extra_special_tokens": {},
+  "model_input_names": [
+    "input_ids",
+    "attention_mask"
+  ],
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "</s>",
+  "tokenizer_class": "PreTrainedTokenizer",
+  "unk_token": "<unk>"
+}
diff --git a/checkpoint-700/trainer_state.json b/checkpoint-700/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..015958cd933aab81c0207f0586c5cc404d9ff647
--- /dev/null
+++ b/checkpoint-700/trainer_state.json
@@ -0,0 +1,8784 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 0.9790209790209791,
+  "eval_steps": 500,
+  "global_step": 700,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 94.14286041259766,
+      "epoch": 0.0013986013986013986,
+      "grad_norm": 0.09098726750251449,
+      "kl": 0.0003147125244140625,
+      "learning_rate": 3.496503496503497e-07,
+      "loss": -0.0001,
+      "num_tokens": 19338.0,
+      "reward": 1.3853150606155396,
+      "reward_std": 0.4827312231063843,
+      "rewards/check_gptzero_func": 0.3214285671710968,
+      "rewards/check_perplexity_diff_func": 0.6428571343421936,
+      "rewards/check_winston_local_func": 0.42102929949760437,
+      "step": 1
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.002797202797202797,
+      "grad_norm": 0.09100260462019598,
+      "kl": 0.0003147125244140625,
+      "learning_rate": 6.993006993006994e-07,
+      "loss": -0.0001,
+      "step": 2
+    },
+    {
+      "clip_ratio": 0.0031990089919418097,
+      "completion_length": 97.73214721679688,
+      "epoch": 0.004195804195804196,
+      "grad_norm": 0.16978190242842756,
+      "kl": 0.0003662109375,
+      "learning_rate": 1.0489510489510491e-06,
+      "loss": -0.0017,
+      "num_tokens": 39487.0,
+      "reward": 0.9710169434547424,
+      "reward_std": 0.6376833319664001,
+      "rewards/check_gptzero_func": 0.125,
+      "rewards/check_perplexity_diff_func": 0.5,
+      "rewards/check_winston_local_func": 0.34601688385009766,
+      "step": 3
+    },
+    {
+      "clip_ratio": 0.0018373647471889853,
+      "epoch": 0.005594405594405594,
+      "grad_norm": 0.17663660241053314,
+      "kl": 0.000423431396484375,
+      "learning_rate": 1.3986013986013987e-06,
+      "loss": -0.0013,
+      "step": 4
+    },
+    {
+      "clip_ratio": 0.0016871786210685968,
+      "completion_length": 72.83928680419922,
+      "epoch": 0.006993006993006993,
+      "grad_norm": 0.09466520543424979,
+      "kl": 0.00034332275390625,
+      "learning_rate": 1.7482517482517485e-06,
+      "loss": 0.0008,
+      "num_tokens": 56082.0,
+      "reward": 1.0736479759216309,
+      "reward_std": 0.6862176656723022,
+      "rewards/check_gptzero_func": 0.2321428507566452,
+      "rewards/check_perplexity_diff_func": 0.3928571343421936,
+      "rewards/check_winston_local_func": 0.4486479163169861,
+      "step": 5
+    },
+    {
+      "clip_ratio": 0.0010098000057041645,
+      "epoch": 0.008391608391608392,
+      "grad_norm": 0.09539550614656178,
+      "kl": 0.0003376007080078125,
+      "learning_rate": 2.0979020979020983e-06,
+      "loss": 0.001,
+      "step": 6
+    },
+    {
+      "clip_ratio": 0.00184189947322011,
+      "completion_length": 95.92857360839844,
+      "epoch": 0.009790209790209791,
+      "grad_norm": 0.10573631695411624,
+      "kl": 0.0003757476806640625,
+      "learning_rate": 2.4475524475524477e-06,
+      "loss": 0.001,
+      "num_tokens": 75906.0,
+      "reward": 1.135968804359436,
+      "reward_std": 0.48974505066871643,
+      "rewards/check_gptzero_func": 0.2142857164144516,
+      "rewards/check_perplexity_diff_func": 0.5,
+      "rewards/check_winston_local_func": 0.42168304324150085,
+      "step": 7
+    },
+    {
+      "clip_ratio": 0.0018530809320509434,
+      "epoch": 0.011188811188811189,
+      "grad_norm": 0.10614532740192534,
+      "kl": 0.0004119873046875,
+      "learning_rate": 2.7972027972027974e-06,
+      "loss": 0.0011,
+      "step": 8
+    },
+    {
+      "clip_ratio": 0.0017129909247159958,
+      "completion_length": 106.76786041259766,
+      "epoch": 0.012587412587412588,
+      "grad_norm": 0.07916450383171113,
+      "kl": 0.000377655029296875,
+      "learning_rate": 3.1468531468531472e-06,
+      "loss": -0.0012,
+      "num_tokens": 96989.0,
+      "reward": 0.9590713977813721,
+      "reward_std": 0.5206417441368103,
+      "rewards/check_gptzero_func": 0.2321428507566452,
+      "rewards/check_perplexity_diff_func": 0.3928571343421936,
+      "rewards/check_winston_local_func": 0.3340713679790497,
+      "step": 9
+    },
+    {
+      "clip_ratio": 0.0034840807784348726,
+      "epoch": 0.013986013986013986,
+      "grad_norm": 0.08061029037170311,
+      "kl": 0.00037384033203125,
+      "learning_rate": 3.496503496503497e-06,
+      "loss": -0.0008,
+      "step": 10
+    },
+    {
+      "clip_ratio": 0.0017933619674295187,
+      "completion_length": 102.85714721679688,
+      "epoch": 0.015384615384615385,
+      "grad_norm": 0.07930375667041344,
+      "kl": 0.00030517578125,
+      "learning_rate": 3.846153846153847e-06,
+      "loss": -0.0038,
+      "num_tokens": 117271.0,
+      "reward": 1.1903148889541626,
+      "reward_std": 0.4722224473953247,
+      "rewards/check_gptzero_func": 0.2857142984867096,
+      "rewards/check_perplexity_diff_func": 0.5357142686843872,
+      "rewards/check_winston_local_func": 0.368886262178421,
+      "step": 11
+    },
+    {
+      "clip_ratio": 0.0015239757485687733,
+      "epoch": 0.016783216783216783,
+      "grad_norm": 0.07898151410343454,
+      "kl": 0.0003871917724609375,
+      "learning_rate": 4.195804195804197e-06,
+      "loss": -0.0041,
+      "step": 12
+    },
+    {
+      "clip_ratio": 0.0029670600779354572,
+      "completion_length": 104.37500762939453,
+      "epoch": 0.01818181818181818,
+      "grad_norm": 0.09247854491588844,
+      "kl": 0.0003986358642578125,
+      "learning_rate": 4.5454545454545455e-06,
+      "loss": -0.0049,
+      "num_tokens": 138174.0,
+      "reward": 0.9787324070930481,
+      "reward_std": 0.63657546043396,
+      "rewards/check_gptzero_func": 0.1964285671710968,
+      "rewards/check_perplexity_diff_func": 0.5714285969734192,
+      "rewards/check_winston_local_func": 0.21087531745433807,
+      "step": 13
+    },
+    {
+      "clip_ratio": 0.003217793768271804,
+      "epoch": 0.019580419580419582,
+      "grad_norm": 0.09293749933034297,
+      "kl": 0.0003204345703125,
+      "learning_rate": 4.895104895104895e-06,
+      "loss": -0.0045,
+      "step": 14
+    },
+    {
+      "clip_ratio": 0.002677519340068102,
+      "completion_length": 67.41072082519531,
+      "epoch": 0.02097902097902098,
+      "grad_norm": 0.14020673218962584,
+      "kl": 0.000499725341796875,
+      "learning_rate": 5.244755244755245e-06,
+      "loss": -0.008,
+      "num_tokens": 154385.0,
+      "reward": 1.0573337078094482,
+      "reward_std": 0.5388314127922058,
+      "rewards/check_gptzero_func": 0.2857142984867096,
+      "rewards/check_perplexity_diff_func": 0.3571428656578064,
+      "rewards/check_winston_local_func": 0.41447654366493225,
+      "step": 15
+    },
+    {
+      "clip_ratio": 0.0021408216562122107,
+      "epoch": 0.022377622377622378,
+      "grad_norm": 0.16110555979669544,
+      "kl": 0.000530242919921875,
+      "learning_rate": 5.594405594405595e-06,
+      "loss": -0.0081,
+      "step": 16
+    },
+    {
+      "clip_ratio": 0.002852825215086341,
+      "completion_length": 89.83928680419922,
+      "epoch": 0.023776223776223775,
+      "grad_norm": 0.1066932843002634,
+      "kl": 0.000415802001953125,
+      "learning_rate": 5.944055944055945e-06,
+      "loss": 0.0166,
+      "num_tokens": 173158.0,
+      "reward": 1.2226747274398804,
+      "reward_std": 0.45578521490097046,
+      "rewards/check_gptzero_func": 0.3392857015132904,
+      "rewards/check_perplexity_diff_func": 0.5,
+      "rewards/check_winston_local_func": 0.3833889663219452,
+      "step": 17
+    },
+    {
+      "clip_ratio": 0.002018069615587592,
+      "epoch": 0.025174825174825177,
+      "grad_norm": 0.10578254537451186,
+      "kl": 0.0003833770751953125,
+      "learning_rate": 6.2937062937062944e-06,
+      "loss": 0.0164,
+      "step": 18
+    },
+    {
+      "clip_ratio": 0.0032663617748767138,
+      "completion_length": 105.83928680419922,
+      "epoch": 0.026573426573426574,
+      "grad_norm": 0.11554707163597037,
+      "kl": 0.0003719329833984375,
+      "learning_rate": 6.643356643356643e-06,
+      "loss": -0.0005,
+      "num_tokens": 194649.0,
+      "reward": 0.8957912921905518,
+      "reward_std": 0.47702187299728394,
+      "rewards/check_gptzero_func": 0.1785714328289032,
+      "rewards/check_perplexity_diff_func": 0.3571428656578064,
+      "rewards/check_winston_local_func": 0.3600769340991974,
+      "step": 19
+    },
+    {
+      "clip_ratio": 0.0024422537535429,
+      "epoch": 0.027972027972027972,
+      "grad_norm": 0.11417666172406592,
+      "kl": 0.0003814697265625,
+      "learning_rate": 6.993006993006994e-06,
+      "loss": -0.0008,
+      "step": 20
+    },
+    {
+      "clip_ratio": 0.0013639701064676046,
+      "completion_length": 107.39286041259766,
+      "epoch": 0.02937062937062937,
+      "grad_norm": 0.09634196153318929,
+      "kl": 0.000396728515625,
+      "learning_rate": 7.342657342657343e-06,
+      "loss": -0.0011,
+      "num_tokens": 215987.0,
+      "reward": 0.9197577238082886,
+      "reward_std": 0.39100831747055054,
+      "rewards/check_gptzero_func": 0.2321428507566452,
+      "rewards/check_perplexity_diff_func": 0.3571428656578064,
+      "rewards/check_winston_local_func": 0.33047202229499817,
+      "step": 21
+    },
+    {
+      "clip_ratio": 0.003595889313146472,
+      "epoch": 0.03076923076923077,
+      "grad_norm": 0.09779857975187953,
+      "kl": 0.00052642822265625,
+      "learning_rate": 7.692307692307694e-06,
+      "loss": -0.0013,
+      "step": 22
+    },
+    {
+      "clip_ratio": 0.0024316231720149517,
+      "completion_length": 89.08928680419922,
+      "epoch": 0.032167832167832165,
+      "grad_norm": 0.13189477135999747,
+      "kl": 0.0003414154052734375,
+      "learning_rate": 8.041958041958042e-06,
+      "loss": -0.0009,
+      "num_tokens": 234678.0,
+      "reward": 0.9593304395675659,
+      "reward_std": 0.5972030758857727,
+      "rewards/check_gptzero_func": 0.1964285671710968,
+      "rewards/check_perplexity_diff_func": 0.4285714328289032,
+      "rewards/check_winston_local_func": 0.33433040976524353,
+      "step": 23
+    },
+    {
+      "clip_ratio": 0.000728283659555018,
+      "epoch": 0.033566433566433566,
+      "grad_norm": 0.12953291972672934,
+      "kl": 0.000335693359375,
+      "learning_rate": 8.391608391608393e-06,
+      "loss": -0.0011,
+      "step": 24
+    },
+    {
+      "clip_ratio": 0.0021546650677919388,
+      "completion_length": 90.98214721679688,
+      "epoch": 0.03496503496503497,
+      "grad_norm": 0.07682474826469807,
+      "kl": 0.0004558563232421875,
+      "learning_rate": 8.741258741258741e-06,
+      "loss": 0.0077,
+      "num_tokens": 253833.0,
+      "reward": 0.8255766034126282,
+      "reward_std": 0.3988816440105438,
+      "rewards/check_gptzero_func": 0.2678571343421936,
+      "rewards/check_perplexity_diff_func": 0.2857142984867096,
+      "rewards/check_winston_local_func": 0.272005170583725,
+      "step": 25
+    },
+    {
+      "clip_ratio": 0.0019250252516940236,
+      "epoch": 0.03636363636363636,
+      "grad_norm": 0.07873511079065725,
+      "kl": 0.0003490447998046875,
+      "learning_rate": 9.090909090909091e-06,
+      "loss": 0.0075,
+      "step": 26
+    },
+    {
+      "clip_ratio": 0.001886485842987895,
+      "completion_length": 74.4464340209961,
+      "epoch": 0.03776223776223776,
+      "grad_norm": 0.06765277577538054,
+      "kl": 0.000354766845703125,
+      "learning_rate": 9.44055944055944e-06,
+      "loss": -0.0003,
+      "num_tokens": 270908.0,
+      "reward": 1.3086735010147095,
+      "reward_std": 0.3881511986255646,
+      "rewards/check_gptzero_func": 0.4107142984867096,
+      "rewards/check_perplexity_diff_func": 0.4642857015132904,
+      "rewards/check_winston_local_func": 0.4336733818054199,
+      "step": 27
+    },
+    {
+      "clip_ratio": 0.0020898371003568172,
+      "epoch": 0.039160839160839164,
+      "grad_norm": 0.06722581226675682,
+      "kl": 0.000453948974609375,
+      "learning_rate": 9.79020979020979e-06,
+      "loss": -0.0006,
+      "step": 28
+    },
+    {
+      "clip_ratio": 0.002855924190953374,
+      "completion_length": 103.64286041259766,
+      "epoch": 0.04055944055944056,
+      "grad_norm": 0.16301390061045865,
+      "kl": 0.0003948211669921875,
+      "learning_rate": 1.013986013986014e-05,
+      "loss": 0.0063,
+      "num_tokens": 292060.0,
+      "reward": 1.22488272190094,
+      "reward_std": 0.8093323111534119,
+      "rewards/check_gptzero_func": 0.2678571343421936,
+      "rewards/check_perplexity_diff_func": 0.6071428656578064,
+      "rewards/check_winston_local_func": 0.34988275170326233,
+      "step": 29
+    },
+    {
+      "clip_ratio": 0.0016208746237680316,
+      "epoch": 0.04195804195804196,
+      "grad_norm": 0.17335125351231856,
+      "kl": 0.000507354736328125,
+      "learning_rate": 1.048951048951049e-05,
+      "loss": 0.0059,
+      "step": 30
+    },
+    {
+      "clip_ratio": 0.0011572305811569095,
+      "completion_length": 82.91072082519531,
+      "epoch": 0.043356643356643354,
+      "grad_norm": 0.11830097508590764,
+      "kl": 0.0006256103515625,
+      "learning_rate": 1.0839160839160838e-05,
+      "loss": 0.002,
+      "num_tokens": 310163.0,
+      "reward": 0.9069231748580933,
+      "reward_std": 0.4621748924255371,
+      "rewards/check_gptzero_func": 0.25,
+      "rewards/check_perplexity_diff_func": 0.3571428656578064,
+      "rewards/check_winston_local_func": 0.29978030920028687,
+      "step": 31
+    },
+    {
+      "clip_ratio": 0.00402362085878849,
+      "epoch": 0.044755244755244755,
+      "grad_norm": 0.11861334422671065,
+      "kl": 0.000762939453125,
+      "learning_rate": 1.118881118881119e-05,
+      "loss": 0.0019,
+      "step": 32
+    },
+    {
+      "clip_ratio": 0.0017474278574809432,
+      "completion_length": 90.9464340209961,
+      "epoch": 0.046153846153846156,
+      "grad_norm": 0.1025496804751534,
+      "kl": 0.000701904296875,
+      "learning_rate": 1.153846153846154e-05,
+      "loss": -0.0014,
+      "num_tokens": 328922.0,
+      "reward": 1.1198338270187378,
+      "reward_std": 0.5045111775398254,
+      "rewards/check_gptzero_func": 0.0892857164144516,
+      "rewards/check_perplexity_diff_func": 0.7857142686843872,
+      "rewards/check_winston_local_func": 0.24483375251293182,
+      "step": 33
+    },
+    {
+      "clip_ratio": 0.0012307984288781881,
+      "epoch": 0.04755244755244755,
+      "grad_norm": 0.10274008519956918,
+      "kl": 0.00087738037109375,
+      "learning_rate": 1.188811188811189e-05,
+      "loss": -0.0015,
+      "step": 34
+    },
+    {
+      "clip_ratio": 0.0029886537231504917,
+      "completion_length": 93.00000762939453,
+      "epoch": 0.04895104895104895,
+      "grad_norm": 0.1311802256012859,
+      "kl": 0.001190185546875,
+      "learning_rate": 1.2237762237762239e-05,
+      "loss": -0.0141,
+      "num_tokens": 348852.0,
+      "reward": 0.9456299543380737,
+      "reward_std": 0.6374980807304382,
+      "rewards/check_gptzero_func": 0.2321428507566452,
+      "rewards/check_perplexity_diff_func": 0.4285714328289032,
+      "rewards/check_winston_local_func": 0.28491565585136414,
+      "step": 35
+    },
+    {
+      "clip_ratio": 0.0028984802775084972,
+      "epoch": 0.05034965034965035,
+      "grad_norm": 0.13267305607921692,
+      "kl": 0.00145721435546875,
+      "learning_rate": 1.2587412587412589e-05,
+      "loss": -0.0147,
+      "step": 36
+    },
+    {
+      "clip_ratio": 0.0018746532732620835,
+      "completion_length": 89.85714721679688,
+      "epoch": 0.05174825174825175,
+      "grad_norm": 0.1497182815907709,
+      "kl": 0.002044677734375,
+      "learning_rate": 1.2937062937062939e-05,
+      "loss": -0.0253,
+      "num_tokens": 367482.0,
+      "reward": 1.2605940103530884,
+      "reward_std": 0.66156405210495,
+      "rewards/check_gptzero_func": 0.1785714328289032,
+      "rewards/check_perplexity_diff_func": 0.75,
+      "rewards/check_winston_local_func": 0.3320225179195404,
+      "step": 37
+    },
+    {
+      "clip_ratio": 0.0020914783235639334,
+      "epoch": 0.05314685314685315,
+      "grad_norm": 0.15027543310666272,
+      "kl": 0.0026092529296875,
+      "learning_rate": 1.3286713286713287e-05,
+      "loss": -0.0257,
+      "step": 38
+    },
+    {
+      "clip_ratio": 0.0017028081929311156,
+      "completion_length": 101.58928680419922,
+      "epoch": 0.05454545454545454,
+      "grad_norm": 0.1306748742984494,
+      "kl": 0.0027008056640625,
+      "learning_rate": 1.3636363636363637e-05,
+      "loss": -0.0091,
+      "num_tokens": 388365.0,
+      "reward": 1.1187876462936401,
+      "reward_std": 0.4044211804866791,
+      "rewards/check_gptzero_func": 0.2321428507566452,
+      "rewards/check_perplexity_diff_func": 0.5357142686843872,
+      "rewards/check_winston_local_func": 0.3509305417537689,
+      "step": 39
+    },
+    {
+      "clip_ratio": 0.003386714030057192,
+      "epoch": 0.055944055944055944,
+      "grad_norm": 0.12880289464580882,
+      "kl": 0.003265380859375,
+      "learning_rate": 1.3986013986013988e-05,
+      "loss": -0.0086,
+      "step": 40
+    },
+    {
+      "clip_ratio": 0.0024523374158889055,
+      "completion_length": 100.41072082519531,
+      "epoch": 0.057342657342657345,
+      "grad_norm": 0.12385816200639352,
+      "kl": 0.004547119140625,
+      "learning_rate": 1.4335664335664336e-05,
+      "loss": -0.0161,
+      "num_tokens": 409074.0,
+      "reward": 0.804709792137146,
+      "reward_std": 0.5129754543304443,
+      "rewards/check_gptzero_func": 0.1428571492433548,
+      "rewards/check_perplexity_diff_func": 0.3928571343421936,
+      "rewards/check_winston_local_func": 0.2689954936504364,
+      "step": 41
+    },
+    {
+      "clip_ratio": 0.004267544951289892,
+      "epoch": 0.05874125874125874,
+      "grad_norm": 0.12301160152684054,
+      "kl": 0.005218505859375,
+      "learning_rate": 1.4685314685314686e-05,
+      "loss": -0.0161,
+      "step": 42
+    },
+    {
+      "clip_ratio": 0.0019341636216267943,
+      "completion_length": 90.46428680419922,
+      "epoch": 0.06013986013986014,
+      "grad_norm": 0.1390908713431379,
+      "kl": 0.0048828125,
+      "learning_rate": 1.5034965034965034e-05,
+      "loss": 0.0068,
+      "num_tokens": 428262.0,
+      "reward": 1.087609887123108,
+      "reward_std": 0.6448312997817993,
+      "rewards/check_gptzero_func": 0.2321428507566452,
+      "rewards/check_perplexity_diff_func": 0.3928571343421936,
+      "rewards/check_winston_local_func": 0.46260982751846313,
+      "step": 43
+    },
+    {
+      "clip_ratio": 0.0016832423862069845,
+      "epoch": 0.06153846153846154,
+      "grad_norm": 0.13873080950458352,
+      "kl": 0.00567626953125,
+      "learning_rate": 1.5384615384615387e-05,
+      "loss": 0.0065,
+      "step": 44
+    },
+    {
+      "clip_ratio": 0.0019954824820160866,
+      "completion_length": 96.92857360839844,
+      "epoch": 0.06293706293706294,
+      "grad_norm": 0.11630124487143968,
+      "kl": 0.00897216796875,
+      "learning_rate": 1.5734265734265734e-05,
+      "loss": 0.008,
+      "num_tokens": 448480.0,
+      "reward": 1.0604994297027588,
+      "reward_std": 0.5378546118736267,
+      "rewards/check_gptzero_func": 0.2321428507566452,
+      "rewards/check_perplexity_diff_func": 0.4642857015132904,
+      "rewards/check_winston_local_func": 0.36407074332237244,
+      "step": 45
+    },
+    {
+      "clip_ratio": 0.002663462422788143,
+      "epoch": 0.06433566433566433,
+      "grad_norm": 0.1149796219701467,
+      "kl": 0.01007080078125,
+      "learning_rate": 1.6083916083916083e-05,
+      "loss": 0.0081,
+      "step": 46
+    },
+    {
+      "clip_ratio": 0.0018039485439658165,
+      "completion_length": 79.51786041259766,
+      "epoch": 0.06573426573426573,
+      "grad_norm": 0.12721969778213826,
+      "kl": 0.01123046875,
+      "learning_rate": 1.6433566433566433e-05,
+      "loss": 0.0112,
+      "num_tokens": 465671.0,
+      "reward": 0.9921315312385559,
+      "reward_std": 0.38579052686691284,
+      "rewards/check_gptzero_func": 0.1607142835855484,
+      "rewards/check_perplexity_diff_func": 0.3928571343421936,
+      "rewards/check_winston_local_func": 0.43856000900268555,
+      "step": 47
+    },
+    {
+      "clip_ratio": 0.0027954555116593838,
+      "epoch": 0.06713286713286713,
+      "grad_norm": 0.12930272448754576,
+      "kl": 0.01300048828125,
+      "learning_rate": 1.6783216783216786e-05,
+      "loss": 0.0107,
+      "step": 48
+    },
+    {
+      "clip_ratio": 0.0028269642498344183,
+      "completion_length": 116.5714340209961,
+      "epoch": 0.06853146853146853,
+      "grad_norm": 0.12430052891449103,
+      "kl": 0.01226806640625,
+      "learning_rate": 1.7132867132867133e-05,
+      "loss": 0.025,
+      "num_tokens": 488367.0,
+      "reward": 0.8830849528312683,
+      "reward_std": 0.5276607871055603,
+      "rewards/check_gptzero_func": 0.25,
+      "rewards/check_perplexity_diff_func": 0.3214285671710968,
+      "rewards/check_winston_local_func": 0.31165632605552673,
+      "step": 49
+    },
+    {
+      "clip_ratio": 0.0023628019262105227,
+      "epoch": 0.06993006993006994,
+      "grad_norm": 0.12293590613717971,
+      "kl": 0.0142822265625,
+      "learning_rate": 1.7482517482517483e-05,
+      "loss": 0.0249,
+      "step": 50
+    },
+    {
+      "clip_ratio": 0.0033425339497625828,
+      "completion_length": 83.39286041259766,
+      "epoch": 0.07132867132867132,
+      "grad_norm": 0.11106847343390151,
+      "kl": 0.01177978515625,
+      "learning_rate": 1.7832167832167836e-05,
+      "loss": -0.0041,
+      "num_tokens": 506551.0,
+      "reward": 1.029296875,
+      "reward_std": 0.4901208281517029,
+      "rewards/check_gptzero_func": 0.2857142984867096,
+      "rewards/check_perplexity_diff_func": 0.3214285671710968,
+      "rewards/check_winston_local_func": 0.42215394973754883,
+      "step": 51
+    },
+    {
+      "clip_ratio": 0.0022109400015324354,
+      "epoch": 0.07272727272727272,
+      "grad_norm": 0.11017795132415774,
+      "kl": 0.013671875,
+      "learning_rate": 1.8181818181818182e-05,
+      "loss": -0.0043,
+      "step": 52
+    },
+    {
+      "clip_ratio": 0.0019256824161857367,
+      "completion_length": 86.5714340209961,
+      "epoch": 0.07412587412587412,
+      "grad_norm": 0.11010712395827207,
+      "kl": 0.0208740234375,
+      "learning_rate": 1.8531468531468532e-05,
+      "loss": -0.0032,
+      "num_tokens": 524863.0,
+      "reward": 0.962996244430542,
+      "reward_std": 0.6607940196990967,
+      "rewards/check_gptzero_func": 0.1071428582072258,
+      "rewards/check_perplexity_diff_func": 0.5,
+      "rewards/check_winston_local_func": 0.355853408575058,
+      "step": 53
+    },
+    {
+      "clip_ratio": 0.0025832760147750378,
+      "epoch": 0.07552447552447553,
+      "grad_norm": 0.10695378216267973,
+      "kl": 0.0250244140625,
+      "learning_rate": 1.888111888111888e-05,
+      "loss": -0.0034,
+      "step": 54
+    },
+    {
+      "clip_ratio": 0.0017639577854424715,
+      "completion_length": 96.08928680419922,
+      "epoch": 0.07692307692307693,
+      "grad_norm": 0.17472941592839772,
+      "kl": 0.0322265625,
+      "learning_rate": 1.923076923076923e-05,
+      "loss": -0.0141,
+      "num_tokens": 544736.0,
+      "reward": 1.3478963375091553,
+      "reward_std": 0.8002303242683411,
+      "rewards/check_gptzero_func": 0.2857142984867096,
+      "rewards/check_perplexity_diff_func": 0.7142857313156128,
+      "rewards/check_winston_local_func": 0.34789615869522095,
+      "step": 55
+    },
+    {
+      "clip_ratio": 0.0034089265391230583,
+      "epoch": 0.07832167832167833,
+      "grad_norm": 0.14666884899599142,
+      "kl": 0.041748046875,
+      "learning_rate": 1.958041958041958e-05,
+      "loss": -0.0155,
+      "step": 56
+    },
+    {
+      "clip_ratio": 0.0026615143287926912,
+      "completion_length": 96.91072082519531,
+      "epoch": 0.07972027972027972,
+      "grad_norm": 0.1400337476406121,
+      "kl": 0.035400390625,
+      "learning_rate": 1.993006993006993e-05,
+      "loss": -0.027,
+      "num_tokens": 564803.0,
+      "reward": 1.0594887733459473,
+      "reward_std": 0.6912091970443726,
+      "rewards/check_gptzero_func": 0.2857142984867096,
+      "rewards/check_perplexity_diff_func": 0.4285714328289032,
+      "rewards/check_winston_local_func": 0.34520307183265686,
+      "step": 57
+    },
+    {
+      "clip_ratio": 0.003153084311634302,
+      "epoch": 0.08111888111888112,
+      "grad_norm": 0.138816359725178,
+      "kl": 0.042236328125,
+      "learning_rate": 2.027972027972028e-05,
+      "loss": -0.0274,
+      "step": 58
+    },
+    {
+      "clip_ratio": 0.0012626759707927704,
+      "completion_length": 76.35714721679688,
+      "epoch": 0.08251748251748252,
+      "grad_norm": 0.15940476459351663,
+      "kl": 0.0556640625,
+      "learning_rate": 2.062937062937063e-05,
+      "loss": -0.0057,
+      "num_tokens": 582265.0,
+      "reward": 0.992591381072998,
+      "reward_std": 0.5938137769699097,
+      "rewards/check_gptzero_func": 0.125,
+      "rewards/check_perplexity_diff_func": 0.4642857015132904,
+      "rewards/check_winston_local_func": 0.4033055901527405,
+      "step": 59
+    },
+    {
+      "clip_ratio": 0.004643784370273352,
+      "epoch": 0.08391608391608392,
+      "grad_norm": 0.15305834024157844,
+      "kl": 0.07421875,
+      "learning_rate": 2.097902097902098e-05,
+      "loss": -0.0064,
+      "step": 60
+    },
+    {
+      "clip_ratio": 0.0032869603019207716,
+      "completion_length": 103.33928680419922,
+      "epoch": 0.08531468531468532,
+      "grad_norm": 0.12260760902895611,
+      "kl": 0.0810546875,
+      "learning_rate": 2.132867132867133e-05,
+      "loss": 0.0066,
+      "num_tokens": 603454.0,
+      "reward": 1.191124439239502,
+      "reward_std": 0.5557684898376465,
+      "rewards/check_gptzero_func": 0.2321428507566452,
+      "rewards/check_perplexity_diff_func": 0.5357142686843872,
+      "rewards/check_winston_local_func": 0.4232672154903412,
+      "step": 61
+    },
+    {
+      "clip_ratio": 0.008058370091021061,
+      "epoch": 0.08671328671328671,
+      "grad_norm": 0.10224499878092017,
+      "kl": 0.1162109375,
+      "learning_rate": 2.1678321678321677e-05,
+      "loss": 0.0058,
+      "step": 62
+    },
+    {
+      "clip_ratio": 0.002202474046498537,
+      "completion_length": 91.4464340209961,
+      "epoch": 0.08811188811188811,
+      "grad_norm": 0.09776882919444682,
+      "kl": 0.2138671875,
+      "learning_rate": 2.202797202797203e-05,
+      "loss": 0.0021,
+      "num_tokens": 622705.0,
+      "reward": 1.1320958137512207,
+      "reward_std": 0.4060821831226349,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 0.3214285671710968,
+      "rewards/check_winston_local_func": 0.41781002283096313,
+      "step": 63
+    },
+    {
+      "clip_ratio": 0.0025903189089149237,
+      "epoch": 0.08951048951048951,
+      "grad_norm": 0.0963149975780195,
+      "kl": 0.2109375,
+      "learning_rate": 2.237762237762238e-05,
+      "loss": 0.0014,
+      "step": 64
+    },
+    {
+      "clip_ratio": 0.0016666523879393935,
+      "completion_length": 100.00000762939453,
+      "epoch": 0.09090909090909091,
+      "grad_norm": 0.12490001498293682,
+      "kl": 0.1220703125,
+      "learning_rate": 2.272727272727273e-05,
+      "loss": 0.0081,
+      "num_tokens": 643191.0,
+      "reward": 1.274580478668213,
+      "reward_std": 0.5277792811393738,
+      "rewards/check_gptzero_func": 0.2321428507566452,
+      "rewards/check_perplexity_diff_func": 0.5357142686843872,
+      "rewards/check_winston_local_func": 0.5067232847213745,
+      "step": 65
+    },
+    {
+      "clip_ratio": 0.0032784033101052046,
+      "epoch": 0.09230769230769231,
+      "grad_norm": 0.12114304348028067,
+      "kl": 0.134765625,
+      "learning_rate": 2.307692307692308e-05,
+      "loss": 0.0074,
+      "step": 66
+    },
+    {
+      "clip_ratio": 0.0017520035617053509,
+      "completion_length": 95.16072082519531,
+      "epoch": 0.0937062937062937,
+      "grad_norm": 0.14292853522837679,
+      "kl": 0.216796875,
+      "learning_rate": 2.342657342657343e-05,
+      "loss": -0.0018,
+      "num_tokens": 662736.0,
+      "reward": 1.1007850170135498,
+      "reward_std": 0.45985397696495056,
+      "rewards/check_gptzero_func": 0.25,
+      "rewards/check_perplexity_diff_func": 0.4642857015132904,
+      "rewards/check_winston_local_func": 0.38649922609329224,
+      "step": 67
+    },
+    {
+      "clip_ratio": 0.00579434959217906,
+      "epoch": 0.0951048951048951,
+      "grad_norm": 0.14340586759573087,
+      "kl": 0.2431640625,
+      "learning_rate": 2.377622377622378e-05,
+      "loss": -0.0029,
+      "step": 68
+    },
+    {
+      "clip_ratio": 0.0022546499967575073,
+      "completion_length": 86.17857360839844,
+      "epoch": 0.0965034965034965,
+      "grad_norm": 0.17287042335415018,
+      "kl": 0.26171875,
+      "learning_rate": 2.4125874125874125e-05,
+      "loss": 0.0059,
+      "num_tokens": 681030.0,
+      "reward": 1.4391711950302124,
+      "reward_std": 0.6516181230545044,
+      "rewards/check_gptzero_func": 0.5178571343421936,
+      "rewards/check_perplexity_diff_func": 0.4642857015132904,
+      "rewards/check_winston_local_func": 0.45702821016311646,
+      "step": 69
+    },
+    {
+      "clip_ratio": 0.004240955226123333,
+      "epoch": 0.0979020979020979,
+      "grad_norm": 0.16371494613893736,
+      "kl": 0.28125,
+      "learning_rate": 2.4475524475524478e-05,
+      "loss": 0.0048,
+      "step": 70
+    },
+    {
+      "clip_ratio": 0.003548440057784319,
+      "completion_length": 102.42857360839844,
+      "epoch": 0.0993006993006993,
+      "grad_norm": 0.1615418523026911,
+      "kl": 0.1708984375,
+      "learning_rate": 2.4825174825174828e-05,
+      "loss": 0.0015,
+      "num_tokens": 701874.0,
+      "reward": 1.4929367303848267,
+      "reward_std": 0.6083499193191528,
+      "rewards/check_gptzero_func": 0.3214285671710968,
+      "rewards/check_perplexity_diff_func": 0.7142857313156128,
+      "rewards/check_winston_local_func": 0.4572224020957947,
+      "step": 71
+    },
+    {
+      "clip_ratio": 0.004331245087087154,
+      "epoch": 0.1006993006993007,
+      "grad_norm": 0.12876017762018707,
+      "kl": 0.216796875,
+      "learning_rate": 2.5174825174825178e-05,
+      "loss": 0.0,
+      "step": 72
+    },
+    {
+      "clip_ratio": 0.003951852675527334,
+      "completion_length": 73.46428680419922,
+      "epoch": 0.1020979020979021,
+      "grad_norm": 0.20791748542634808,
+      "kl": 0.365234375,
+      "learning_rate": 2.5524475524475528e-05,
+      "loss": -0.0037,
+      "num_tokens": 718488.0,
+      "reward": 1.7953797578811646,
+      "reward_std": 0.5393027067184448,
+      "rewards/check_gptzero_func": 0.4464285671710968,
+      "rewards/check_perplexity_diff_func": 0.8571428656578064,
+      "rewards/check_winston_local_func": 0.49180838465690613,
+      "step": 73
+    },
+    {
+      "clip_ratio": 0.008312534540891647,
+      "epoch": 0.1034965034965035,
+      "grad_norm": 0.19232465302245624,
+      "kl": 0.5390625,
+      "learning_rate": 2.5874125874125877e-05,
+      "loss": -0.0053,
+      "step": 74
+    },
+    {
+      "clip_ratio": 0.0021736263297498226,
+      "completion_length": 97.73214721679688,
+      "epoch": 0.1048951048951049,
+      "grad_norm": 0.18504738647098173,
+      "kl": 0.2421875,
+      "learning_rate": 2.6223776223776224e-05,
+      "loss": -0.0023,
+      "num_tokens": 738619.0,
+      "reward": 1.284185767173767,
+      "reward_std": 0.6959513425827026,
+      "rewards/check_gptzero_func": 0.4464285671710968,
+      "rewards/check_perplexity_diff_func": 0.3928571343421936,
+      "rewards/check_winston_local_func": 0.4449000358581543,
+      "step": 75
+    },
+    {
+      "clip_ratio": 0.0033756059128791094,
+      "epoch": 0.1062937062937063,
+      "grad_norm": 0.17585639043073228,
+      "kl": 0.255859375,
+      "learning_rate": 2.6573426573426574e-05,
+      "loss": -0.0043,
+      "step": 76
+    },
+    {
+      "clip_ratio": 0.003740633837878704,
+      "completion_length": 95.67857360839844,
+      "epoch": 0.1076923076923077,
+      "grad_norm": 0.21573775826044048,
+      "kl": 0.2451171875,
+      "learning_rate": 2.6923076923076923e-05,
+      "loss": 0.0105,
+      "num_tokens": 758111.0,
+      "reward": 1.4672702550888062,
+      "reward_std": 0.85318922996521,
+      "rewards/check_gptzero_func": 0.25,
+      "rewards/check_perplexity_diff_func": 0.7857142686843872,
+      "rewards/check_winston_local_func": 0.43155592679977417,
+      "step": 77
+    },
+    {
+      "clip_ratio": 0.004517109598964453,
+      "epoch": 0.10909090909090909,
+      "grad_norm": 0.2071235680791843,
+      "kl": 0.302734375,
+      "learning_rate": 2.7272727272727273e-05,
+      "loss": 0.009,
+      "step": 78
+    },
+    {
+      "clip_ratio": 0.004641580395400524,
+      "completion_length": 117.75000762939453,
+      "epoch": 0.11048951048951049,
+      "grad_norm": 0.19278645376555428,
+      "kl": 0.259765625,
+      "learning_rate": 2.762237762237762e-05,
+      "loss": -0.0004,
+      "num_tokens": 780403.0,
+      "reward": 1.4974778890609741,
+      "reward_std": 0.59557044506073,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 0.6785714030265808,
+      "rewards/check_winston_local_func": 0.39033493399620056,
+      "step": 79
+    },
+    {
+      "clip_ratio": 0.006305322516709566,
+      "epoch": 0.11188811188811189,
+      "grad_norm": 0.13786850726930885,
+      "kl": 0.3125,
+      "learning_rate": 2.7972027972027976e-05,
+      "loss": -0.0019,
+      "step": 80
+    },
+    {
+      "clip_ratio": 0.0017778041074052453,
+      "completion_length": 101.03572082519531,
+      "epoch": 0.11328671328671329,
+      "grad_norm": 0.18099911272050398,
+      "kl": 1.7578125,
+      "learning_rate": 2.8321678321678326e-05,
+      "loss": 0.0053,
+      "num_tokens": 800231.0,
+      "reward": 1.420699954032898,
+      "reward_std": 0.5104399919509888,
+      "rewards/check_gptzero_func": 0.2678571343421936,
+      "rewards/check_perplexity_diff_func": 0.6428571343421936,
+      "rewards/check_winston_local_func": 0.5099858045578003,
+      "step": 81
+    },
+    {
+      "clip_ratio": 0.0030425102449953556,
+      "epoch": 0.11468531468531469,
+      "grad_norm": 0.10725025403922342,
+      "kl": 0.9921875,
+      "learning_rate": 2.8671328671328672e-05,
+      "loss": 0.0043,
+      "step": 82
+    },
+    {
+      "clip_ratio": 0.0030319676734507084,
+      "completion_length": 93.3214340209961,
+      "epoch": 0.11608391608391608,
+      "grad_norm": 0.22858957476463732,
+      "kl": 0.35546875,
+      "learning_rate": 2.9020979020979022e-05,
+      "loss": 0.0021,
+      "num_tokens": 819773.0,
+      "reward": 1.4201573133468628,
+      "reward_std": 0.761705756187439,
+      "rewards/check_gptzero_func": 0.3392857015132904,
+      "rewards/check_perplexity_diff_func": 0.5714285969734192,
+      "rewards/check_winston_local_func": 0.5094431042671204,
+      "step": 83
+    },
+    {
+      "clip_ratio": 0.0027519434224814177,
+      "epoch": 0.11748251748251748,
+      "grad_norm": 0.180287994009811,
+      "kl": 0.392578125,
+      "learning_rate": 2.9370629370629372e-05,
+      "loss": -0.0002,
+      "step": 84
+    },
+    {
+      "clip_ratio": 0.0034523813519626856,
+      "completion_length": 102.66072082519531,
+      "epoch": 0.11888111888111888,
+      "grad_norm": 0.25571068617285075,
+      "kl": 0.494140625,
+      "learning_rate": 2.972027972027972e-05,
+      "loss": -0.0103,
+      "num_tokens": 840280.0,
+      "reward": 1.7226576805114746,
+      "reward_std": 0.7545516490936279,
+      "rewards/check_gptzero_func": 0.5714285969734192,
+      "rewards/check_perplexity_diff_func": 0.6785714030265808,
+      "rewards/check_winston_local_func": 0.4726576805114746,
+      "step": 85
+    },
+    {
+      "clip_ratio": 0.00493992306292057,
+      "epoch": 0.12027972027972028,
+      "grad_norm": 0.2172932526081326,
+      "kl": 0.490234375,
+      "learning_rate": 3.0069930069930068e-05,
+      "loss": -0.013,
+      "step": 86
+    },
+    {
+      "clip_ratio": 0.003154533449560404,
+      "completion_length": 98.96428680419922,
+      "epoch": 0.12167832167832168,
+      "grad_norm": 0.3651883135439141,
+      "kl": 0.5859375,
+      "learning_rate": 3.0419580419580425e-05,
+      "loss": -0.0129,
+      "num_tokens": 860422.0,
+      "reward": 1.6449967622756958,
+      "reward_std": 0.7163441777229309,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 0.7142857313156128,
+      "rewards/check_winston_local_func": 0.5378537178039551,
+      "step": 87
+    },
+    {
+      "clip_ratio": 0.007247431669384241,
+      "epoch": 0.12307692307692308,
+      "grad_norm": 0.2167770735292921,
+      "kl": 0.61328125,
+      "learning_rate": 3.0769230769230774e-05,
+      "loss": -0.0168,
+      "step": 88
+    },
+    {
+      "clip_ratio": 0.0027036736719310284,
+      "completion_length": 108.03572082519531,
+      "epoch": 0.12447552447552447,
+      "grad_norm": 0.12604084880729077,
+      "kl": 0.291015625,
+      "learning_rate": 3.111888111888112e-05,
+      "loss": -0.0039,
+      "num_tokens": 881364.0,
+      "reward": 1.3799673318862915,
+      "reward_std": 0.6881024837493896,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 0.5714285969734192,
+      "rewards/check_winston_local_func": 0.4156815707683563,
+      "step": 89
+    },
+    {
+      "clip_ratio": 0.005238416139036417,
+      "epoch": 0.1258741258741259,
+      "grad_norm": 0.12103303780018655,
+      "kl": 0.30859375,
+      "learning_rate": 3.146853146853147e-05,
+      "loss": -0.005,
+      "step": 90
+    },
+    {
+      "clip_ratio": 0.003768512513488531,
+      "completion_length": 122.80357360839844,
+      "epoch": 0.12727272727272726,
+      "grad_norm": 0.18423815092289067,
+      "kl": 0.67578125,
+      "learning_rate": 3.181818181818182e-05,
+      "loss": -0.0171,
+      "num_tokens": 904487.0,
+      "reward": 1.6101170778274536,
+      "reward_std": 0.6241902709007263,
+      "rewards/check_gptzero_func": 0.3392857015132904,
+      "rewards/check_perplexity_diff_func": 0.7857142686843872,
+      "rewards/check_winston_local_func": 0.48511695861816406,
+      "step": 91
+    },
+    {
+      "clip_ratio": 0.005407070741057396,
+      "epoch": 0.12867132867132866,
+      "grad_norm": 0.15457101650134836,
+      "kl": 0.66796875,
+      "learning_rate": 3.216783216783217e-05,
+      "loss": -0.0182,
+      "step": 92
+    },
+    {
+      "clip_ratio": 0.002042532665655017,
+      "completion_length": 108.17857360839844,
+      "epoch": 0.13006993006993006,
+      "grad_norm": 0.19588538280583437,
+      "kl": 1.75,
+      "learning_rate": 3.251748251748252e-05,
+      "loss": -0.0023,
+      "num_tokens": 925633.0,
+      "reward": 1.3932582139968872,
+      "reward_std": 0.5087055563926697,
+      "rewards/check_gptzero_func": 0.4107142984867096,
+      "rewards/check_perplexity_diff_func": 0.4642857015132904,
+      "rewards/check_winston_local_func": 0.5182580351829529,
+      "step": 93
+    },
+    {
+      "clip_ratio": 0.00876330491155386,
+      "epoch": 0.13146853146853146,
+      "grad_norm": 0.1506256243597478,
+      "kl": 0.80078125,
+      "learning_rate": 3.2867132867132866e-05,
+      "loss": -0.0035,
+      "step": 94
+    },
+    {
+      "clip_ratio": 0.003118924330919981,
+      "completion_length": 77.4464340209961,
+      "epoch": 0.13286713286713286,
+      "grad_norm": 0.34784170699852607,
+      "kl": 0.72265625,
+      "learning_rate": 3.321678321678322e-05,
+      "loss": -0.0134,
+      "num_tokens": 942892.0,
+      "reward": 1.832355260848999,
+      "reward_std": 1.064102292060852,
+      "rewards/check_gptzero_func": 0.4821428656578064,
+      "rewards/check_perplexity_diff_func": 0.8214285969734192,
+      "rewards/check_winston_local_func": 0.5287837982177734,
+      "step": 95
+    },
+    {
+      "clip_ratio": 0.00509096821770072,
+      "epoch": 0.13426573426573427,
+      "grad_norm": 0.25968424971577175,
+      "kl": 0.96875,
+      "learning_rate": 3.356643356643357e-05,
+      "loss": -0.0176,
+      "step": 96
+    },
+    {
+      "clip_ratio": 0.0037272910121828318,
+      "completion_length": 95.26786041259766,
+      "epoch": 0.13566433566433567,
+      "grad_norm": 0.16914051199341906,
+      "kl": 0.349609375,
+      "learning_rate": 3.391608391608392e-05,
+      "loss": 0.0037,
+      "num_tokens": 962317.0,
+      "reward": 1.805726170539856,
+      "reward_std": 0.831261396408081,
+      "rewards/check_gptzero_func": 0.4107142984867096,
+      "rewards/check_perplexity_diff_func": 0.9285714030265808,
+      "rewards/check_winston_local_func": 0.46644020080566406,
+      "step": 97
+    },
+    {
+      "clip_ratio": 0.0034601751249283552,
+      "epoch": 0.13706293706293707,
+      "grad_norm": 0.16538870434198547,
+      "kl": 0.388671875,
+      "learning_rate": 3.4265734265734265e-05,
+      "loss": 0.0019,
+      "step": 98
+    },
+    {
+      "clip_ratio": 0.004487877711653709,
+      "completion_length": 111.8214340209961,
+      "epoch": 0.13846153846153847,
+      "grad_norm": 0.12415078751808917,
+      "kl": 0.384765625,
+      "learning_rate": 3.461538461538462e-05,
+      "loss": -0.001,
+      "num_tokens": 983513.0,
+      "reward": 1.2062333822250366,
+      "reward_std": 0.591974139213562,
+      "rewards/check_gptzero_func": 0.3214285671710968,
+      "rewards/check_perplexity_diff_func": 0.3571428656578064,
+      "rewards/check_winston_local_func": 0.5276618599891663,
+      "step": 99
+    },
+    {
+      "clip_ratio": 0.00284260674379766,
+      "epoch": 0.13986013986013987,
+      "grad_norm": 0.1197068572237121,
+      "kl": 0.376953125,
+      "learning_rate": 3.4965034965034965e-05,
+      "loss": -0.0017,
+      "step": 100
+    },
+    {
+      "clip_ratio": 0.0023206709884107113,
+      "completion_length": 101.08928680419922,
+      "epoch": 0.14125874125874127,
+      "grad_norm": 0.16400691003183243,
+      "kl": 0.30078125,
+      "learning_rate": 3.531468531468531e-05,
+      "loss": 0.0033,
+      "num_tokens": 1003878.0,
+      "reward": 1.631854772567749,
+      "reward_std": 0.5223387479782104,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 0.7142857313156128,
+      "rewards/check_winston_local_func": 0.5247119665145874,
+      "step": 101
+    },
+    {
+      "clip_ratio": 0.0016618981026113033,
+      "epoch": 0.14265734265734265,
+      "grad_norm": 0.1573929247787687,
+      "kl": 0.326171875,
+      "learning_rate": 3.566433566433567e-05,
+      "loss": 0.0012,
+      "step": 102
+    },
+    {
+      "clip_ratio": 0.0022201864048838615,
+      "completion_length": 124.46429443359375,
+      "epoch": 0.14405594405594405,
+      "grad_norm": 0.1353861256020412,
+      "kl": 0.3828125,
+      "learning_rate": 3.601398601398602e-05,
+      "loss": 0.005,
+      "num_tokens": 1027066.0,
+      "reward": 1.6920486688613892,
+      "reward_std": 0.48560601472854614,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 0.75,
+      "rewards/check_winston_local_func": 0.5491914749145508,
+      "step": 103
+    },
+    {
+      "clip_ratio": 0.0029075751081109047,
+      "epoch": 0.14545454545454545,
+      "grad_norm": 0.13328591024866146,
+      "kl": 0.39453125,
+      "learning_rate": 3.6363636363636364e-05,
+      "loss": 0.0034,
+      "step": 104
+    },
+    {
+      "clip_ratio": 0.0019614642951637506,
+      "completion_length": 85.26786041259766,
+      "epoch": 0.14685314685314685,
+      "grad_norm": 0.19674478002774554,
+      "kl": 0.333984375,
+      "learning_rate": 3.671328671328672e-05,
+      "loss": -0.0188,
+      "num_tokens": 1045567.0,
+      "reward": 1.6822034120559692,
+      "reward_std": 0.7559517621994019,
+      "rewards/check_gptzero_func": 0.5357142686843872,
+      "rewards/check_perplexity_diff_func": 0.6071428656578064,
+      "rewards/check_winston_local_func": 0.5393460988998413,
+      "step": 105
+    },
+    {
+      "clip_ratio": 0.004534203093498945,
+      "epoch": 0.14825174825174825,
+      "grad_norm": 0.16787611299305724,
+      "kl": 0.3671875,
+      "learning_rate": 3.7062937062937064e-05,
+      "loss": -0.0217,
+      "step": 106
+    },
+    {
+      "clip_ratio": 0.003009920008480549,
+      "completion_length": 106.55357360839844,
+      "epoch": 0.14965034965034965,
+      "grad_norm": 0.19751132540015684,
+      "kl": 0.455078125,
+      "learning_rate": 3.741258741258741e-05,
+      "loss": 0.0111,
+      "num_tokens": 1066466.0,
+      "reward": 2.178619861602783,
+      "reward_std": 0.7016831040382385,
+      "rewards/check_gptzero_func": 0.6071428656578064,
+      "rewards/check_perplexity_diff_func": 0.8928571343421936,
+      "rewards/check_winston_local_func": 0.6786197423934937,
+      "step": 107
+    },
+    {
+      "clip_ratio": 0.004473494831472635,
+      "epoch": 0.15104895104895105,
+      "grad_norm": 0.1431470008572649,
+      "kl": 0.4375,
+      "learning_rate": 3.776223776223776e-05,
+      "loss": 0.0095,
+      "step": 108
+    },
+    {
+      "clip_ratio": 0.0026541000697761774,
+      "completion_length": 100.6964340209961,
+      "epoch": 0.15244755244755245,
+      "grad_norm": 0.33135604940901486,
+      "kl": 0.310546875,
+      "learning_rate": 3.811188811188811e-05,
+      "loss": 0.0138,
+      "num_tokens": 1086751.0,
+      "reward": 1.6315226554870605,
+      "reward_std": 0.7806248068809509,
+      "rewards/check_gptzero_func": 0.4642857015132904,
+      "rewards/check_perplexity_diff_func": 0.6071428656578064,
+      "rewards/check_winston_local_func": 0.5600939393043518,
+      "step": 109
+    },
+    {
+      "clip_ratio": 0.0028892713598906994,
+      "epoch": 0.15384615384615385,
+      "grad_norm": 0.24217207616877234,
+      "kl": 0.361328125,
+      "learning_rate": 3.846153846153846e-05,
+      "loss": 0.0074,
+      "step": 110
+    },
+    {
+      "clip_ratio": 0.0021747422870248556,
+      "completion_length": 84.76786041259766,
+      "epoch": 0.15524475524475526,
+      "grad_norm": 0.15803622964815373,
+      "kl": 0.66796875,
+      "learning_rate": 3.8811188811188816e-05,
+      "loss": -0.0045,
+      "num_tokens": 1104518.0,
+      "reward": 1.6692500114440918,
+      "reward_std": 0.5965060591697693,
+      "rewards/check_gptzero_func": 0.5535714030265808,
+      "rewards/check_perplexity_diff_func": 0.6071428656578064,
+      "rewards/check_winston_local_func": 0.5085356831550598,
+      "step": 111
+    },
+    {
+      "clip_ratio": 0.004626331850886345,
+      "epoch": 0.15664335664335666,
+      "grad_norm": 0.14920406573741435,
+      "kl": 0.7578125,
+      "learning_rate": 3.916083916083916e-05,
+      "loss": -0.0064,
+      "step": 112
+    },
+    {
+      "clip_ratio": 0.002694516209885478,
+      "completion_length": 92.17857360839844,
+      "epoch": 0.15804195804195803,
+      "grad_norm": 0.161427063551978,
+      "kl": 0.361328125,
+      "learning_rate": 3.9510489510489516e-05,
+      "loss": 0.0058,
+      "num_tokens": 1123466.0,
+      "reward": 1.353637456893921,
+      "reward_std": 0.5352396965026855,
+      "rewards/check_gptzero_func": 0.4821428656578064,
+      "rewards/check_perplexity_diff_func": 0.3571428656578064,
+      "rewards/check_winston_local_func": 0.5143517255783081,
+      "step": 113
+    },
+    {
+      "clip_ratio": 0.00391918933019042,
+      "epoch": 0.15944055944055943,
+      "grad_norm": 0.14561571783883442,
+      "kl": 0.39453125,
+      "learning_rate": 3.986013986013986e-05,
+      "loss": 0.0036,
+      "step": 114
+    },
+    {
+      "clip_ratio": 0.001278667594306171,
+      "completion_length": 107.12500762939453,
+      "epoch": 0.16083916083916083,
+      "grad_norm": 0.1805624176716803,
+      "kl": 0.3046875,
+      "learning_rate": 4.020979020979021e-05,
+      "loss": -0.001,
+      "num_tokens": 1144619.0,
+      "reward": 1.6282455921173096,
+      "reward_std": 0.7325619459152222,
+      "rewards/check_gptzero_func": 0.4464285671710968,
+      "rewards/check_perplexity_diff_func": 0.5357142686843872,
+      "rewards/check_winston_local_func": 0.6461027264595032,
+      "step": 115
+    },
+    {
+      "clip_ratio": 0.0017050534952431917,
+      "epoch": 0.16223776223776223,
+      "grad_norm": 0.16942268791902212,
+      "kl": 0.337890625,
+      "learning_rate": 4.055944055944056e-05,
+      "loss": -0.0027,
+      "step": 116
+    },
+    {
+      "clip_ratio": 0.0015462420415133238,
+      "completion_length": 114.8214340209961,
+      "epoch": 0.16363636363636364,
+      "grad_norm": 0.1777133538888581,
+      "kl": 0.322265625,
+      "learning_rate": 4.0909090909090915e-05,
+      "loss": -0.0028,
+      "num_tokens": 1166185.0,
+      "reward": 1.566156268119812,
+      "reward_std": 0.38663557171821594,
+      "rewards/check_gptzero_func": 0.4642857015132904,
+      "rewards/check_perplexity_diff_func": 0.5,
+      "rewards/check_winston_local_func": 0.6018705368041992,
+      "step": 117
+    },
+    {
+      "clip_ratio": 0.0030793240293860435,
+      "epoch": 0.16503496503496504,
+      "grad_norm": 0.15084856022622706,
+      "kl": 0.2890625,
+      "learning_rate": 4.125874125874126e-05,
+      "loss": -0.0048,
+      "step": 118
+    },
+    {
+      "clip_ratio": 0.0018360918620601296,
+      "completion_length": 100.25000762939453,
+      "epoch": 0.16643356643356644,
+      "grad_norm": 0.10943003413749244,
+      "kl": 0.83984375,
+      "learning_rate": 4.1608391608391614e-05,
+      "loss": 0.0043,
+      "num_tokens": 1186317.0,
+      "reward": 1.7977957725524902,
+      "reward_std": 0.5172301530838013,
+      "rewards/check_gptzero_func": 0.5714285969734192,
+      "rewards/check_perplexity_diff_func": 0.7142857313156128,
+      "rewards/check_winston_local_func": 0.5120813250541687,
+      "step": 119
+    },
+    {
+      "clip_ratio": 0.0025589358992874622,
+      "epoch": 0.16783216783216784,
+      "grad_norm": 0.10431032316313237,
+      "kl": 0.68359375,
+      "learning_rate": 4.195804195804196e-05,
+      "loss": 0.0029,
+      "step": 120
+    },
+    {
+      "clip_ratio": 0.002800893737003207,
+      "completion_length": 120.91072082519531,
+      "epoch": 0.16923076923076924,
+      "grad_norm": 0.1567960287886374,
+      "kl": 0.37890625,
+      "learning_rate": 4.230769230769231e-05,
+      "loss": -0.0016,
+      "num_tokens": 1208530.0,
+      "reward": 1.5369055271148682,
+      "reward_std": 0.7073518633842468,
+      "rewards/check_gptzero_func": 0.3214285671710968,
+      "rewards/check_perplexity_diff_func": 0.6785714030265808,
+      "rewards/check_winston_local_func": 0.5369054079055786,
+      "step": 121
+    },
+    {
+      "clip_ratio": 0.0026409339625388384,
+      "epoch": 0.17062937062937064,
+      "grad_norm": 0.15013150065673506,
+      "kl": 0.37890625,
+      "learning_rate": 4.265734265734266e-05,
+      "loss": -0.0035,
+      "step": 122
+    },
+    {
+      "clip_ratio": 0.002211854327470064,
+      "completion_length": 110.66072082519531,
+      "epoch": 0.17202797202797201,
+      "grad_norm": 0.17274409069851862,
+      "kl": 0.380859375,
+      "learning_rate": 4.300699300699301e-05,
+      "loss": -0.0062,
+      "num_tokens": 1229171.0,
+      "reward": 1.6140996217727661,
+      "reward_std": 0.7221139669418335,
+      "rewards/check_gptzero_func": 0.4642857015132904,
+      "rewards/check_perplexity_diff_func": 0.5357142686843872,
+      "rewards/check_winston_local_func": 0.6140995025634766,
+      "step": 123
+    },
+    {
+      "clip_ratio": 0.005159804597496986,
+      "epoch": 0.17342657342657342,
+      "grad_norm": 0.13306053466541726,
+      "kl": 0.443359375,
+      "learning_rate": 4.335664335664335e-05,
+      "loss": -0.009,
+      "step": 124
+    },
+    {
+      "clip_ratio": 0.0031610180158168077,
+      "completion_length": 104.30357360839844,
+      "epoch": 0.17482517482517482,
+      "grad_norm": 0.19261275892706695,
+      "kl": 0.30078125,
+      "learning_rate": 4.370629370629371e-05,
+      "loss": -0.0129,
+      "num_tokens": 1249350.0,
+      "reward": 1.9047484397888184,
+      "reward_std": 0.6767317652702332,
+      "rewards/check_gptzero_func": 0.6071428656578064,
+      "rewards/check_perplexity_diff_func": 0.6071428656578064,
+      "rewards/check_winston_local_func": 0.6904626488685608,
+      "step": 125
+    },
+    {
+      "clip_ratio": 0.004231306724250317,
+      "epoch": 0.17622377622377622,
+      "grad_norm": 0.1752737652506695,
+      "kl": 0.33203125,
+      "learning_rate": 4.405594405594406e-05,
+      "loss": -0.0162,
+      "step": 126
+    },
+    {
+      "clip_ratio": 0.003981932066380978,
+      "completion_length": 109.53572082519531,
+      "epoch": 0.17762237762237762,
+      "grad_norm": 0.1423250496935692,
+      "kl": 0.48828125,
+      "learning_rate": 4.4405594405594406e-05,
+      "loss": -0.0102,
+      "num_tokens": 1269848.0,
+      "reward": 1.7591207027435303,
+      "reward_std": 0.5321380496025085,
+      "rewards/check_gptzero_func": 0.5,
+      "rewards/check_perplexity_diff_func": 0.6785714030265808,
+      "rewards/check_winston_local_func": 0.5805493593215942,
+      "step": 127
+    },
+    {
+      "clip_ratio": 0.005287667270749807,
+      "epoch": 0.17902097902097902,
+      "grad_norm": 0.13255848380623775,
+      "kl": 0.498046875,
+      "learning_rate": 4.475524475524476e-05,
+      "loss": -0.0121,
+      "step": 128
+    },
+    {
+      "clip_ratio": 0.0025025398936122656,
+      "completion_length": 105.30357360839844,
+      "epoch": 0.18041958041958042,
+      "grad_norm": 0.1177341259986552,
+      "kl": 0.283203125,
+      "learning_rate": 4.5104895104895105e-05,
+      "loss": -0.0016,
+      "num_tokens": 1290033.0,
+      "reward": 1.7174798250198364,
+      "reward_std": 0.5096268057823181,
+      "rewards/check_gptzero_func": 0.5357142686843872,
+      "rewards/check_perplexity_diff_func": 0.5357142686843872,
+      "rewards/check_winston_local_func": 0.6460510492324829,
+      "step": 129
+    },
+    {
+      "clip_ratio": 0.0033291254658252,
+      "epoch": 0.18181818181818182,
+      "grad_norm": 0.11141937591016414,
+      "kl": 0.294921875,
+      "learning_rate": 4.545454545454546e-05,
+      "loss": -0.0032,
+      "step": 130
+    },
+    {
+      "clip_ratio": 0.0024425899609923363,
+      "completion_length": 96.30357360839844,
+      "epoch": 0.18321678321678322,
+      "grad_norm": 0.23593816935965856,
+      "kl": 0.341796875,
+      "learning_rate": 4.5804195804195805e-05,
+      "loss": -0.006,
+      "num_tokens": 1309356.0,
+      "reward": 1.7568891048431396,
+      "reward_std": 0.7255779504776001,
+      "rewards/check_gptzero_func": 0.5892857313156128,
+      "rewards/check_perplexity_diff_func": 0.5357142686843872,
+      "rewards/check_winston_local_func": 0.6318890452384949,
+      "step": 131
+    },
+    {
+      "clip_ratio": 0.004331678152084351,
+      "epoch": 0.18461538461538463,
+      "grad_norm": 0.2135627059179365,
+      "kl": 0.349609375,
+      "learning_rate": 4.615384615384616e-05,
+      "loss": -0.0089,
+      "step": 132
+    },
+    {
+      "clip_ratio": 0.0028827630449086428,
+      "completion_length": 111.6964340209961,
+      "epoch": 0.18601398601398603,
+      "grad_norm": 0.1622976058137894,
+      "kl": 0.369140625,
+      "learning_rate": 4.6503496503496505e-05,
+      "loss": 0.014,
+      "num_tokens": 1330017.0,
+      "reward": 1.6692737340927124,
+      "reward_std": 0.628279983997345,
+      "rewards/check_gptzero_func": 0.4464285671710968,
+      "rewards/check_perplexity_diff_func": 0.6428571343421936,
+      "rewards/check_winston_local_func": 0.5799878835678101,
+      "step": 133
+    },
+    {
+      "clip_ratio": 0.004608546383678913,
+      "epoch": 0.1874125874125874,
+      "grad_norm": 0.16503328062657166,
+      "kl": 0.4140625,
+      "learning_rate": 4.685314685314686e-05,
+      "loss": 0.0121,
+      "step": 134
+    },
+    {
+      "clip_ratio": 0.0027426625601947308,
+      "completion_length": 107.58928680419922,
+      "epoch": 0.1888111888111888,
+      "grad_norm": 0.27926954484170907,
+      "kl": 0.5859375,
+      "learning_rate": 4.7202797202797204e-05,
+      "loss": 0.0139,
+      "num_tokens": 1350164.0,
+      "reward": 1.9428951740264893,
+      "reward_std": 0.6565932035446167,
+      "rewards/check_gptzero_func": 0.6785714030265808,
+      "rewards/check_perplexity_diff_func": 0.5714285969734192,
+      "rewards/check_winston_local_func": 0.6928950548171997,
+      "step": 135
+    },
+    {
+      "clip_ratio": 0.00439854059368372,
+      "epoch": 0.1902097902097902,
+      "grad_norm": 0.1925514269864402,
+      "kl": 0.5546875,
+      "learning_rate": 4.755244755244756e-05,
+      "loss": 0.0074,
+      "step": 136
+    },
+    {
+      "clip_ratio": 0.0018738384824246168,
+      "completion_length": 116.4464340209961,
+      "epoch": 0.1916083916083916,
+      "grad_norm": 0.14918417137816656,
+      "kl": 0.58984375,
+      "learning_rate": 4.7902097902097904e-05,
+      "loss": 0.0037,
+      "num_tokens": 1371507.0,
+      "reward": 2.0137577056884766,
+      "reward_std": 0.5378711819648743,
+      "rewards/check_gptzero_func": 0.6071428656578064,
+      "rewards/check_perplexity_diff_func": 0.6785714030265808,
+      "rewards/check_winston_local_func": 0.7280434370040894,
+      "step": 137
+    },
+    {
+      "clip_ratio": 0.003529219189658761,
+      "epoch": 0.193006993006993,
+      "grad_norm": 0.1301060608041501,
+      "kl": 0.6171875,
+      "learning_rate": 4.825174825174825e-05,
+      "loss": 0.0016,
+      "step": 138
+    },
+    {
+      "clip_ratio": 0.0011985624441877007,
+      "completion_length": 112.6964340209961,
+      "epoch": 0.1944055944055944,
+      "grad_norm": 0.13034934146859325,
+      "kl": 0.322265625,
+      "learning_rate": 4.86013986013986e-05,
+      "loss": -0.0024,
+      "num_tokens": 1392414.0,
+      "reward": 1.9355616569519043,
+      "reward_std": 0.4726971983909607,
+      "rewards/check_gptzero_func": 0.5892857313156128,
+      "rewards/check_perplexity_diff_func": 0.75,
+      "rewards/check_winston_local_func": 0.596275806427002,
+      "step": 139
+    },
+    {
+      "clip_ratio": 0.0032699662260711193,
+      "epoch": 0.1958041958041958,
+      "grad_norm": 0.12091032490165694,
+      "kl": 0.318359375,
+      "learning_rate": 4.8951048951048956e-05,
+      "loss": -0.0039,
+      "step": 140
+    },
+    {
+      "clip_ratio": 0.0013384540798142552,
+      "completion_length": 132.05357360839844,
+      "epoch": 0.1972027972027972,
+      "grad_norm": 0.12728573210952063,
+      "kl": 0.59765625,
+      "learning_rate": 4.93006993006993e-05,
+      "loss": 0.0045,
+      "num_tokens": 1416177.0,
+      "reward": 1.7924094200134277,
+      "reward_std": 0.7202263474464417,
+      "rewards/check_gptzero_func": 0.5714285969734192,
+      "rewards/check_perplexity_diff_func": 0.7142857313156128,
+      "rewards/check_winston_local_func": 0.5066950917243958,
+      "step": 141
+    },
+    {
+      "clip_ratio": 0.0017050639726221561,
+      "epoch": 0.1986013986013986,
+      "grad_norm": 0.12231361158637834,
+      "kl": 0.65234375,
+      "learning_rate": 4.9650349650349656e-05,
+      "loss": 0.0025,
+      "step": 142
+    },
+    {
+      "clip_ratio": 0.0015860958956182003,
+      "completion_length": 99.73214721679688,
+      "epoch": 0.2,
+      "grad_norm": 0.1853946177277763,
+      "kl": 0.58984375,
+      "learning_rate": 5e-05,
+      "loss": 0.0021,
+      "num_tokens": 1435732.0,
+      "reward": 2.0557591915130615,
+      "reward_std": 0.503829836845398,
+      "rewards/check_gptzero_func": 0.6964285969734192,
+      "rewards/check_perplexity_diff_func": 0.75,
+      "rewards/check_winston_local_func": 0.6093305945396423,
+      "step": 143
+    },
+    {
+      "clip_ratio": 0.002123563550412655,
+      "epoch": 0.2013986013986014,
+      "grad_norm": 0.13470126319919157,
+      "kl": 0.423828125,
+      "learning_rate": 5.0349650349650356e-05,
+      "loss": -0.0009,
+      "step": 144
+    },
+    {
+      "clip_ratio": 0.0018757216166704893,
+      "completion_length": 95.73214721679688,
+      "epoch": 0.20279720279720279,
+      "grad_norm": 0.14930190515465938,
+      "kl": 0.388671875,
+      "learning_rate": 5.06993006993007e-05,
+      "loss": -0.0026,
+      "num_tokens": 1454705.0,
+      "reward": 2.0560998916625977,
+      "reward_std": 0.5541732311248779,
+      "rewards/check_gptzero_func": 0.6964285969734192,
+      "rewards/check_perplexity_diff_func": 0.5714285969734192,
+      "rewards/check_winston_local_func": 0.788242518901825,
+      "step": 145
+    },
+    {
+      "clip_ratio": 0.003540371311828494,
+      "epoch": 0.2041958041958042,
+      "grad_norm": 0.1304251509197979,
+      "kl": 0.41796875,
+      "learning_rate": 5.1048951048951055e-05,
+      "loss": -0.0046,
+      "step": 146
+    },
+    {
+      "clip_ratio": 0.0031265008728951216,
+      "completion_length": 80.9464340209961,
+      "epoch": 0.2055944055944056,
+      "grad_norm": 0.2087929657840581,
+      "kl": 1.1328125,
+      "learning_rate": 5.1398601398601395e-05,
+      "loss": 0.0026,
+      "num_tokens": 1472182.0,
+      "reward": 1.9076076745986938,
+      "reward_std": 0.46633273363113403,
+      "rewards/check_gptzero_func": 0.6428571343421936,
+      "rewards/check_perplexity_diff_func": 0.5,
+      "rewards/check_winston_local_func": 0.7647504210472107,
+      "step": 147
+    },
+    {
+      "clip_ratio": 0.0029583375435322523,
+      "epoch": 0.206993006993007,
+      "grad_norm": 0.178151823438605,
+      "kl": 0.66015625,
+      "learning_rate": 5.1748251748251755e-05,
+      "loss": -0.0014,
+      "step": 148
+    },
+    {
+      "clip_ratio": 0.0020277961157262325,
+      "completion_length": 126.58929443359375,
+      "epoch": 0.2083916083916084,
+      "grad_norm": 0.13736153026044381,
+      "kl": 0.373046875,
+      "learning_rate": 5.2097902097902094e-05,
+      "loss": 0.0127,
+      "num_tokens": 1494581.0,
+      "reward": 1.870171308517456,
+      "reward_std": 0.4412032663822174,
+      "rewards/check_gptzero_func": 0.4464285671710968,
+      "rewards/check_perplexity_diff_func": 0.8214285969734192,
+      "rewards/check_winston_local_func": 0.6023141741752625,
+      "step": 149
+    },
+    {
+      "clip_ratio": 0.0022643795236945152,
+      "epoch": 0.2097902097902098,
+      "grad_norm": 0.12942677801199462,
+      "kl": 0.40625,
+      "learning_rate": 5.244755244755245e-05,
+      "loss": 0.0106,
+      "step": 150
+    },
+    {
+      "clip_ratio": 0.001886948710307479,
+      "completion_length": 95.14286041259766,
+      "epoch": 0.2111888111888112,
+      "grad_norm": 0.15555312890089615,
+      "kl": 0.62890625,
+      "learning_rate": 5.279720279720281e-05,
+      "loss": 0.0013,
+      "num_tokens": 1513403.0,
+      "reward": 1.9735006093978882,
+      "reward_std": 0.5218394994735718,
+      "rewards/check_gptzero_func": 0.5535714030265808,
+      "rewards/check_perplexity_diff_func": 0.6071428656578064,
+      "rewards/check_winston_local_func": 0.8127861022949219,
+      "step": 151
+    },
+    {
+      "clip_ratio": 0.003476199461147189,
+      "epoch": 0.2125874125874126,
+      "grad_norm": 0.13833602394582134,
+      "kl": 0.5390625,
+      "learning_rate": 5.314685314685315e-05,
+      "loss": -0.0015,
+      "step": 152
+    },
+    {
+      "clip_ratio": 0.0029247915372252464,
+      "completion_length": 104.1964340209961,
+      "epoch": 0.213986013986014,
+      "grad_norm": 0.18332277821774023,
+      "kl": 0.451171875,
+      "learning_rate": 5.34965034965035e-05,
+      "loss": -0.0047,
+      "num_tokens": 1533154.0,
+      "reward": 1.977178931236267,
+      "reward_std": 0.5758928656578064,
+      "rewards/check_gptzero_func": 0.6785714030265808,
+      "rewards/check_perplexity_diff_func": 0.6785714030265808,
+      "rewards/check_winston_local_func": 0.6200361251831055,
+      "step": 153
+    },
+    {
+      "clip_ratio": 0.002894646255299449,
+      "epoch": 0.2153846153846154,
+      "grad_norm": 0.12417162413759981,
+      "kl": 0.48828125,
+      "learning_rate": 5.384615384615385e-05,
+      "loss": -0.0082,
+      "step": 154
+    },
+    {
+      "clip_ratio": 0.002533347113057971,
+      "completion_length": 106.6964340209961,
+      "epoch": 0.21678321678321677,
+      "grad_norm": 0.2165608004438418,
+      "kl": 0.4140625,
+      "learning_rate": 5.41958041958042e-05,
+      "loss": 0.0367,
+      "num_tokens": 1554005.0,
+      "reward": 2.254620313644409,
+      "reward_std": 0.6860859394073486,
+      "rewards/check_gptzero_func": 0.7857142686843872,
+      "rewards/check_perplexity_diff_func": 0.7142857313156128,
+      "rewards/check_winston_local_func": 0.7546200752258301,
+      "step": 155
+    },
+    {
+      "clip_ratio": 0.005156368017196655,
+      "epoch": 0.21818181818181817,
+      "grad_norm": 0.16185524517286934,
+      "kl": 0.4609375,
+      "learning_rate": 5.4545454545454546e-05,
+      "loss": 0.0322,
+      "step": 156
+    },
+    {
+      "clip_ratio": 0.002192203886806965,
+      "completion_length": 129.55357360839844,
+      "epoch": 0.21958041958041957,
+      "grad_norm": 0.13686573639431518,
+      "kl": 0.421875,
+      "learning_rate": 5.48951048951049e-05,
+      "loss": 0.0095,
+      "num_tokens": 1576988.0,
+      "reward": 2.202296495437622,
+      "reward_std": 0.4840867817401886,
+      "rewards/check_gptzero_func": 0.6428571343421936,
+      "rewards/check_perplexity_diff_func": 0.8928571343421936,
+      "rewards/check_winston_local_func": 0.6665821075439453,
+      "step": 157
+    },
+    {
+      "clip_ratio": 0.0037054666317999363,
+      "epoch": 0.22097902097902097,
+      "grad_norm": 0.12040805378810834,
+      "kl": 0.45703125,
+      "learning_rate": 5.524475524475524e-05,
+      "loss": 0.0077,
+      "step": 158
+    },
+    {
+      "clip_ratio": 0.002324200002476573,
+      "completion_length": 141.10714721679688,
+      "epoch": 0.22237762237762237,
+      "grad_norm": 0.1722210877266639,
+      "kl": 0.9765625,
+      "learning_rate": 5.55944055944056e-05,
+      "loss": -0.0049,
+      "num_tokens": 1600826.0,
+      "reward": 2.0245378017425537,
+      "reward_std": 0.45877185463905334,
+      "rewards/check_gptzero_func": 0.5892857313156128,
+      "rewards/check_perplexity_diff_func": 0.75,
+      "rewards/check_winston_local_func": 0.6852518916130066,
+      "step": 159
+    },
+    {
+      "clip_ratio": 0.003470373572781682,
+      "epoch": 0.22377622377622378,
+      "grad_norm": 0.13356363193434526,
+      "kl": 0.65234375,
+      "learning_rate": 5.594405594405595e-05,
+      "loss": -0.0085,
+      "step": 160
+    },
+    {
+      "clip_ratio": 0.0026535126380622387,
+      "completion_length": 109.76786041259766,
+      "epoch": 0.22517482517482518,
+      "grad_norm": 0.409404915148595,
+      "kl": 0.515625,
+      "learning_rate": 5.629370629370629e-05,
+      "loss": -0.0051,
+      "num_tokens": 1621103.0,
+      "reward": 2.0040969848632812,
+      "reward_std": 0.5942196249961853,
+      "rewards/check_gptzero_func": 0.5178571343421936,
+      "rewards/check_perplexity_diff_func": 0.7857142686843872,
+      "rewards/check_winston_local_func": 0.7005256414413452,
+      "step": 161
+    },
+    {
+      "clip_ratio": 0.0058495416305959225,
+      "epoch": 0.22657342657342658,
+      "grad_norm": 0.16376328147020022,
+      "kl": 0.5078125,
+      "learning_rate": 5.664335664335665e-05,
+      "loss": -0.0129,
+      "step": 162
+    },
+    {
+      "clip_ratio": 0.002924378262832761,
+      "completion_length": 114.3214340209961,
+      "epoch": 0.22797202797202798,
+      "grad_norm": 0.17698045766822795,
+      "kl": 0.62109375,
+      "learning_rate": 5.699300699300699e-05,
+      "loss": -0.0053,
+      "num_tokens": 1642629.0,
+      "reward": 2.0913121700286865,
+      "reward_std": 0.533393144607544,
+      "rewards/check_gptzero_func": 0.6785714030265808,
+      "rewards/check_perplexity_diff_func": 0.6785714030265808,
+      "rewards/check_winston_local_func": 0.7341693639755249,
+      "step": 163
+    },
+    {
+      "clip_ratio": 0.006533453240990639,
+      "epoch": 0.22937062937062938,
+      "grad_norm": 0.15566114152311913,
+      "kl": 0.66796875,
+      "learning_rate": 5.7342657342657345e-05,
+      "loss": -0.0085,
+      "step": 164
+    },
+    {
+      "clip_ratio": 0.002033352619037032,
+      "completion_length": 110.48214721679688,
+      "epoch": 0.23076923076923078,
+      "grad_norm": 0.19340316318251602,
+      "kl": 0.4453125,
+      "learning_rate": 5.769230769230769e-05,
+      "loss": 0.0054,
+      "num_tokens": 1663000.0,
+      "reward": 1.809409499168396,
+      "reward_std": 0.5209853053092957,
+      "rewards/check_gptzero_func": 0.6428571343421936,
+      "rewards/check_perplexity_diff_func": 0.3571428656578064,
+      "rewards/check_winston_local_func": 0.809409499168396,
+      "step": 165
+    },
+    {
+      "clip_ratio": 0.005329853855073452,
+      "epoch": 0.23216783216783216,
+      "grad_norm": 0.15299329983895626,
+      "kl": 0.55078125,
+      "learning_rate": 5.8041958041958044e-05,
+      "loss": 0.0018,
+      "step": 166
+    },
+    {
+      "clip_ratio": 0.0025733087677508593,
+      "completion_length": 95.73214721679688,
+      "epoch": 0.23356643356643356,
+      "grad_norm": 0.16102285972903455,
+      "kl": 0.458984375,
+      "learning_rate": 5.83916083916084e-05,
+      "loss": -0.0008,
+      "num_tokens": 1682123.0,
+      "reward": 1.9705681800842285,
+      "reward_std": 0.5465096235275269,
+      "rewards/check_gptzero_func": 0.6428571343421936,
+      "rewards/check_perplexity_diff_func": 0.6428571343421936,
+      "rewards/check_winston_local_func": 0.6848538517951965,
+      "step": 167
+    },
+    {
+      "clip_ratio": 0.0034487086813896894,
+      "epoch": 0.23496503496503496,
+      "grad_norm": 0.12896214559568192,
+      "kl": 0.474609375,
+      "learning_rate": 5.8741258741258744e-05,
+      "loss": -0.0039,
+      "step": 168
+    },
+    {
+      "clip_ratio": 0.0012721805833280087,
+      "completion_length": 72.125,
+      "epoch": 0.23636363636363636,
+      "grad_norm": 0.19976827264055144,
+      "kl": 0.5703125,
+      "learning_rate": 5.90909090909091e-05,
+      "loss": -0.0046,
+      "num_tokens": 1698078.0,
+      "reward": 1.8698246479034424,
+      "reward_std": 0.4430284798145294,
+      "rewards/check_gptzero_func": 0.7142857313156128,
+      "rewards/check_perplexity_diff_func": 0.3571428656578064,
+      "rewards/check_winston_local_func": 0.7983959913253784,
+      "step": 169
+    },
+    {
+      "clip_ratio": 0.005269515328109264,
+      "epoch": 0.23776223776223776,
+      "grad_norm": 0.14855282442633314,
+      "kl": 0.61328125,
+      "learning_rate": 5.944055944055944e-05,
+      "loss": -0.0079,
+      "step": 170
+    },
+    {
+      "clip_ratio": 0.0027731256559491158,
+      "completion_length": 117.83929443359375,
+      "epoch": 0.23916083916083916,
+      "grad_norm": 0.23058249563979205,
+      "kl": 0.412109375,
+      "learning_rate": 5.9790209790209796e-05,
+      "loss": 0.0055,
+      "num_tokens": 1719673.0,
+      "reward": 2.2663450241088867,
+      "reward_std": 0.46630793809890747,
+      "rewards/check_gptzero_func": 0.6785714030265808,
+      "rewards/check_perplexity_diff_func": 0.7142857313156128,
+      "rewards/check_winston_local_func": 0.8734878301620483,
+      "step": 171
+    },
+    {
+      "clip_ratio": 0.0034511894918978214,
+      "epoch": 0.24055944055944056,
+      "grad_norm": 0.17368207545699044,
+      "kl": 0.373046875,
+      "learning_rate": 6.0139860139860136e-05,
+      "loss": -0.0007,
+      "step": 172
+    },
+    {
+      "clip_ratio": 0.0032431341242045164,
+      "completion_length": 126.85714721679688,
+      "epoch": 0.24195804195804196,
+      "grad_norm": 0.17995690811350934,
+      "kl": 0.353515625,
+      "learning_rate": 6.048951048951049e-05,
+      "loss": -0.0011,
+      "num_tokens": 1742569.0,
+      "reward": 2.258990526199341,
+      "reward_std": 0.6008436679840088,
+      "rewards/check_gptzero_func": 0.75,
+      "rewards/check_perplexity_diff_func": 0.75,
+      "rewards/check_winston_local_func": 0.7589904069900513,
+      "step": 173
+    },
+    {
+      "clip_ratio": 0.006950656417757273,
+      "epoch": 0.24335664335664337,
+      "grad_norm": 0.13892848528506246,
+      "kl": 0.359375,
+      "learning_rate": 6.083916083916085e-05,
+      "loss": -0.0033,
+      "step": 174
+    },
+    {
+      "clip_ratio": 0.0027662119828164577,
+      "completion_length": 130.8928680419922,
+      "epoch": 0.24475524475524477,
+      "grad_norm": 0.10040261340554028,
+      "kl": 0.345703125,
+      "learning_rate": 6.118881118881119e-05,
+      "loss": -0.0032,
+      "num_tokens": 1765643.0,
+      "reward": 2.074061632156372,
+      "reward_std": 0.3745954632759094,
+      "rewards/check_gptzero_func": 0.6607142686843872,
+      "rewards/check_perplexity_diff_func": 0.6785714030265808,
+      "rewards/check_winston_local_func": 0.7347758412361145,
+      "step": 175
+    },
+    {
+      "clip_ratio": 0.0030742601957172155,
+      "epoch": 0.24615384615384617,
+      "grad_norm": 0.0921134439221548,
+      "kl": 0.3359375,
+      "learning_rate": 6.153846153846155e-05,
+      "loss": -0.0044,
+      "step": 176
+    },
+    {
+      "clip_ratio": 0.002027927665039897,
+      "completion_length": 89.73214721679688,
+      "epoch": 0.24755244755244754,
+      "grad_norm": 0.1709976566266231,
+      "kl": 0.50390625,
+      "learning_rate": 6.188811188811188e-05,
+      "loss": 0.0051,
+      "num_tokens": 1783674.0,
+      "reward": 1.7810758352279663,
+      "reward_std": 0.4943030774593353,
+      "rewards/check_gptzero_func": 0.6428571343421936,
+      "rewards/check_perplexity_diff_func": 0.4285714328289032,
+      "rewards/check_winston_local_func": 0.7096471786499023,
+      "step": 177
+    },
+    {
+      "clip_ratio": 0.0020612890366464853,
+      "epoch": 0.24895104895104894,
+      "grad_norm": 0.1477275348028994,
+      "kl": 0.50390625,
+      "learning_rate": 6.223776223776224e-05,
+      "loss": 0.0011,
+      "step": 178
+    },
+    {
+      "clip_ratio": 0.0021104796323925257,
+      "completion_length": 132.10714721679688,
+      "epoch": 0.25034965034965034,
+      "grad_norm": 0.08842848405697774,
+      "kl": 0.39453125,
+      "learning_rate": 6.258741258741259e-05,
+      "loss": 0.0005,
+      "num_tokens": 1806754.0,
+      "reward": 2.107767105102539,
+      "reward_std": 0.269815057516098,
+      "rewards/check_gptzero_func": 0.75,
+      "rewards/check_perplexity_diff_func": 0.5357142686843872,
+      "rewards/check_winston_local_func": 0.8220529556274414,
+      "step": 179
+    },
+    {
+      "clip_ratio": 0.004407648928463459,
+      "epoch": 0.2517482517482518,
+      "grad_norm": 0.07542595290297384,
+      "kl": 0.416015625,
+      "learning_rate": 6.293706293706293e-05,
+      "loss": -0.0006,
+      "step": 180
+    },
+    {
+      "clip_ratio": 0.0018615310546010733,
+      "completion_length": 89.30357360839844,
+      "epoch": 0.25314685314685315,
+      "grad_norm": 0.2895549842864223,
+      "kl": 0.498046875,
+      "learning_rate": 6.32867132867133e-05,
+      "loss": -0.0082,
+      "num_tokens": 1824795.0,
+      "reward": 1.8801069259643555,
+      "reward_std": 0.5523244738578796,
+      "rewards/check_gptzero_func": 0.6071428656578064,
+      "rewards/check_perplexity_diff_func": 0.5,
+      "rewards/check_winston_local_func": 0.7729640603065491,
+      "step": 181
+    },
+    {
+      "clip_ratio": 0.007555535528808832,
+      "epoch": 0.2545454545454545,
+      "grad_norm": 0.20874610098022578,
+      "kl": 0.59375,
+      "learning_rate": 6.363636363636364e-05,
+      "loss": -0.016,
+      "step": 182
+    },
+    {
+      "clip_ratio": 0.0016618422232568264,
+      "completion_length": 84.46428680419922,
+      "epoch": 0.25594405594405595,
+      "grad_norm": 0.20653753743630154,
+      "kl": 0.5625,
+      "learning_rate": 6.398601398601399e-05,
+      "loss": 0.0058,
+      "num_tokens": 1842573.0,
+      "reward": 2.1319751739501953,
+      "reward_std": 0.5030335187911987,
+      "rewards/check_gptzero_func": 0.7321428656578064,
+      "rewards/check_perplexity_diff_func": 0.5714285969734192,
+      "rewards/check_winston_local_func": 0.8284037709236145,
+      "step": 183
+    },
+    {
+      "clip_ratio": 0.00579726742580533,
+      "epoch": 0.2573426573426573,
+      "grad_norm": 0.15782001907156346,
+      "kl": 0.578125,
+      "learning_rate": 6.433566433566433e-05,
+      "loss": 0.0001,
+      "step": 184
+    },
+    {
+      "clip_ratio": 0.0016176491044461727,
+      "completion_length": 125.39286041259766,
+      "epoch": 0.25874125874125875,
+      "grad_norm": 0.12385736188741701,
+      "kl": 0.4921875,
+      "learning_rate": 6.46853146853147e-05,
+      "loss": -0.0027,
+      "num_tokens": 1864297.0,
+      "reward": 1.7732529640197754,
+      "reward_std": 0.3774341642856598,
+      "rewards/check_gptzero_func": 0.625,
+      "rewards/check_perplexity_diff_func": 0.4285714328289032,
+      "rewards/check_winston_local_func": 0.7196813821792603,
+      "step": 185
+    },
+    {
+      "clip_ratio": 0.00623862212523818,
+      "epoch": 0.2601398601398601,
+      "grad_norm": 0.10877531810926387,
+      "kl": 0.53125,
+      "learning_rate": 6.503496503496504e-05,
+      "loss": -0.0048,
+      "step": 186
+    },
+    {
+      "clip_ratio": 0.0031279984395951033,
+      "completion_length": 98.00000762939453,
+      "epoch": 0.26153846153846155,
+      "grad_norm": 0.3661429387824046,
+      "kl": 0.6328125,
+      "learning_rate": 6.538461538461539e-05,
+      "loss": -0.0063,
+      "num_tokens": 1882931.0,
+      "reward": 2.0694050788879395,
+      "reward_std": 0.6090536117553711,
+      "rewards/check_gptzero_func": 0.6785714030265808,
+      "rewards/check_perplexity_diff_func": 0.5714285969734192,
+      "rewards/check_winston_local_func": 0.819405198097229,
+      "step": 187
+    },
+    {
+      "clip_ratio": 0.010017934255301952,
+      "epoch": 0.2629370629370629,
+      "grad_norm": 0.34977841113727764,
+      "kl": 0.65234375,
+      "learning_rate": 6.573426573426573e-05,
+      "loss": -0.0185,
+      "step": 188
+    },
+    {
+      "clip_ratio": 0.00222155568189919,
+      "completion_length": 133.44644165039062,
+      "epoch": 0.26433566433566436,
+      "grad_norm": 0.18715243199492715,
+      "kl": 0.46484375,
+      "learning_rate": 6.608391608391609e-05,
+      "loss": 0.0062,
+      "num_tokens": 1906084.0,
+      "reward": 2.364468812942505,
+      "reward_std": 0.5593475699424744,
+      "rewards/check_gptzero_func": 0.6964285969734192,
+      "rewards/check_perplexity_diff_func": 0.8571428656578064,
+      "rewards/check_winston_local_func": 0.8108974099159241,
+      "step": 189
+    },
+    {
+      "clip_ratio": 0.008103223517537117,
+      "epoch": 0.26573426573426573,
+      "grad_norm": 0.14058389303467,
+      "kl": 0.45703125,
+      "learning_rate": 6.643356643356644e-05,
+      "loss": 0.0025,
+      "step": 190
+    },
+    {
+      "clip_ratio": 0.0019668787717819214,
+      "completion_length": 142.07144165039062,
+      "epoch": 0.26713286713286716,
+      "grad_norm": 0.14399812178023874,
+      "kl": 0.43359375,
+      "learning_rate": 6.678321678321679e-05,
+      "loss": -0.0031,
+      "num_tokens": 1930224.0,
+      "reward": 2.1098926067352295,
+      "reward_std": 0.5390018820762634,
+      "rewards/check_gptzero_func": 0.8214285969734192,
+      "rewards/check_perplexity_diff_func": 0.4642857015132904,
+      "rewards/check_winston_local_func": 0.8241782784461975,
+      "step": 191
+    },
+    {
+      "clip_ratio": 0.005082913674414158,
+      "epoch": 0.26853146853146853,
+      "grad_norm": 0.12024703070308183,
+      "kl": 0.447265625,
+      "learning_rate": 6.713286713286715e-05,
+      "loss": -0.0055,
+      "step": 192
+    },
+    {
+      "clip_ratio": 0.0032251765951514244,
+      "completion_length": 84.98214721679688,
+      "epoch": 0.2699300699300699,
+      "grad_norm": 0.32527529059153654,
+      "kl": 0.6640625,
+      "learning_rate": 6.748251748251748e-05,
+      "loss": 0.0186,
+      "num_tokens": 1947399.0,
+      "reward": 1.9443087577819824,
+      "reward_std": 0.6417390704154968,
+      "rewards/check_gptzero_func": 0.7678571343421936,
+      "rewards/check_perplexity_diff_func": 0.2857142984867096,
+      "rewards/check_winston_local_func": 0.8907372355461121,
+      "step": 193
+    },
+    {
+      "clip_ratio": 0.008085422217845917,
+      "epoch": 0.27132867132867133,
+      "grad_norm": 0.21806984416977268,
+      "kl": 0.69140625,
+      "learning_rate": 6.783216783216784e-05,
+      "loss": 0.0068,
+      "step": 194
+    },
+    {
+      "clip_ratio": 0.0025381618179380894,
+      "completion_length": 91.66072082519531,
+      "epoch": 0.2727272727272727,
+      "grad_norm": 0.1831840025015104,
+      "kl": 0.53125,
+      "learning_rate": 6.818181818181818e-05,
+      "loss": 0.0014,
+      "num_tokens": 1965402.0,
+      "reward": 2.3979477882385254,
+      "reward_std": 0.45264866948127747,
+      "rewards/check_gptzero_func": 0.7678571343421936,
+      "rewards/check_perplexity_diff_func": 0.7857142686843872,
+      "rewards/check_winston_local_func": 0.8443759083747864,
+      "step": 195
+    },
+    {
+      "clip_ratio": 0.005082620773464441,
+      "epoch": 0.27412587412587414,
+      "grad_norm": 0.14399845631168656,
+      "kl": 0.515625,
+      "learning_rate": 6.853146853146853e-05,
+      "loss": -0.0023,
+      "step": 196
+    },
+    {
+      "clip_ratio": 0.0025359569117426872,
+      "completion_length": 118.80357360839844,
+      "epoch": 0.2755244755244755,
+      "grad_norm": 0.1617404225919446,
+      "kl": 0.5625,
+      "learning_rate": 6.888111888111889e-05,
+      "loss": 0.0146,
+      "num_tokens": 1986731.0,
+      "reward": 2.078895330429077,
+      "reward_std": 0.31289052963256836,
+      "rewards/check_gptzero_func": 0.8035714030265808,
+      "rewards/check_perplexity_diff_func": 0.3928571343421936,
+      "rewards/check_winston_local_func": 0.882466733455658,
+      "step": 197
+    },
+    {
+      "clip_ratio": 0.0062335156835615635,
+      "epoch": 0.27692307692307694,
+      "grad_norm": 0.12006362566688389,
+      "kl": 0.69921875,
+      "learning_rate": 6.923076923076924e-05,
+      "loss": 0.0125,
+      "step": 198
+    },
+    {
+      "clip_ratio": 0.0023628328926861286,
+      "completion_length": 113.76786041259766,
+      "epoch": 0.2783216783216783,
+      "grad_norm": 0.19235514876641358,
+      "kl": 0.58984375,
+      "learning_rate": 6.958041958041958e-05,
+      "loss": -0.0069,
+      "num_tokens": 2007822.0,
+      "reward": 2.0453529357910156,
+      "reward_std": 0.576248288154602,
+      "rewards/check_gptzero_func": 0.7321428656578064,
+      "rewards/check_perplexity_diff_func": 0.5714285969734192,
+      "rewards/check_winston_local_func": 0.7417814135551453,
+      "step": 199
+    },
+    {
+      "clip_ratio": 0.0030245708767324686,
+      "epoch": 0.27972027972027974,
+      "grad_norm": 0.15320753759426498,
+      "kl": 0.578125,
+      "learning_rate": 6.993006993006993e-05,
+      "loss": -0.0115,
+      "step": 200
+    },
+    {
+      "clip_ratio": 0.0028001146856695414,
+      "completion_length": 121.51786041259766,
+      "epoch": 0.2811188811188811,
+      "grad_norm": 0.2770839119590159,
+      "kl": 0.5859375,
+      "learning_rate": 7.027972027972029e-05,
+      "loss": -0.0321,
+      "num_tokens": 2030153.0,
+      "reward": 2.223505735397339,
+      "reward_std": 0.580276370048523,
+      "rewards/check_gptzero_func": 0.8392857313156128,
+      "rewards/check_perplexity_diff_func": 0.5,
+      "rewards/check_winston_local_func": 0.8842198252677917,
+      "step": 201
+    },
+    {
+      "clip_ratio": 0.004435424692928791,
+      "epoch": 0.28251748251748254,
+      "grad_norm": 0.1752123363810394,
+      "kl": 0.61328125,
+      "learning_rate": 7.062937062937062e-05,
+      "loss": -0.0398,
+      "step": 202
+    },
+    {
+      "clip_ratio": 0.003775001736357808,
+      "completion_length": 132.96429443359375,
+      "epoch": 0.2839160839160839,
+      "grad_norm": 0.20528144857768493,
+      "kl": 0.5390625,
+      "learning_rate": 7.097902097902098e-05,
+      "loss": -0.0054,
+      "num_tokens": 2053611.0,
+      "reward": 2.4398131370544434,
+      "reward_std": 0.6120996475219727,
+      "rewards/check_gptzero_func": 0.8214285969734192,
+      "rewards/check_perplexity_diff_func": 0.8571428656578064,
+      "rewards/check_winston_local_func": 0.761241614818573,
+      "step": 203
+    },
+    {
+      "clip_ratio": 0.0037133130244910717,
+      "epoch": 0.2853146853146853,
+      "grad_norm": 0.14916355666745199,
+      "kl": 0.55859375,
+      "learning_rate": 7.132867132867134e-05,
+      "loss": -0.0112,
+      "step": 204
+    },
+    {
+      "clip_ratio": 0.002573356730863452,
+      "completion_length": 109.51786041259766,
+      "epoch": 0.2867132867132867,
+      "grad_norm": 0.15672915743695726,
+      "kl": 1.4921875,
+      "learning_rate": 7.167832167832168e-05,
+      "loss": 0.0032,
+      "num_tokens": 2073846.0,
+      "reward": 2.221467971801758,
+      "reward_std": 0.492183119058609,
+      "rewards/check_gptzero_func": 0.6964285969734192,
+      "rewards/check_perplexity_diff_func": 0.6428571343421936,
+      "rewards/check_winston_local_func": 0.8821821212768555,
+      "step": 205
+    },
+    {
+      "clip_ratio": 0.0036338225472718477,
+      "epoch": 0.2881118881118881,
+      "grad_norm": 0.27785877634425765,
+      "kl": 0.74609375,
+      "learning_rate": 7.202797202797204e-05,
+      "loss": 0.0013,
+      "step": 206
+    },
+    {
+      "clip_ratio": 0.004580673761665821,
+      "completion_length": 95.05357360839844,
+      "epoch": 0.2895104895104895,
+      "grad_norm": 0.1746155777114636,
+      "kl": 0.671875,
+      "learning_rate": 7.237762237762238e-05,
+      "loss": -0.0068,
+      "num_tokens": 2092355.0,
+      "reward": 2.0429205894470215,
+      "reward_std": 0.36572587490081787,
+      "rewards/check_gptzero_func": 0.6964285969734192,
+      "rewards/check_perplexity_diff_func": 0.4642857015132904,
+      "rewards/check_winston_local_func": 0.8822061419487,
+      "step": 207
+    },
+    {
+      "clip_ratio": 0.007289381232112646,
+      "epoch": 0.2909090909090909,
+      "grad_norm": 0.12704093759252294,
+      "kl": 0.75390625,
+      "learning_rate": 7.272727272727273e-05,
+      "loss": -0.0111,
+      "step": 208
+    },
+    {
+      "clip_ratio": 0.0038105440326035023,
+      "completion_length": 113.4464340209961,
+      "epoch": 0.2923076923076923,
+      "grad_norm": 0.20958397036994925,
+      "kl": 0.6015625,
+      "learning_rate": 7.307692307692307e-05,
+      "loss": 0.0082,
+      "num_tokens": 2113542.0,
+      "reward": 2.162785530090332,
+      "reward_std": 0.49807849526405334,
+      "rewards/check_gptzero_func": 0.75,
+      "rewards/check_perplexity_diff_func": 0.5714285969734192,
+      "rewards/check_winston_local_func": 0.8413568139076233,
+      "step": 209
+    },
+    {
+      "clip_ratio": 0.007606918923556805,
+      "epoch": 0.2937062937062937,
+      "grad_norm": 0.1766466418371376,
+      "kl": 0.63671875,
+      "learning_rate": 7.342657342657343e-05,
+      "loss": 0.0025,
+      "step": 210
+    },
+    {
+      "clip_ratio": 0.0019986790139228106,
+      "completion_length": 92.73214721679688,
+      "epoch": 0.2951048951048951,
+      "grad_norm": 0.2716166901527731,
+      "kl": 0.6171875,
+      "learning_rate": 7.377622377622378e-05,
+      "loss": -0.0005,
+      "num_tokens": 2131873.0,
+      "reward": 2.099912405014038,
+      "reward_std": 0.5661742687225342,
+      "rewards/check_gptzero_func": 0.8214285969734192,
+      "rewards/check_perplexity_diff_func": 0.4285714328289032,
+      "rewards/check_winston_local_func": 0.8499122858047485,
+      "step": 211
+    },
+    {
+      "clip_ratio": 0.005917501635849476,
+      "epoch": 0.2965034965034965,
+      "grad_norm": 0.19270405158731238,
+      "kl": 0.62890625,
+      "learning_rate": 7.412587412587413e-05,
+      "loss": -0.0079,
+      "step": 212
+    },
+    {
+      "clip_ratio": 0.002530448604375124,
+      "completion_length": 96.51786041259766,
+      "epoch": 0.29790209790209793,
+      "grad_norm": 0.22723256705117592,
+      "kl": 0.69921875,
+      "learning_rate": 7.447552447552449e-05,
+      "loss": -0.009,
+      "num_tokens": 2150462.0,
+      "reward": 2.1355700492858887,
+      "reward_std": 0.3738899827003479,
+      "rewards/check_gptzero_func": 0.8571428656578064,
+      "rewards/check_perplexity_diff_func": 0.3571428656578064,
+      "rewards/check_winston_local_func": 0.9212842583656311,
+      "step": 213
+    },
+    {
+      "clip_ratio": 0.011397747322916985,
+      "epoch": 0.2993006993006993,
+      "grad_norm": 0.18189402035580596,
+      "kl": 0.7265625,
+      "learning_rate": 7.482517482517482e-05,
+      "loss": -0.0139,
+      "step": 214
+    },
+    {
+      "clip_ratio": 0.004513947293162346,
+      "completion_length": 94.66072082519531,
+      "epoch": 0.3006993006993007,
+      "grad_norm": 0.2675295151019955,
+      "kl": 0.67578125,
+      "learning_rate": 7.517482517482518e-05,
+      "loss": 0.0014,
+      "num_tokens": 2169243.0,
+      "reward": 2.2817916870117188,
+      "reward_std": 0.4118366539478302,
+      "rewards/check_gptzero_func": 0.8571428656578064,
+      "rewards/check_perplexity_diff_func": 0.4642857015132904,
+      "rewards/check_winston_local_func": 0.9603630900382996,
+      "step": 215
+    },
+    {
+      "clip_ratio": 0.010964194312691689,
+      "epoch": 0.3020979020979021,
+      "grad_norm": 0.17031623307626814,
+      "kl": 0.875,
+      "learning_rate": 7.552447552447553e-05,
+      "loss": -0.0057,
+      "step": 216
+    },
+    {
+      "clip_ratio": 0.003371638245880604,
+      "completion_length": 118.17857360839844,
+      "epoch": 0.3034965034965035,
+      "grad_norm": 0.1847364030092435,
+      "kl": 0.60546875,
+      "learning_rate": 7.587412587412587e-05,
+      "loss": -0.0022,
+      "num_tokens": 2190571.0,
+      "reward": 2.1225900650024414,
+      "reward_std": 0.5403507947921753,
+      "rewards/check_gptzero_func": 0.75,
+      "rewards/check_perplexity_diff_func": 0.4642857015132904,
+      "rewards/check_winston_local_func": 0.9083043336868286,
+      "step": 217
+    },
+    {
+      "clip_ratio": 0.01002263929694891,
+      "epoch": 0.3048951048951049,
+      "grad_norm": 0.1449002041728082,
+      "kl": 0.61328125,
+      "learning_rate": 7.622377622377622e-05,
+      "loss": -0.0067,
+      "step": 218
+    },
+    {
+      "clip_ratio": 0.0013253266224637628,
+      "completion_length": 129.33929443359375,
+      "epoch": 0.3062937062937063,
+      "grad_norm": 0.1504069021149342,
+      "kl": 0.578125,
+      "learning_rate": 7.657342657342658e-05,
+      "loss": -0.0075,
+      "num_tokens": 2213408.0,
+      "reward": 2.1974618434906006,
+      "reward_std": 0.4602964520454407,
+      "rewards/check_gptzero_func": 0.7857142686843872,
+      "rewards/check_perplexity_diff_func": 0.5,
+      "rewards/check_winston_local_func": 0.911747395992279,
+      "step": 219
+    },
+    {
+      "clip_ratio": 0.006876418832689524,
+      "epoch": 0.3076923076923077,
+      "grad_norm": 0.11992971275284438,
+      "kl": 0.57421875,
+      "learning_rate": 7.692307692307693e-05,
+      "loss": -0.0101,
+      "step": 220
+    },
+    {
+      "clip_ratio": 0.0024474726524204016,
+      "completion_length": 89.67857360839844,
+      "epoch": 0.3090909090909091,
+      "grad_norm": 0.46915411380362765,
+      "kl": 0.73046875,
+      "learning_rate": 7.727272727272727e-05,
+      "loss": 0.0333,
+      "num_tokens": 2231248.0,
+      "reward": 2.161350727081299,
+      "reward_std": 0.47136664390563965,
+      "rewards/check_gptzero_func": 0.8214285969734192,
+      "rewards/check_perplexity_diff_func": 0.4642857015132904,
+      "rewards/check_winston_local_func": 0.8756363987922668,
+      "step": 221
+    },
+    {
+      "clip_ratio": 0.02464432455599308,
+      "epoch": 0.3104895104895105,
+      "grad_norm": 0.30204889343007607,
+      "kl": 0.8046875,
+      "learning_rate": 7.762237762237763e-05,
+      "loss": 0.0211,
+      "step": 222
+    },
+    {
+      "clip_ratio": 0.0017527465242892504,
+      "completion_length": 134.17857360839844,
+      "epoch": 0.3118881118881119,
+      "grad_norm": 0.18948229022832336,
+      "kl": 0.6328125,
+      "learning_rate": 7.797202797202798e-05,
+      "loss": -0.0049,
+      "num_tokens": 2255162.0,
+      "reward": 2.4446465969085693,
+      "reward_std": 0.5171672105789185,
+      "rewards/check_gptzero_func": 0.7678571343421936,
+      "rewards/check_perplexity_diff_func": 0.7857142686843872,
+      "rewards/check_winston_local_func": 0.8910752534866333,
+      "step": 223
+    },
+    {
+      "clip_ratio": 0.006593752186745405,
+      "epoch": 0.3132867132867133,
+      "grad_norm": 0.15013602627394823,
+      "kl": 0.640625,
+      "learning_rate": 7.832167832167832e-05,
+      "loss": -0.0097,
+      "step": 224
+    },
+    {
+      "clip_ratio": 0.0019926591776311398,
+      "completion_length": 127.00000762939453,
+      "epoch": 0.3146853146853147,
+      "grad_norm": 0.22875481051435598,
+      "kl": 0.6875,
+      "learning_rate": 7.867132867132867e-05,
+      "loss": 0.0021,
+      "num_tokens": 2277320.0,
+      "reward": 2.2040371894836426,
+      "reward_std": 0.5528277158737183,
+      "rewards/check_gptzero_func": 0.7321428656578064,
+      "rewards/check_perplexity_diff_func": 0.5357142686843872,
+      "rewards/check_winston_local_func": 0.936180055141449,
+      "step": 225
+    },
+    {
+      "clip_ratio": 0.005839359946548939,
+      "epoch": 0.31608391608391606,
+      "grad_norm": 0.17829200080123941,
+      "kl": 0.68359375,
+      "learning_rate": 7.902097902097903e-05,
+      "loss": -0.0053,
+      "step": 226
+    },
+    {
+      "clip_ratio": 0.0026252593379467726,
+      "completion_length": 130.0357208251953,
+      "epoch": 0.3174825174825175,
+      "grad_norm": 0.16076101957107508,
+      "kl": 0.66015625,
+      "learning_rate": 7.937062937062938e-05,
+      "loss": -0.0033,
+      "num_tokens": 2299818.0,
+      "reward": 2.3278117179870605,
+      "reward_std": 0.48680880665779114,
+      "rewards/check_gptzero_func": 0.8035714030265808,
+      "rewards/check_perplexity_diff_func": 0.6428571343421936,
+      "rewards/check_winston_local_func": 0.8813830614089966,
+      "step": 227
+    },
+    {
+      "clip_ratio": 0.004954828415066004,
+      "epoch": 0.31888111888111886,
+      "grad_norm": 0.12734264227867195,
+      "kl": 0.6875,
+      "learning_rate": 7.972027972027972e-05,
+      "loss": -0.0071,
+      "step": 228
+    },
+    {
+      "clip_ratio": 0.0034341278951615095,
+      "completion_length": 133.7857208251953,
+      "epoch": 0.3202797202797203,
+      "grad_norm": 0.24065024754984402,
+      "kl": 0.6875,
+      "learning_rate": 8.006993006993007e-05,
+      "loss": -0.0052,
+      "num_tokens": 2322948.0,
+      "reward": 2.5632615089416504,
+      "reward_std": 0.6206622123718262,
+      "rewards/check_gptzero_func": 0.9642857313156128,
+      "rewards/check_perplexity_diff_func": 0.6785714030265808,
+      "rewards/check_winston_local_func": 0.9204041361808777,
+      "step": 229
+    },
+    {
+      "clip_ratio": 0.006521092262119055,
+      "epoch": 0.32167832167832167,
+      "grad_norm": 0.18602361781154253,
+      "kl": 0.6796875,
+      "learning_rate": 8.041958041958042e-05,
+      "loss": -0.013,
+      "step": 230
+    },
+    {
+      "clip_ratio": 0.004207184072583914,
+      "completion_length": 116.53572082519531,
+      "epoch": 0.3230769230769231,
+      "grad_norm": 0.36704569865090053,
+      "kl": 0.69921875,
+      "learning_rate": 8.076923076923078e-05,
+      "loss": 0.0055,
+      "num_tokens": 2344106.0,
+      "reward": 2.4917171001434326,
+      "reward_std": 0.6073458790779114,
+      "rewards/check_gptzero_func": 0.8392857313156128,
+      "rewards/check_perplexity_diff_func": 0.75,
+      "rewards/check_winston_local_func": 0.902431309223175,
+      "step": 231
+    },
+    {
+      "clip_ratio": 0.013317321427166462,
+      "epoch": 0.32447552447552447,
+      "grad_norm": 0.13844634287775134,
+      "kl": 0.73046875,
+      "learning_rate": 8.111888111888112e-05,
+      "loss": -0.0039,
+      "step": 232
+    },
+    {
+      "clip_ratio": 0.004024635534733534,
+      "completion_length": 134.92857360839844,
+      "epoch": 0.3258741258741259,
+      "grad_norm": 0.34168858248313216,
+      "kl": 0.64453125,
+      "learning_rate": 8.146853146853147e-05,
+      "loss": -0.0057,
+      "num_tokens": 2367724.0,
+      "reward": 2.20210599899292,
+      "reward_std": 0.5888614654541016,
+      "rewards/check_gptzero_func": 0.7321428656578064,
+      "rewards/check_perplexity_diff_func": 0.6071428656578064,
+      "rewards/check_winston_local_func": 0.8628200888633728,
+      "step": 233
+    },
+    {
+      "clip_ratio": 0.009708845987915993,
+      "epoch": 0.32727272727272727,
+      "grad_norm": 0.19135959991877635,
+      "kl": 0.67578125,
+      "learning_rate": 8.181818181818183e-05,
+      "loss": -0.0146,
+      "step": 234
+    },
+    {
+      "clip_ratio": 0.005571329966187477,
+      "completion_length": 98.75000762939453,
+      "epoch": 0.32867132867132864,
+      "grad_norm": 0.32328761096657005,
+      "kl": 0.77734375,
+      "learning_rate": 8.216783216783218e-05,
+      "loss": 0.0128,
+      "num_tokens": 2386636.0,
+      "reward": 2.2358410358428955,
+      "reward_std": 0.599101185798645,
+      "rewards/check_gptzero_func": 0.8214285969734192,
+      "rewards/check_perplexity_diff_func": 0.5357142686843872,
+      "rewards/check_winston_local_func": 0.8786982297897339,
+      "step": 235
+    },
+    {
+      "clip_ratio": 0.009436835534870625,
+      "epoch": 0.3300699300699301,
+      "grad_norm": 0.19557085227388898,
+      "kl": 0.8515625,
+      "learning_rate": 8.251748251748252e-05,
+      "loss": 0.0022,
+      "step": 236
+    },
+    {
+      "clip_ratio": 0.006164635997265577,
+      "completion_length": 79.75,
+      "epoch": 0.33146853146853145,
+      "grad_norm": 0.3250433111655094,
+      "kl": 0.98828125,
+      "learning_rate": 8.286713286713287e-05,
+      "loss": -0.0126,
+      "num_tokens": 2403400.0,
+      "reward": 2.1074206829071045,
+      "reward_std": 0.41642776131629944,
+      "rewards/check_gptzero_func": 0.9107142686843872,
+      "rewards/check_perplexity_diff_func": 0.25,
+      "rewards/check_winston_local_func": 0.946706235408783,
+      "step": 237
+    },
+    {
+      "clip_ratio": 0.0066314926370978355,
+      "epoch": 0.3328671328671329,
+      "grad_norm": 0.1612667131666411,
+      "kl": 0.98828125,
+      "learning_rate": 8.321678321678323e-05,
+      "loss": -0.0225,
+      "step": 238
+    },
+    {
+      "clip_ratio": 0.003079179208725691,
+      "completion_length": 117.10714721679688,
+      "epoch": 0.33426573426573425,
+      "grad_norm": 0.2505534626607256,
+      "kl": 0.6796875,
+      "learning_rate": 8.356643356643356e-05,
+      "loss": -0.0103,
+      "num_tokens": 2425354.0,
+      "reward": 2.3166799545288086,
+      "reward_std": 0.5270359516143799,
+      "rewards/check_gptzero_func": 0.8214285969734192,
+      "rewards/check_perplexity_diff_func": 0.5714285969734192,
+      "rewards/check_winston_local_func": 0.9238227605819702,
+      "step": 239
+    },
+    {
+      "clip_ratio": 0.006801524665206671,
+      "epoch": 0.3356643356643357,
+      "grad_norm": 0.16826419365505382,
+      "kl": 0.671875,
+      "learning_rate": 8.391608391608392e-05,
+      "loss": -0.0182,
+      "step": 240
+    },
+    {
+      "clip_ratio": 0.004480496048927307,
+      "completion_length": 125.05357360839844,
+      "epoch": 0.33706293706293705,
+      "grad_norm": 0.26807661638826913,
+      "kl": 0.60546875,
+      "learning_rate": 8.426573426573428e-05,
+      "loss": -0.0086,
+      "num_tokens": 2447333.0,
+      "reward": 2.2939677238464355,
+      "reward_std": 0.39558398723602295,
+      "rewards/check_gptzero_func": 0.9107142686843872,
+      "rewards/check_perplexity_diff_func": 0.5357142686843872,
+      "rewards/check_winston_local_func": 0.8475390672683716,
+      "step": 241
+    },
+    {
+      "clip_ratio": 0.007587091531604528,
+      "epoch": 0.3384615384615385,
+      "grad_norm": 0.13602673124742348,
+      "kl": 0.57421875,
+      "learning_rate": 8.461538461538461e-05,
+      "loss": -0.0152,
+      "step": 242
+    },
+    {
+      "clip_ratio": 0.0034923183266073465,
+      "completion_length": 98.71428680419922,
+      "epoch": 0.33986013986013985,
+      "grad_norm": 0.7773422467135167,
+      "kl": 6.5,
+      "learning_rate": 8.496503496503497e-05,
+      "loss": 0.0071,
+      "num_tokens": 2466435.0,
+      "reward": 2.4026196002960205,
+      "reward_std": 0.4264836013317108,
+      "rewards/check_gptzero_func": 0.8571428656578064,
+      "rewards/check_perplexity_diff_func": 0.6785714030265808,
+      "rewards/check_winston_local_func": 0.8669052124023438,
+      "step": 243
+    },
+    {
+      "clip_ratio": 0.026189187541604042,
+      "epoch": 0.3412587412587413,
+      "grad_norm": 130.9971047023429,
+      "kl": 0.70703125,
+      "learning_rate": 8.531468531468532e-05,
+      "loss": 1.8998,
+      "step": 244
+    },
+    {
+      "clip_ratio": 0.004780410788953304,
+      "completion_length": 135.0,
+      "epoch": 0.34265734265734266,
+      "grad_norm": 0.2329479243774761,
+      "kl": 0.87109375,
+      "learning_rate": 8.566433566433567e-05,
+      "loss": 0.0011,
+      "num_tokens": 2489651.0,
+      "reward": 2.2166569232940674,
+      "reward_std": 0.5797281265258789,
+      "rewards/check_gptzero_func": 0.9107142686843872,
+      "rewards/check_perplexity_diff_func": 0.4285714328289032,
+      "rewards/check_winston_local_func": 0.8773713111877441,
+      "step": 245
+    },
+    {
+      "clip_ratio": 0.021273447200655937,
+      "epoch": 0.34405594405594403,
+      "grad_norm": 0.22672502397660207,
+      "kl": 0.9609375,
+      "learning_rate": 8.601398601398601e-05,
+      "loss": -0.0045,
+      "step": 246
+    },
+    {
+      "clip_ratio": 0.002856873208656907,
+      "completion_length": 122.66072082519531,
+      "epoch": 0.34545454545454546,
+      "grad_norm": 0.19772681964626543,
+      "kl": 0.984375,
+      "learning_rate": 8.636363636363637e-05,
+      "loss": 0.0028,
+      "num_tokens": 2510506.0,
+      "reward": 2.323245048522949,
+      "reward_std": 0.46726977825164795,
+      "rewards/check_gptzero_func": 0.9107142686843872,
+      "rewards/check_perplexity_diff_func": 0.5,
+      "rewards/check_winston_local_func": 0.9125306010246277,
+      "step": 247
+    },
+    {
+      "clip_ratio": 0.02299562282860279,
+      "epoch": 0.34685314685314683,
+      "grad_norm": 0.1812005085005865,
+      "kl": 0.87890625,
+      "learning_rate": 8.67132867132867e-05,
+      "loss": -0.001,
+      "step": 248
+    },
+    {
+      "clip_ratio": 0.0026335662696510553,
+      "completion_length": 116.85714721679688,
+      "epoch": 0.34825174825174826,
+      "grad_norm": 0.28731547808119623,
+      "kl": 0.76953125,
+      "learning_rate": 8.706293706293707e-05,
+      "loss": 0.024,
+      "num_tokens": 2531254.0,
+      "reward": 2.5312695503234863,
+      "reward_std": 0.5851892828941345,
+      "rewards/check_gptzero_func": 0.8571428656578064,
+      "rewards/check_perplexity_diff_func": 0.8214285969734192,
+      "rewards/check_winston_local_func": 0.8526979088783264,
+      "step": 249
+    },
+    {
+      "clip_ratio": 0.012148472480475903,
+      "epoch": 0.34965034965034963,
+      "grad_norm": 0.21250398409595742,
+      "kl": 0.734375,
+      "learning_rate": 8.741258741258743e-05,
+      "loss": 0.0156,
+      "step": 250
+    },
+    {
+      "clip_ratio": 0.0036881309933960438,
+      "completion_length": 109.12500762939453,
+      "epoch": 0.35104895104895106,
+      "grad_norm": 0.278787459589393,
+      "kl": 1.3671875,
+      "learning_rate": 8.776223776223776e-05,
+      "loss": -0.0127,
+      "num_tokens": 2551265.0,
+      "reward": 2.24423885345459,
+      "reward_std": 0.6221095323562622,
+      "rewards/check_gptzero_func": 0.875,
+      "rewards/check_perplexity_diff_func": 0.5,
+      "rewards/check_winston_local_func": 0.8692389130592346,
+      "step": 251
+    },
+    {
+      "clip_ratio": 0.011438491754233837,
+      "epoch": 0.35244755244755244,
+      "grad_norm": 0.21324998809887244,
+      "kl": 1.0859375,
+      "learning_rate": 8.811188811188812e-05,
+      "loss": -0.023,
+      "step": 252
+    },
+    {
+      "clip_ratio": 0.0039367591962218285,
+      "completion_length": 112.14286041259766,
+      "epoch": 0.35384615384615387,
+      "grad_norm": 0.31069817037713615,
+      "kl": 0.91015625,
+      "learning_rate": 8.846153846153847e-05,
+      "loss": 0.0167,
+      "num_tokens": 2572119.0,
+      "reward": 2.492161989212036,
+      "reward_std": 0.6407585144042969,
+      "rewards/check_gptzero_func": 0.875,
+      "rewards/check_perplexity_diff_func": 0.75,
+      "rewards/check_winston_local_func": 0.8671619296073914,
+      "step": 253
+    },
+    {
+      "clip_ratio": 0.005880096461623907,
+      "epoch": 0.35524475524475524,
+      "grad_norm": 0.19008007229653096,
+      "kl": 0.90234375,
+      "learning_rate": 8.881118881118881e-05,
+      "loss": 0.0057,
+      "step": 254
+    },
+    {
+      "clip_ratio": 0.003151817014440894,
+      "completion_length": 120.78572082519531,
+      "epoch": 0.35664335664335667,
+      "grad_norm": 0.4113945542101958,
+      "kl": 0.890625,
+      "learning_rate": 8.916083916083916e-05,
+      "loss": 0.0079,
+      "num_tokens": 2593081.0,
+      "reward": 2.3228745460510254,
+      "reward_std": 0.5329480171203613,
+      "rewards/check_gptzero_func": 0.875,
+      "rewards/check_perplexity_diff_func": 0.6428571343421936,
+      "rewards/check_winston_local_func": 0.8050170540809631,
+      "step": 255
+    },
+    {
+      "clip_ratio": 0.01224872563034296,
+      "epoch": 0.35804195804195804,
+      "grad_norm": 0.24677625747594412,
+      "kl": 1.0078125,
+      "learning_rate": 8.951048951048952e-05,
+      "loss": 0.0006,
+      "step": 256
+    },
+    {
+      "clip_ratio": 0.003666934324428439,
+      "completion_length": 146.5178680419922,
+      "epoch": 0.3594405594405594,
+      "grad_norm": 0.21537926236506497,
+      "kl": 0.484375,
+      "learning_rate": 8.986013986013986e-05,
+      "loss": 0.0068,
+      "num_tokens": 2618758.0,
+      "reward": 2.198406934738159,
+      "reward_std": 0.3875991106033325,
+      "rewards/check_gptzero_func": 0.9285714030265808,
+      "rewards/check_perplexity_diff_func": 0.4285714328289032,
+      "rewards/check_winston_local_func": 0.8412641882896423,
+      "step": 257
+    },
+    {
+      "clip_ratio": 0.012731654569506645,
+      "epoch": 0.36083916083916084,
+      "grad_norm": 0.1543455831571161,
+      "kl": 0.478515625,
+      "learning_rate": 9.020979020979021e-05,
+      "loss": 0.0025,
+      "step": 258
+    },
+    {
+      "clip_ratio": 0.0012050624936819077,
+      "completion_length": 129.375,
+      "epoch": 0.3622377622377622,
+      "grad_norm": 0.16107873823673136,
+      "kl": 1.6640625,
+      "learning_rate": 9.055944055944057e-05,
+      "loss": 0.0054,
+      "num_tokens": 2641815.0,
+      "reward": 2.4805288314819336,
+      "reward_std": 0.3510296940803528,
+      "rewards/check_gptzero_func": 0.8392857313156128,
+      "rewards/check_perplexity_diff_func": 0.8214285969734192,
+      "rewards/check_winston_local_func": 0.8198142647743225,
+      "step": 259
+    },
+    {
+      "clip_ratio": 0.006208478473126888,
+      "epoch": 0.36363636363636365,
+      "grad_norm": 0.10546346547579268,
+      "kl": 0.8046875,
+      "learning_rate": 9.090909090909092e-05,
+      "loss": 0.0019,
+      "step": 260
+    },
+    {
+      "clip_ratio": 0.0025593352038413286,
+      "completion_length": 128.57144165039062,
+      "epoch": 0.365034965034965,
+      "grad_norm": 0.1986264685447053,
+      "kl": 0.6171875,
+      "learning_rate": 9.125874125874126e-05,
+      "loss": -0.0025,
+      "num_tokens": 2664343.0,
+      "reward": 2.261420249938965,
+      "reward_std": 0.5704914927482605,
+      "rewards/check_gptzero_func": 0.7857142686843872,
+      "rewards/check_perplexity_diff_func": 0.6428571343421936,
+      "rewards/check_winston_local_func": 0.8328486680984497,
+      "step": 261
+    },
+    {
+      "clip_ratio": 0.002767725382000208,
+      "epoch": 0.36643356643356645,
+      "grad_norm": 0.14267481856078115,
+      "kl": 0.66015625,
+      "learning_rate": 9.160839160839161e-05,
+      "loss": -0.0071,
+      "step": 262
+    },
+    {
+      "clip_ratio": 0.0036019759718328714,
+      "completion_length": 108.30357360839844,
+      "epoch": 0.3678321678321678,
+      "grad_norm": 0.25429899311591847,
+      "kl": 0.61328125,
+      "learning_rate": 9.195804195804196e-05,
+      "loss": 0.0114,
+      "num_tokens": 2684460.0,
+      "reward": 2.1853349208831787,
+      "reward_std": 0.5545978546142578,
+      "rewards/check_gptzero_func": 0.7142857313156128,
+      "rewards/check_perplexity_diff_func": 0.6428571343421936,
+      "rewards/check_winston_local_func": 0.828191876411438,
+      "step": 263
+    },
+    {
+      "clip_ratio": 0.002995865885168314,
+      "epoch": 0.36923076923076925,
+      "grad_norm": 0.17454926732818832,
+      "kl": 0.62109375,
+      "learning_rate": 9.230769230769232e-05,
+      "loss": 0.0028,
+      "step": 264
+    },
+    {
+      "clip_ratio": 0.002558046253398061,
+      "completion_length": 102.80357360839844,
+      "epoch": 0.3706293706293706,
+      "grad_norm": 0.18701691067416304,
+      "kl": 0.734375,
+      "learning_rate": 9.265734265734266e-05,
+      "loss": 0.0401,
+      "num_tokens": 2704225.0,
+      "reward": 2.1770143508911133,
+      "reward_std": 0.573693037033081,
+      "rewards/check_gptzero_func": 0.8035714030265808,
+      "rewards/check_perplexity_diff_func": 0.5357142686843872,
+      "rewards/check_winston_local_func": 0.8377286195755005,
+      "step": 265
+    },
+    {
+      "clip_ratio": 0.006838452070951462,
+      "epoch": 0.37202797202797205,
+      "grad_norm": 0.13942096443359253,
+      "kl": 0.71875,
+      "learning_rate": 9.300699300699301e-05,
+      "loss": 0.0354,
+      "step": 266
+    },
+    {
+      "clip_ratio": 0.00375689216889441,
+      "completion_length": 106.37500762939453,
+      "epoch": 0.3734265734265734,
+      "grad_norm": 0.25266992310982356,
+      "kl": 0.7421875,
+      "learning_rate": 9.335664335664336e-05,
+      "loss": -0.0156,
+      "num_tokens": 2724530.0,
+      "reward": 2.11887526512146,
+      "reward_std": 0.5366143584251404,
+      "rewards/check_gptzero_func": 0.875,
+      "rewards/check_perplexity_diff_func": 0.4642857015132904,
+      "rewards/check_winston_local_func": 0.7795897126197815,
+      "step": 267
+    },
+    {
+      "clip_ratio": 0.006593361962586641,
+      "epoch": 0.3748251748251748,
+      "grad_norm": 0.18497903526945206,
+      "kl": 0.7265625,
+      "learning_rate": 9.370629370629372e-05,
+      "loss": -0.0243,
+      "step": 268
+    },
+    {
+      "clip_ratio": 0.004569682292640209,
+      "completion_length": 124.39286041259766,
+      "epoch": 0.37622377622377623,
+      "grad_norm": 0.21331332062489702,
+      "kl": 0.953125,
+      "learning_rate": 9.405594405594406e-05,
+      "loss": 0.0346,
+      "num_tokens": 2746246.0,
+      "reward": 2.403602123260498,
+      "reward_std": 0.5621734857559204,
+      "rewards/check_gptzero_func": 0.7857142686843872,
+      "rewards/check_perplexity_diff_func": 0.75,
+      "rewards/check_winston_local_func": 0.8678879141807556,
+      "step": 269
+    },
+    {
+      "clip_ratio": 0.006006812676787376,
+      "epoch": 0.3776223776223776,
+      "grad_norm": 0.15420322253654764,
+      "kl": 0.8671875,
+      "learning_rate": 9.440559440559441e-05,
+      "loss": 0.0283,
+      "step": 270
+    },
+    {
+      "clip_ratio": 0.002397142816334963,
+      "completion_length": 102.41072082519531,
+      "epoch": 0.37902097902097903,
+      "grad_norm": 0.22567114655173642,
+      "kl": 0.609375,
+      "learning_rate": 9.475524475524477e-05,
+      "loss": 0.0231,
+      "num_tokens": 2765203.0,
+      "reward": 2.649076223373413,
+      "reward_std": 0.47521141171455383,
+      "rewards/check_gptzero_func": 0.875,
+      "rewards/check_perplexity_diff_func": 1.0,
+      "rewards/check_winston_local_func": 0.7740762829780579,
+      "step": 271
+    },
+    {
+      "clip_ratio": 0.007639365270733833,
+      "epoch": 0.3804195804195804,
+      "grad_norm": 0.18662930190408594,
+      "kl": 0.578125,
+      "learning_rate": 9.510489510489511e-05,
+      "loss": 0.0165,
+      "step": 272
+    },
+    {
+      "clip_ratio": 0.0010056205792352557,
+      "completion_length": 115.28572082519531,
+      "epoch": 0.38181818181818183,
+      "grad_norm": 0.14954423121719349,
+      "kl": 0.625,
+      "learning_rate": 9.545454545454546e-05,
+      "loss": 0.0237,
+      "num_tokens": 2785985.0,
+      "reward": 2.3541975021362305,
+      "reward_std": 0.5445494651794434,
+      "rewards/check_gptzero_func": 0.8392857313156128,
+      "rewards/check_perplexity_diff_func": 0.75,
+      "rewards/check_winston_local_func": 0.7649118304252625,
+      "step": 273
+    },
+    {
+      "clip_ratio": 0.003297280054539442,
+      "epoch": 0.3832167832167832,
+      "grad_norm": 0.11713806873969078,
+      "kl": 0.640625,
+      "learning_rate": 9.580419580419581e-05,
+      "loss": 0.0206,
+      "step": 274
+    },
+    {
+      "clip_ratio": 0.0016172031173482537,
+      "completion_length": 128.73214721679688,
+      "epoch": 0.38461538461538464,
+      "grad_norm": 0.155009547701029,
+      "kl": 0.51171875,
+      "learning_rate": 9.615384615384617e-05,
+      "loss": 0.0191,
+      "num_tokens": 2808414.0,
+      "reward": 2.376694679260254,
+      "reward_std": 0.33215656876564026,
+      "rewards/check_gptzero_func": 0.8392857313156128,
+      "rewards/check_perplexity_diff_func": 0.8571428656578064,
+      "rewards/check_winston_local_func": 0.6802659630775452,
+      "step": 275
+    },
+    {
+      "clip_ratio": 0.004294655751436949,
+      "epoch": 0.386013986013986,
+      "grad_norm": 0.11409067239204185,
+      "kl": 0.515625,
+      "learning_rate": 9.65034965034965e-05,
+      "loss": 0.0155,
+      "step": 276
+    },
+    {
+      "clip_ratio": 0.002925862092524767,
+      "completion_length": 116.41072082519531,
+      "epoch": 0.38741258741258744,
+      "grad_norm": 0.2764503095198425,
+      "kl": 0.578125,
+      "learning_rate": 9.685314685314686e-05,
+      "loss": 0.01,
+      "num_tokens": 2829285.0,
+      "reward": 2.3352210521698,
+      "reward_std": 0.7207943201065063,
+      "rewards/check_gptzero_func": 0.6607142686843872,
+      "rewards/check_perplexity_diff_func": 0.9642857313156128,
+      "rewards/check_winston_local_func": 0.7102211713790894,
+      "step": 277
+    },
+    {
+      "clip_ratio": 0.005557883996516466,
+      "epoch": 0.3888111888111888,
+      "grad_norm": 0.2140511595370052,
+      "kl": 0.5859375,
+      "learning_rate": 9.72027972027972e-05,
+      "loss": -0.0004,
+      "step": 278
+    },
+    {
+      "clip_ratio": 0.0029403052758425474,
+      "completion_length": 114.25000762939453,
+      "epoch": 0.3902097902097902,
+      "grad_norm": 0.21363596109825803,
+      "kl": 0.63671875,
+      "learning_rate": 9.755244755244755e-05,
+      "loss": 0.0259,
+      "num_tokens": 2849989.0,
+      "reward": 2.6570937633514404,
+      "reward_std": 0.5007120370864868,
+      "rewards/check_gptzero_func": 0.75,
+      "rewards/check_perplexity_diff_func": 1.2142857313156128,
+      "rewards/check_winston_local_func": 0.6928080320358276,
+      "step": 279
+    },
+    {
+      "clip_ratio": 0.00384224159643054,
+      "epoch": 0.3916083916083916,
+      "grad_norm": 0.15102706915256958,
+      "kl": 0.6875,
+      "learning_rate": 9.790209790209791e-05,
+      "loss": 0.0185,
+      "step": 280
+    },
+    {
+      "clip_ratio": 0.0027283646631985903,
+      "completion_length": 90.25000762939453,
+      "epoch": 0.393006993006993,
+      "grad_norm": 0.3585134615134873,
+      "kl": 0.73046875,
+      "learning_rate": 9.825174825174826e-05,
+      "loss": -0.006,
+      "num_tokens": 2867567.0,
+      "reward": 2.4528698921203613,
+      "reward_std": 0.6383817791938782,
+      "rewards/check_gptzero_func": 0.6071428656578064,
+      "rewards/check_perplexity_diff_func": 1.1428571939468384,
+      "rewards/check_winston_local_func": 0.7028695940971375,
+      "step": 281
+    },
+    {
+      "clip_ratio": 0.016295742243528366,
+      "epoch": 0.3944055944055944,
+      "grad_norm": 0.2358766583499307,
+      "kl": 0.73046875,
+      "learning_rate": 9.86013986013986e-05,
+      "loss": -0.0151,
+      "step": 282
+    },
+    {
+      "clip_ratio": 0.0025735762901604176,
+      "completion_length": 132.125,
+      "epoch": 0.3958041958041958,
+      "grad_norm": 0.12427534341566092,
+      "kl": 0.515625,
+      "learning_rate": 9.895104895104895e-05,
+      "loss": 0.006,
+      "num_tokens": 2890954.0,
+      "reward": 2.172079086303711,
+      "reward_std": 0.31105130910873413,
+      "rewards/check_gptzero_func": 0.6607142686843872,
+      "rewards/check_perplexity_diff_func": 0.7857142686843872,
+      "rewards/check_winston_local_func": 0.7256504893302917,
+      "step": 283
+    },
+    {
+      "clip_ratio": 0.008568070828914642,
+      "epoch": 0.3972027972027972,
+      "grad_norm": 0.10751149742872082,
+      "kl": 0.5234375,
+      "learning_rate": 9.930069930069931e-05,
+      "loss": 0.0038,
+      "step": 284
+    },
+    {
+      "clip_ratio": 0.0013303400482982397,
+      "completion_length": 115.50000762939453,
+      "epoch": 0.3986013986013986,
+      "grad_norm": 0.11155145645226237,
+      "kl": 0.427734375,
+      "learning_rate": 9.965034965034964e-05,
+      "loss": 0.0136,
+      "num_tokens": 2912320.0,
+      "reward": 2.499699354171753,
+      "reward_std": 0.22870054841041565,
+      "rewards/check_gptzero_func": 0.6071428656578064,
+      "rewards/check_perplexity_diff_func": 1.3214285373687744,
+      "rewards/check_winston_local_func": 0.5711276531219482,
+      "step": 285
+    },
+    {
+      "clip_ratio": 0.004690885543823242,
+      "epoch": 0.4,
+      "grad_norm": 0.08246302090980155,
+      "kl": 0.443359375,
+      "learning_rate": 0.0001,
+      "loss": 0.0117,
+      "step": 286
+    },
+    {
+      "clip_ratio": 0.0013457606546580791,
+      "completion_length": 153.60714721679688,
+      "epoch": 0.4013986013986014,
+      "grad_norm": 0.0959414180998777,
+      "kl": 0.458984375,
+      "learning_rate": 9.999996275889018e-05,
+      "loss": -0.0037,
+      "num_tokens": 2937286.0,
+      "reward": 2.60290789604187,
+      "reward_std": 0.2643657624721527,
+      "rewards/check_gptzero_func": 0.625,
+      "rewards/check_perplexity_diff_func": 1.3214285373687744,
+      "rewards/check_winston_local_func": 0.6564791798591614,
+      "step": 287
+    },
+    {
+      "clip_ratio": 0.0025680752005428076,
+      "epoch": 0.4027972027972028,
+      "grad_norm": 0.08295696905412306,
+      "kl": 0.453125,
+      "learning_rate": 9.999985103561615e-05,
+      "loss": -0.0053,
+      "step": 288
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 106.39286041259766,
+      "epoch": 0.4041958041958042,
+      "grad_norm": 0.12524695486629228,
+      "kl": 0.3671875,
+      "learning_rate": 9.999966483034437e-05,
+      "loss": -0.0031,
+      "num_tokens": 2957572.0,
+      "reward": 2.607825994491577,
+      "reward_std": 0.2113000899553299,
+      "rewards/check_gptzero_func": 0.5535714030265808,
+      "rewards/check_perplexity_diff_func": 1.4642857313156128,
+      "rewards/check_winston_local_func": 0.5899689793586731,
+      "step": 289
+    },
+    {
+      "clip_ratio": 0.0015743181575089693,
+      "epoch": 0.40559440559440557,
+      "grad_norm": 0.08958540244405325,
+      "kl": 0.36328125,
+      "learning_rate": 9.999940414335222e-05,
+      "loss": -0.0046,
+      "step": 290
+    },
+    {
+      "clip_ratio": 0.0017556955572217703,
+      "completion_length": 96.0714340209961,
+      "epoch": 0.406993006993007,
+      "grad_norm": 0.19185238369967877,
+      "kl": 2.109375,
+      "learning_rate": 9.999906897502803e-05,
+      "loss": 0.008,
+      "num_tokens": 2976076.0,
+      "reward": 2.6148557662963867,
+      "reward_std": 0.24822747707366943,
+      "rewards/check_gptzero_func": 0.5178571343421936,
+      "rewards/check_perplexity_diff_func": 1.4285714626312256,
+      "rewards/check_winston_local_func": 0.6684269309043884,
+      "step": 291
+    },
+    {
+      "clip_ratio": 0.00337741756811738,
+      "epoch": 0.4083916083916084,
+      "grad_norm": 0.08769186584743054,
+      "kl": 0.671875,
+      "learning_rate": 9.999865932587107e-05,
+      "loss": 0.0046,
+      "step": 292
+    },
+    {
+      "clip_ratio": 0.0012849332997575402,
+      "completion_length": 91.87500762939453,
+      "epoch": 0.4097902097902098,
+      "grad_norm": 0.13036265265723118,
+      "kl": 0.4453125,
+      "learning_rate": 9.999817519649158e-05,
+      "loss": 0.0127,
+      "num_tokens": 2994773.0,
+      "reward": 2.552300214767456,
+      "reward_std": 0.14500018954277039,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 1.5,
+      "rewards/check_winston_local_func": 0.6237286329269409,
+      "step": 293
+    },
+    {
+      "clip_ratio": 0.0027297178748995066,
+      "epoch": 0.4111888111888112,
+      "grad_norm": 0.0976002240930887,
+      "kl": 0.439453125,
+      "learning_rate": 9.99976165876107e-05,
+      "loss": 0.0101,
+      "step": 294
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 113.35714721679688,
+      "epoch": 0.4125874125874126,
+      "grad_norm": 0.02609480806956132,
+      "kl": 0.1591796875,
+      "learning_rate": 9.999698350006063e-05,
+      "loss": 0.0012,
+      "num_tokens": 3015993.0,
+      "reward": 2.874926805496216,
+      "reward_std": 0.0874238833785057,
+      "rewards/check_gptzero_func": 0.4821428656578064,
+      "rewards/check_perplexity_diff_func": 1.8928571939468384,
+      "rewards/check_winston_local_func": 0.4999266266822815,
+      "step": 295
+    },
+    {
+      "clip_ratio": 9.294544724980369e-05,
+      "epoch": 0.413986013986014,
+      "grad_norm": 0.02050573548257188,
+      "kl": 0.16015625,
+      "learning_rate": 9.999627593478442e-05,
+      "loss": 0.0011,
+      "step": 296
+    },
+    {
+      "clip_ratio": 0.00012664640962611884,
+      "completion_length": 98.5714340209961,
+      "epoch": 0.4153846153846154,
+      "grad_norm": 0.01680753751461254,
+      "kl": 0.201171875,
+      "learning_rate": 9.999549389283606e-05,
+      "loss": -0.0003,
+      "num_tokens": 3035155.0,
+      "reward": 3.0224545001983643,
+      "reward_std": 0.016296973451972008,
+      "rewards/check_gptzero_func": 0.4107142984867096,
+      "rewards/check_perplexity_diff_func": 1.9642857313156128,
+      "rewards/check_winston_local_func": 0.647454559803009,
+      "step": 297
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.4167832167832168,
+      "grad_norm": 0.012884367061070623,
+      "kl": 0.203125,
+      "learning_rate": 9.999463737538053e-05,
+      "loss": -0.0003,
+      "step": 298
+    },
+    {
+      "clip_ratio": 0.0003541912592481822,
+      "completion_length": 95.76786041259766,
+      "epoch": 0.41818181818181815,
+      "grad_norm": 0.04526732622569909,
+      "kl": 0.291015625,
+      "learning_rate": 9.999370638369377e-05,
+      "loss": 0.0002,
+      "num_tokens": 3054174.0,
+      "reward": 2.3303959369659424,
+      "reward_std": 0.08971218019723892,
+      "rewards/check_gptzero_func": 0.3035714328289032,
+      "rewards/check_perplexity_diff_func": 1.6071428060531616,
+      "rewards/check_winston_local_func": 0.4196813404560089,
+      "step": 299
+    },
+    {
+      "clip_ratio": 0.00011806376278400421,
+      "epoch": 0.4195804195804196,
+      "grad_norm": 0.03862986955837794,
+      "kl": 0.296875,
+      "learning_rate": 9.999270091916257e-05,
+      "loss": -0.0002,
+      "step": 300
+    },
+    {
+      "clip_ratio": 0.0012040403671562672,
+      "completion_length": 101.73214721679688,
+      "epoch": 0.42097902097902096,
+      "grad_norm": 0.12616717395678975,
+      "kl": 0.35546875,
+      "learning_rate": 9.999162098328474e-05,
+      "loss": 0.0033,
+      "num_tokens": 3073401.0,
+      "reward": 2.8153316974639893,
+      "reward_std": 0.18918544054031372,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 1.7142857313156128,
+      "rewards/check_winston_local_func": 0.6724745631217957,
+      "step": 301
+    },
+    {
+      "clip_ratio": 0.0013847867958247662,
+      "epoch": 0.4223776223776224,
+      "grad_norm": 0.06929991057149441,
+      "kl": 0.412109375,
+      "learning_rate": 9.999046657766903e-05,
+      "loss": 0.0012,
+      "step": 302
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 81.3214340209961,
+      "epoch": 0.42377622377622376,
+      "grad_norm": 0.06730614630696258,
+      "kl": 0.2099609375,
+      "learning_rate": 9.998923770403505e-05,
+      "loss": -0.0019,
+      "num_tokens": 3090625.0,
+      "reward": 2.802220106124878,
+      "reward_std": 0.03781326860189438,
+      "rewards/check_gptzero_func": 0.375,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.42722010612487793,
+      "step": 303
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.4251748251748252,
+      "grad_norm": 0.022175315902565728,
+      "kl": 0.2109375,
+      "learning_rate": 9.99879343642134e-05,
+      "loss": -0.0024,
+      "step": 304
+    },
+    {
+      "clip_ratio": 0.00013683634460903704,
+      "completion_length": 105.53572082519531,
+      "epoch": 0.42657342657342656,
+      "grad_norm": 0.04867704636152166,
+      "kl": 0.173828125,
+      "learning_rate": 9.998655656014561e-05,
+      "loss": 0.0117,
+      "num_tokens": 3110877.0,
+      "reward": 2.6083197593688965,
+      "reward_std": 0.13502858579158783,
+      "rewards/check_gptzero_func": 0.3571428656578064,
+      "rewards/check_perplexity_diff_func": 1.8928571939468384,
+      "rewards/check_winston_local_func": 0.35831958055496216,
+      "step": 305
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.427972027972028,
+      "grad_norm": 0.043987709695075865,
+      "kl": 0.1728515625,
+      "learning_rate": 9.99851042938841e-05,
+      "loss": 0.0113,
+      "step": 306
+    },
+    {
+      "clip_ratio": 0.0004155792994424701,
+      "completion_length": 124.03572082519531,
+      "epoch": 0.42937062937062936,
+      "grad_norm": 0.03684760048760383,
+      "kl": 0.2080078125,
+      "learning_rate": 9.998357756759222e-05,
+      "loss": -0.0258,
+      "num_tokens": 3134097.0,
+      "reward": 2.7439539432525635,
+      "reward_std": 0.11075940728187561,
+      "rewards/check_gptzero_func": 0.4642857015132904,
+      "rewards/check_perplexity_diff_func": 1.7857142686843872,
+      "rewards/check_winston_local_func": 0.4939536452293396,
+      "step": 307
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.4307692307692308,
+      "grad_norm": 0.013648247572273129,
+      "kl": 0.20703125,
+      "learning_rate": 9.998197638354428e-05,
+      "loss": -0.026,
+      "step": 308
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 89.12500762939453,
+      "epoch": 0.43216783216783217,
+      "grad_norm": 0.02465210474674629,
+      "kl": 0.443359375,
+      "learning_rate": 9.998030074412545e-05,
+      "loss": 0.0014,
+      "num_tokens": 3152358.0,
+      "reward": 2.420307159423828,
+      "reward_std": 0.11704181134700775,
+      "rewards/check_gptzero_func": 0.2321428507566452,
+      "rewards/check_perplexity_diff_func": 1.75,
+      "rewards/check_winston_local_func": 0.4381641745567322,
+      "step": 309
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.43356643356643354,
+      "grad_norm": 0.021184461362474653,
+      "kl": 0.48828125,
+      "learning_rate": 9.997855065183184e-05,
+      "loss": 0.0014,
+      "step": 310
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 115.73214721679688,
+      "epoch": 0.43496503496503497,
+      "grad_norm": 0.0049134080725686995,
+      "kl": 0.162109375,
+      "learning_rate": 9.997672610927047e-05,
+      "loss": -0.0047,
+      "num_tokens": 3173997.0,
+      "reward": 2.61786150932312,
+      "reward_std": 0.013112460263073444,
+      "rewards/check_gptzero_func": 0.375,
+      "rewards/check_perplexity_diff_func": 1.8214285373687744,
+      "rewards/check_winston_local_func": 0.4214327931404114,
+      "step": 311
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.43636363636363634,
+      "grad_norm": 0.00595537475288129,
+      "kl": 0.173828125,
+      "learning_rate": 9.997482711915927e-05,
+      "loss": -0.0047,
+      "step": 312
+    },
+    {
+      "clip_ratio": 0.0001360544265480712,
+      "completion_length": 90.42857360839844,
+      "epoch": 0.43776223776223777,
+      "grad_norm": 0.007409687643468917,
+      "kl": 0.2275390625,
+      "learning_rate": 9.997285368432703e-05,
+      "loss": 0.0006,
+      "num_tokens": 3192417.0,
+      "reward": 2.682650566101074,
+      "reward_std": 0.02417343109846115,
+      "rewards/check_gptzero_func": 0.375,
+      "rewards/check_perplexity_diff_func": 1.8214285373687744,
+      "rewards/check_winston_local_func": 0.4862218499183655,
+      "step": 313
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.43916083916083914,
+      "grad_norm": 0.007129921917131849,
+      "kl": 0.2275390625,
+      "learning_rate": 9.997080580771349e-05,
+      "loss": 0.0006,
+      "step": 314
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 83.14286041259766,
+      "epoch": 0.4405594405594406,
+      "grad_norm": 1.3606081670938062e-05,
+      "kl": 0.2255859375,
+      "learning_rate": 9.996868349236927e-05,
+      "loss": 0.0002,
+      "num_tokens": 3209875.0,
+      "reward": 2.671295404434204,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3571428656578064,
+      "rewards/check_perplexity_diff_func": 1.8571428060531616,
+      "rewards/check_winston_local_func": 0.45700937509536743,
+      "step": 315
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.44195804195804195,
+      "grad_norm": 1.3537787639105168e-05,
+      "kl": 0.2255859375,
+      "learning_rate": 9.996648674145583e-05,
+      "loss": 0.0002,
+      "step": 316
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 126.21429443359375,
+      "epoch": 0.4433566433566434,
+      "grad_norm": 0.00991208286904361,
+      "kl": 0.15625,
+      "learning_rate": 9.996421555824558e-05,
+      "loss": 0.0103,
+      "num_tokens": 3232963.0,
+      "reward": 2.6654393672943115,
+      "reward_std": 0.0725974440574646,
+      "rewards/check_gptzero_func": 0.5357142686843872,
+      "rewards/check_perplexity_diff_func": 1.6071428060531616,
+      "rewards/check_winston_local_func": 0.5225821733474731,
+      "step": 317
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.44475524475524475,
+      "grad_norm": 0.010677817125212348,
+      "kl": 0.154296875,
+      "learning_rate": 9.996186994612176e-05,
+      "loss": 0.0102,
+      "step": 318
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 103.4464340209961,
+      "epoch": 0.4461538461538462,
+      "grad_norm": 5.275581365835224e-05,
+      "kl": 0.16796875,
+      "learning_rate": 9.995944990857849e-05,
+      "loss": 0.0002,
+      "num_tokens": 3252898.0,
+      "reward": 3.0420854091644287,
+      "reward_std": 0.00039901022682897747,
+      "rewards/check_gptzero_func": 0.5714285969734192,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.5420854687690735,
+      "step": 319
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.44755244755244755,
+      "grad_norm": 6.10762506510971e-05,
+      "kl": 0.16796875,
+      "learning_rate": 9.995695544922077e-05,
+      "loss": 0.0002,
+      "step": 320
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 102.28572082519531,
+      "epoch": 0.4489510489510489,
+      "grad_norm": 0.010696195320132886,
+      "kl": 0.1669921875,
+      "learning_rate": 9.995438657176448e-05,
+      "loss": -0.0177,
+      "num_tokens": 3272574.0,
+      "reward": 2.7592146396636963,
+      "reward_std": 0.03917063772678375,
+      "rewards/check_gptzero_func": 0.3571428656578064,
+      "rewards/check_perplexity_diff_func": 1.9642857313156128,
+      "rewards/check_winston_local_func": 0.437785804271698,
+      "step": 321
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.45034965034965035,
+      "grad_norm": 0.01083841197384593,
+      "kl": 0.1669921875,
+      "learning_rate": 9.995174328003631e-05,
+      "loss": -0.0177,
+      "step": 322
+    },
+    {
+      "clip_ratio": 8.473140769638121e-05,
+      "completion_length": 117.41072082519531,
+      "epoch": 0.45174825174825173,
+      "grad_norm": 0.004522641361347185,
+      "kl": 0.1640625,
+      "learning_rate": 9.994902557797382e-05,
+      "loss": -0.0002,
+      "num_tokens": 3294025.0,
+      "reward": 2.92221736907959,
+      "reward_std": 0.025547320023179054,
+      "rewards/check_gptzero_func": 0.5535714030265808,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.36864587664604187,
+      "step": 323
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.45314685314685316,
+      "grad_norm": 0.004737684573095642,
+      "kl": 0.1650390625,
+      "learning_rate": 9.994623346962544e-05,
+      "loss": -0.0002,
+      "step": 324
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 95.26786041259766,
+      "epoch": 0.45454545454545453,
+      "grad_norm": 3.372088632491828e-06,
+      "kl": 0.1796875,
+      "learning_rate": 9.99433669591504e-05,
+      "loss": 0.0002,
+      "num_tokens": 3312902.0,
+      "reward": 2.595499038696289,
+      "reward_std": 4.388691013446078e-05,
+      "rewards/check_gptzero_func": 0.25,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.34549909830093384,
+      "step": 325
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.45594405594405596,
+      "grad_norm": 3.349405120445447e-06,
+      "kl": 0.1796875,
+      "learning_rate": 9.994042605081879e-05,
+      "loss": 0.0002,
+      "step": 326
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 97.80357360839844,
+      "epoch": 0.45734265734265733,
+      "grad_norm": 0.010175912326793437,
+      "kl": 0.177734375,
+      "learning_rate": 9.993741074901153e-05,
+      "loss": 0.0003,
+      "num_tokens": 3331815.0,
+      "reward": 2.496483325958252,
+      "reward_std": 0.04960284009575844,
+      "rewards/check_gptzero_func": 0.2142857164144516,
+      "rewards/check_perplexity_diff_func": 1.9642857313156128,
+      "rewards/check_winston_local_func": 0.31791168451309204,
+      "step": 327
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.45874125874125876,
+      "grad_norm": 0.01017493586702021,
+      "kl": 0.177734375,
+      "learning_rate": 9.993432105822034e-05,
+      "loss": 0.0003,
+      "step": 328
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 96.30357360839844,
+      "epoch": 0.46013986013986014,
+      "grad_norm": 0.0165654292436505,
+      "kl": 0.1572265625,
+      "learning_rate": 9.993115698304774e-05,
+      "loss": 0.0003,
+      "num_tokens": 3351210.0,
+      "reward": 2.8074374198913574,
+      "reward_std": 0.07505974173545837,
+      "rewards/check_gptzero_func": 0.5178571343421936,
+      "rewards/check_perplexity_diff_func": 1.8571428060531616,
+      "rewards/check_winston_local_func": 0.4324372708797455,
+      "step": 329
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.46153846153846156,
+      "grad_norm": 0.014211382809754432,
+      "kl": 0.16015625,
+      "learning_rate": 9.992791852820709e-05,
+      "loss": 0.0003,
+      "step": 330
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 86.1964340209961,
+      "epoch": 0.46293706293706294,
+      "grad_norm": 0.0580909978469744,
+      "kl": 0.2119140625,
+      "learning_rate": 9.992460569852256e-05,
+      "loss": 0.0013,
+      "num_tokens": 3369025.0,
+      "reward": 2.986760377883911,
+      "reward_std": 0.10425443947315216,
+      "rewards/check_gptzero_func": 0.4642857015132904,
+      "rewards/check_perplexity_diff_func": 1.8571428060531616,
+      "rewards/check_winston_local_func": 0.6653318405151367,
+      "step": 331
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.4643356643356643,
+      "grad_norm": 0.0012843504186470114,
+      "kl": 0.2109375,
+      "learning_rate": 9.992121849892904e-05,
+      "loss": 0.0012,
+      "step": 332
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 113.42857360839844,
+      "epoch": 0.46573426573426574,
+      "grad_norm": 0.03054414776989089,
+      "kl": 0.1494140625,
+      "learning_rate": 9.99177569344723e-05,
+      "loss": -0.0,
+      "num_tokens": 3390093.0,
+      "reward": 2.5890934467315674,
+      "reward_std": 0.1023683100938797,
+      "rewards/check_gptzero_func": 0.25,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.41052165627479553,
+      "step": 333
+    },
+    {
+      "clip_ratio": 0.00022784233442507684,
+      "epoch": 0.4671328671328671,
+      "grad_norm": 0.006141110067038759,
+      "kl": 0.150390625,
+      "learning_rate": 9.99142210103088e-05,
+      "loss": -0.0001,
+      "step": 334
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 109.00000762939453,
+      "epoch": 0.46853146853146854,
+      "grad_norm": 0.011702000814979486,
+      "kl": 0.158203125,
+      "learning_rate": 9.991061073170585e-05,
+      "loss": 0.0001,
+      "num_tokens": 3410511.0,
+      "reward": 2.601181745529175,
+      "reward_std": 0.04859553650021553,
+      "rewards/check_gptzero_func": 0.2857142984867096,
+      "rewards/check_perplexity_diff_func": 1.8928571939468384,
+      "rewards/check_winston_local_func": 0.42261001467704773,
+      "step": 335
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.4699300699300699,
+      "grad_norm": 0.004380253375033637,
+      "kl": 0.158203125,
+      "learning_rate": 9.990692610404145e-05,
+      "loss": 0.0001,
+      "step": 336
+    },
+    {
+      "clip_ratio": 0.00015908369095996022,
+      "completion_length": 124.37500762939453,
+      "epoch": 0.47132867132867134,
+      "grad_norm": 0.011635533606345248,
+      "kl": 0.1318359375,
+      "learning_rate": 9.99031671328044e-05,
+      "loss": 0.0002,
+      "num_tokens": 3433250.0,
+      "reward": 2.719287395477295,
+      "reward_std": 0.05448899790644646,
+      "rewards/check_gptzero_func": 0.5,
+      "rewards/check_perplexity_diff_func": 1.75,
+      "rewards/check_winston_local_func": 0.4692873954772949,
+      "step": 337
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.4727272727272727,
+      "grad_norm": 0.009393665715625162,
+      "kl": 0.1318359375,
+      "learning_rate": 9.989933382359422e-05,
+      "loss": 0.0001,
+      "step": 338
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 110.8214340209961,
+      "epoch": 0.47412587412587415,
+      "grad_norm": 6.954019260271828e-07,
+      "kl": 0.15625,
+      "learning_rate": 9.98954261821212e-05,
+      "loss": 0.0002,
+      "num_tokens": 3454202.0,
+      "reward": 2.5842461585998535,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 1.7857142686843872,
+      "rewards/check_winston_local_func": 0.40567442774772644,
+      "step": 339
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.4755244755244755,
+      "grad_norm": 6.417345751201852e-07,
+      "kl": 0.15625,
+      "learning_rate": 9.98914442142063e-05,
+      "loss": 0.0002,
+      "step": 340
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 76.55357360839844,
+      "epoch": 0.47692307692307695,
+      "grad_norm": 0.008322229771779063,
+      "kl": 0.263671875,
+      "learning_rate": 9.988738792578126e-05,
+      "loss": 0.0005,
+      "num_tokens": 3471041.0,
+      "reward": 2.916919231414795,
+      "reward_std": 0.05185602605342865,
+      "rewards/check_gptzero_func": 0.3214285671710968,
+      "rewards/check_perplexity_diff_func": 1.9642857313156128,
+      "rewards/check_winston_local_func": 0.631205141544342,
+      "step": 341
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.4783216783216783,
+      "grad_norm": 0.010116392524351476,
+      "kl": 0.263671875,
+      "learning_rate": 9.988325732288851e-05,
+      "loss": 0.0004,
+      "step": 342
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 122.21429443359375,
+      "epoch": 0.4797202797202797,
+      "grad_norm": 1.4137707959903356e-05,
+      "kl": 0.134765625,
+      "learning_rate": 9.987905241168117e-05,
+      "loss": 0.0001,
+      "num_tokens": 3493483.0,
+      "reward": 2.7157416343688965,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 1.8571428060531616,
+      "rewards/check_winston_local_func": 0.46574172377586365,
+      "step": 343
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.4811188811188811,
+      "grad_norm": 1.3729789968238485e-05,
+      "kl": 0.134765625,
+      "learning_rate": 9.987477319842307e-05,
+      "loss": 0.0001,
+      "step": 344
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 93.53572082519531,
+      "epoch": 0.4825174825174825,
+      "grad_norm": 7.468121696306164e-07,
+      "kl": 0.1845703125,
+      "learning_rate": 9.987041968948869e-05,
+      "loss": 0.0002,
+      "num_tokens": 3512009.0,
+      "reward": 2.5505218505859375,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.1785714328289032,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.44337886571884155,
+      "step": 345
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.48391608391608393,
+      "grad_norm": 7.137709602226676e-07,
+      "kl": 0.1845703125,
+      "learning_rate": 9.986599189136325e-05,
+      "loss": 0.0002,
+      "step": 346
+    },
+    {
+      "clip_ratio": 0.00031959093757905066,
+      "completion_length": 102.58928680419922,
+      "epoch": 0.4853146853146853,
+      "grad_norm": 0.04707770778424934,
+      "kl": 0.1982421875,
+      "learning_rate": 9.986148981064258e-05,
+      "loss": 0.0004,
+      "num_tokens": 3531574.0,
+      "reward": 2.7341363430023193,
+      "reward_std": 0.0231277234852314,
+      "rewards/check_gptzero_func": 0.4107142984867096,
+      "rewards/check_perplexity_diff_func": 1.8928571939468384,
+      "rewards/check_winston_local_func": 0.430564820766449,
+      "step": 347
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.48671328671328673,
+      "grad_norm": 0.020919799034125892,
+      "kl": 0.2060546875,
+      "learning_rate": 9.985691345403316e-05,
+      "loss": 0.0002,
+      "step": 348
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 127.67857360839844,
+      "epoch": 0.4881118881118881,
+      "grad_norm": 6.517786151476589e-05,
+      "kl": 0.1328125,
+      "learning_rate": 9.985226282835216e-05,
+      "loss": 0.0001,
+      "num_tokens": 3554440.0,
+      "reward": 2.926541566848755,
+      "reward_std": 0.0005768488626927137,
+      "rewards/check_gptzero_func": 0.5357142686843872,
+      "rewards/check_perplexity_diff_func": 1.8571428060531616,
+      "rewards/check_winston_local_func": 0.5336844325065613,
+      "step": 349
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.48951048951048953,
+      "grad_norm": 6.370254129427888e-05,
+      "kl": 0.1328125,
+      "learning_rate": 9.984753794052735e-05,
+      "loss": 0.0001,
+      "step": 350
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 98.75000762939453,
+      "epoch": 0.4909090909090909,
+      "grad_norm": 2.085540464425244e-05,
+      "kl": 0.205078125,
+      "learning_rate": 9.984273879759713e-05,
+      "loss": 0.0002,
+      "num_tokens": 3573976.0,
+      "reward": 2.8204286098480225,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.4989997446537018,
+      "step": 351
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.49230769230769234,
+      "grad_norm": 2.009578048518353e-05,
+      "kl": 0.205078125,
+      "learning_rate": 9.983786540671051e-05,
+      "loss": 0.0002,
+      "step": 352
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 85.21428680419922,
+      "epoch": 0.4937062937062937,
+      "grad_norm": 2.1499404490472988e-06,
+      "kl": 0.1826171875,
+      "learning_rate": 9.983291777512711e-05,
+      "loss": 0.0002,
+      "num_tokens": 3591644.0,
+      "reward": 2.8764142990112305,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.44784265756607056,
+      "step": 353
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.4951048951048951,
+      "grad_norm": 2.160124410750654e-06,
+      "kl": 0.1826171875,
+      "learning_rate": 9.982789591021715e-05,
+      "loss": 0.0002,
+      "step": 354
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 112.00000762939453,
+      "epoch": 0.4965034965034965,
+      "grad_norm": 0.00014814082463704463,
+      "kl": 0.1669921875,
+      "learning_rate": 9.982279981946143e-05,
+      "loss": 0.0002,
+      "num_tokens": 3612596.0,
+      "reward": 2.8398191928863525,
+      "reward_std": 0.0006443771999329329,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.5183902382850647,
+      "step": 355
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.4979020979020979,
+      "grad_norm": 0.00014606763289714295,
+      "kl": 0.1669921875,
+      "learning_rate": 9.98176295104513e-05,
+      "loss": 0.0002,
+      "step": 356
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 88.50000762939453,
+      "epoch": 0.4993006993006993,
+      "grad_norm": 3.33550016470454e-07,
+      "kl": 0.1669921875,
+      "learning_rate": 9.98123849908887e-05,
+      "loss": 0.0002,
+      "num_tokens": 3630862.0,
+      "reward": 2.655492067337036,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3214285671710968,
+      "rewards/check_perplexity_diff_func": 1.8571428060531616,
+      "rewards/check_winston_local_func": 0.47692039608955383,
+      "step": 357
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5006993006993007,
+      "grad_norm": 3.3462192763241657e-07,
+      "kl": 0.1669921875,
+      "learning_rate": 9.980706626858607e-05,
+      "loss": 0.0002,
+      "step": 358
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 126.39286041259766,
+      "epoch": 0.5020979020979021,
+      "grad_norm": 3.4860786526471938e-06,
+      "kl": 0.1416015625,
+      "learning_rate": 9.980167335146642e-05,
+      "loss": 0.0001,
+      "num_tokens": 3653738.0,
+      "reward": 2.765826463699341,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.5,
+      "rewards/check_perplexity_diff_func": 1.8571428060531616,
+      "rewards/check_winston_local_func": 0.4086834490299225,
+      "step": 359
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5034965034965035,
+      "grad_norm": 3.71619859471226e-06,
+      "kl": 0.1416015625,
+      "learning_rate": 9.979620624756329e-05,
+      "loss": 0.0001,
+      "step": 360
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 129.92857360839844,
+      "epoch": 0.5048951048951049,
+      "grad_norm": 8.195642286342138e-07,
+      "kl": 0.1376953125,
+      "learning_rate": 9.979066496502074e-05,
+      "loss": 0.0001,
+      "num_tokens": 3677274.0,
+      "reward": 2.8696835041046143,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.5357142686843872,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.4053979218006134,
+      "step": 361
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5062937062937063,
+      "grad_norm": 8.760444459307852e-07,
+      "kl": 0.1376953125,
+      "learning_rate": 9.978504951209327e-05,
+      "loss": 0.0001,
+      "step": 362
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 118.0714340209961,
+      "epoch": 0.5076923076923077,
+      "grad_norm": 1.595860947799347e-06,
+      "kl": 0.162109375,
+      "learning_rate": 9.977935989714595e-05,
+      "loss": 0.0002,
+      "num_tokens": 3698896.0,
+      "reward": 2.6664552688598633,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.2857142984867096,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.45216941833496094,
+      "step": 363
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.509090909090909,
+      "grad_norm": 1.6260194368178069e-06,
+      "kl": 0.162109375,
+      "learning_rate": 9.977359612865423e-05,
+      "loss": 0.0002,
+      "step": 364
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 85.96428680419922,
+      "epoch": 0.5104895104895105,
+      "grad_norm": 4.357301744161908e-06,
+      "kl": 0.2138671875,
+      "learning_rate": 9.976775821520412e-05,
+      "loss": 0.0002,
+      "num_tokens": 3716536.0,
+      "reward": 2.8877956867218018,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.566366970539093,
+      "step": 365
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5118881118881119,
+      "grad_norm": 4.353902612581284e-06,
+      "kl": 0.2138671875,
+      "learning_rate": 9.976184616549203e-05,
+      "loss": 0.0002,
+      "step": 366
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 116.96429443359375,
+      "epoch": 0.5132867132867133,
+      "grad_norm": 1.5148419096170459e-06,
+      "kl": 0.146484375,
+      "learning_rate": 9.97558599883248e-05,
+      "loss": 0.0001,
+      "num_tokens": 3737950.0,
+      "reward": 2.687312602996826,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.36588403582572937,
+      "step": 367
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5146853146853146,
+      "grad_norm": 1.4470246412763422e-06,
+      "kl": 0.146484375,
+      "learning_rate": 9.97497996926197e-05,
+      "loss": 0.0001,
+      "step": 368
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 94.83928680419922,
+      "epoch": 0.5160839160839161,
+      "grad_norm": 0.0004647591690002515,
+      "kl": 0.2138671875,
+      "learning_rate": 9.974366528740441e-05,
+      "loss": 0.0002,
+      "num_tokens": 3756935.0,
+      "reward": 2.634523391723633,
+      "reward_std": 0.0005499019753187895,
+      "rewards/check_gptzero_func": 0.2857142984867096,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.4202377200126648,
+      "step": 369
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5174825174825175,
+      "grad_norm": 0.0004885464229234562,
+      "kl": 0.2138671875,
+      "learning_rate": 9.973745678181705e-05,
+      "loss": 0.0002,
+      "step": 370
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 100.17857360839844,
+      "epoch": 0.5188811188811189,
+      "grad_norm": 3.0418176986251483e-06,
+      "kl": 0.1640625,
+      "learning_rate": 9.973117418510605e-05,
+      "loss": 0.0002,
+      "num_tokens": 3776087.0,
+      "reward": 2.5216457843780518,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.2857142984867096,
+      "rewards/check_perplexity_diff_func": 1.8571428060531616,
+      "rewards/check_winston_local_func": 0.3787885308265686,
+      "step": 371
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5202797202797202,
+      "grad_norm": 3.0723635480561103e-06,
+      "kl": 0.1640625,
+      "learning_rate": 9.972481750663026e-05,
+      "loss": 0.0002,
+      "step": 372
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 89.66072082519531,
+      "epoch": 0.5216783216783217,
+      "grad_norm": 0.008590656277289149,
+      "kl": 0.40234375,
+      "learning_rate": 9.971838675585888e-05,
+      "loss": 0.0007,
+      "num_tokens": 3794550.0,
+      "reward": 2.5267839431762695,
+      "reward_std": 0.05061079189181328,
+      "rewards/check_gptzero_func": 0.3214285671710968,
+      "rewards/check_perplexity_diff_func": 1.8214285373687744,
+      "rewards/check_winston_local_func": 0.383926659822464,
+      "step": 373
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5230769230769231,
+      "grad_norm": 0.003770576536622864,
+      "kl": 0.4296875,
+      "learning_rate": 9.97118819423714e-05,
+      "loss": 0.0007,
+      "step": 374
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 80.17857360839844,
+      "epoch": 0.5244755244755245,
+      "grad_norm": 3.560813902520289e-07,
+      "kl": 0.2216796875,
+      "learning_rate": 9.970530307585773e-05,
+      "loss": 0.0002,
+      "num_tokens": 3811756.0,
+      "reward": 2.883481740951538,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.5263389348983765,
+      "step": 375
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5258741258741259,
+      "grad_norm": 3.5463275242856317e-07,
+      "kl": 0.2216796875,
+      "learning_rate": 9.9698650166118e-05,
+      "loss": 0.0002,
+      "step": 376
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 86.75000762939453,
+      "epoch": 0.5272727272727272,
+      "grad_norm": 4.776682547251814e-07,
+      "kl": 0.205078125,
+      "learning_rate": 9.969192322306271e-05,
+      "loss": 0.0002,
+      "num_tokens": 3829616.0,
+      "reward": 2.9529268741607666,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4642857015132904,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.5600695610046387,
+      "step": 377
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5286713286713287,
+      "grad_norm": 4.804658823710114e-07,
+      "kl": 0.205078125,
+      "learning_rate": 9.96851222567126e-05,
+      "loss": 0.0002,
+      "step": 378
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 117.89286041259766,
+      "epoch": 0.5300699300699301,
+      "grad_norm": 9.143102926453055e-06,
+      "kl": 0.158203125,
+      "learning_rate": 9.96782472771987e-05,
+      "loss": 0.0002,
+      "num_tokens": 3851170.0,
+      "reward": 2.7617037296295166,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.40456071496009827,
+      "step": 379
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5314685314685315,
+      "grad_norm": 9.41817239216292e-06,
+      "kl": 0.158203125,
+      "learning_rate": 9.967129829476228e-05,
+      "loss": 0.0002,
+      "step": 380
+    },
+    {
+      "clip_ratio": 0.000286861730273813,
+      "completion_length": 80.8214340209961,
+      "epoch": 0.5328671328671328,
+      "grad_norm": 0.03337399461775748,
+      "kl": 0.224609375,
+      "learning_rate": 9.966427531975483e-05,
+      "loss": -0.0402,
+      "num_tokens": 3868508.0,
+      "reward": 2.820319652557373,
+      "reward_std": 0.0946396142244339,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 1.8571428060531616,
+      "rewards/check_winston_local_func": 0.5346053838729858,
+      "step": 381
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5342657342657343,
+      "grad_norm": 0.00034351439093039966,
+      "kl": 0.224609375,
+      "learning_rate": 9.965717836263812e-05,
+      "loss": -0.0403,
+      "step": 382
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 95.03572082519531,
+      "epoch": 0.5356643356643357,
+      "grad_norm": 3.3945579906577137e-07,
+      "kl": 0.216796875,
+      "learning_rate": 9.965000743398408e-05,
+      "loss": 0.0002,
+      "num_tokens": 3887044.0,
+      "reward": 2.8529021739959717,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.4243304431438446,
+      "step": 383
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5370629370629371,
+      "grad_norm": 2.8390305863820515e-07,
+      "kl": 0.216796875,
+      "learning_rate": 9.964276254447484e-05,
+      "loss": 0.0002,
+      "step": 384
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 140.1428680419922,
+      "epoch": 0.5384615384615384,
+      "grad_norm": 8.521451122880038e-07,
+      "kl": 0.134765625,
+      "learning_rate": 9.96354437049027e-05,
+      "loss": 0.0001,
+      "num_tokens": 3911264.0,
+      "reward": 2.4770619869232178,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.2857142984867096,
+      "rewards/check_perplexity_diff_func": 1.8571428060531616,
+      "rewards/check_winston_local_func": 0.33420488238334656,
+      "step": 385
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5398601398601398,
+      "grad_norm": 6.753306754213034e-07,
+      "kl": 0.134765625,
+      "learning_rate": 9.962805092617016e-05,
+      "loss": 0.0001,
+      "step": 386
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 97.64286041259766,
+      "epoch": 0.5412587412587413,
+      "grad_norm": 0.08784199050747857,
+      "kl": 2.28125,
+      "learning_rate": 9.962058421928979e-05,
+      "loss": 0.0023,
+      "num_tokens": 3930478.0,
+      "reward": 2.897810459136963,
+      "reward_std": 0.002231778111308813,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.4692386984825134,
+      "step": 387
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5426573426573427,
+      "grad_norm": 0.03620083942117924,
+      "kl": 0.98828125,
+      "learning_rate": 9.961304359538437e-05,
+      "loss": 0.001,
+      "step": 388
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 102.21428680419922,
+      "epoch": 0.544055944055944,
+      "grad_norm": 0.025447754609912356,
+      "kl": 0.1806640625,
+      "learning_rate": 9.96054290656867e-05,
+      "loss": 0.0002,
+      "num_tokens": 3950362.0,
+      "reward": 2.8110532760620117,
+      "reward_std": 0.04895726963877678,
+      "rewards/check_gptzero_func": 0.5357142686843872,
+      "rewards/check_perplexity_diff_func": 1.8214285373687744,
+      "rewards/check_winston_local_func": 0.45391014218330383,
+      "step": 389
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5454545454545454,
+      "grad_norm": 5.1964811610508515e-05,
+      "kl": 0.1806640625,
+      "learning_rate": 9.959774064153977e-05,
+      "loss": 0.0001,
+      "step": 390
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 103.5714340209961,
+      "epoch": 0.5468531468531469,
+      "grad_norm": 3.816406536402076e-07,
+      "kl": 0.173828125,
+      "learning_rate": 9.958997833439657e-05,
+      "loss": 0.0002,
+      "num_tokens": 3970164.0,
+      "reward": 2.5441431999206543,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3214285671710968,
+      "rewards/check_perplexity_diff_func": 1.8571428060531616,
+      "rewards/check_winston_local_func": 0.3655718266963959,
+      "step": 391
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5482517482517483,
+      "grad_norm": 4.199010025491445e-07,
+      "kl": 0.173828125,
+      "learning_rate": 9.958214215582018e-05,
+      "loss": 0.0002,
+      "step": 392
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 126.8214340209961,
+      "epoch": 0.5496503496503496,
+      "grad_norm": 6.309179959293235e-06,
+      "kl": 0.1533203125,
+      "learning_rate": 9.957423211748374e-05,
+      "loss": 0.0002,
+      "num_tokens": 3992994.0,
+      "reward": 2.7432734966278076,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.42184481024742126,
+      "step": 393
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.551048951048951,
+      "grad_norm": 5.547540566387994e-06,
+      "kl": 0.1533203125,
+      "learning_rate": 9.956624823117036e-05,
+      "loss": 0.0002,
+      "step": 394
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 125.62500762939453,
+      "epoch": 0.5524475524475524,
+      "grad_norm": 0.007987942179452279,
+      "kl": 0.142578125,
+      "learning_rate": 9.955819050877321e-05,
+      "loss": 0.0001,
+      "num_tokens": 4015669.0,
+      "reward": 3.091567039489746,
+      "reward_std": 0.018998777493834496,
+      "rewards/check_gptzero_func": 0.5535714030265808,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.6094242334365845,
+      "step": 395
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5538461538461539,
+      "grad_norm": 0.0034758785486018436,
+      "kl": 0.142578125,
+      "learning_rate": 9.955005896229543e-05,
+      "loss": 0.0,
+      "step": 396
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 96.75000762939453,
+      "epoch": 0.5552447552447553,
+      "grad_norm": 0.011495641299277456,
+      "kl": 0.283203125,
+      "learning_rate": 9.954185360385013e-05,
+      "loss": 0.0006,
+      "num_tokens": 4035041.0,
+      "reward": 2.950004816055298,
+      "reward_std": 0.01496803853660822,
+      "rewards/check_gptzero_func": 0.4821428656578064,
+      "rewards/check_perplexity_diff_func": 1.8928571939468384,
+      "rewards/check_winston_local_func": 0.5750047564506531,
+      "step": 397
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5566433566433566,
+      "grad_norm": 0.010082500068390739,
+      "kl": 0.283203125,
+      "learning_rate": 9.953357444566039e-05,
+      "loss": 0.0006,
+      "step": 398
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 110.8214340209961,
+      "epoch": 0.558041958041958,
+      "grad_norm": 8.489376613416e-06,
+      "kl": 0.1787109375,
+      "learning_rate": 9.952522150005919e-05,
+      "loss": 0.0002,
+      "num_tokens": 4055855.0,
+      "reward": 2.913240909576416,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4642857015132904,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.5203836560249329,
+      "step": 399
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5594405594405595,
+      "grad_norm": 8.766485288799225e-06,
+      "kl": 0.1787109375,
+      "learning_rate": 9.951679477948947e-05,
+      "loss": 0.0002,
+      "step": 400
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 105.46428680419922,
+      "epoch": 0.5608391608391609,
+      "grad_norm": 8.652849000461684e-06,
+      "kl": 0.150390625,
+      "learning_rate": 9.9508294296504e-05,
+      "loss": 0.0002,
+      "num_tokens": 4076051.0,
+      "reward": 2.871354818344116,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.44278329610824585,
+      "step": 401
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5622377622377622,
+      "grad_norm": 8.18246821524729e-06,
+      "kl": 0.150390625,
+      "learning_rate": 9.949972006376556e-05,
+      "loss": 0.0002,
+      "step": 402
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 75.25,
+      "epoch": 0.5636363636363636,
+      "grad_norm": 7.351342876079361e-07,
+      "kl": 0.1953125,
+      "learning_rate": 9.949107209404665e-05,
+      "loss": 0.0002,
+      "num_tokens": 4092527.0,
+      "reward": 2.9322519302368164,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3571428656578064,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.5751089453697205,
+      "step": 403
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5650349650349651,
+      "grad_norm": 7.136344464618312e-07,
+      "kl": 0.1953125,
+      "learning_rate": 9.948235040022966e-05,
+      "loss": 0.0002,
+      "step": 404
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 108.08928680419922,
+      "epoch": 0.5664335664335665,
+      "grad_norm": 0.007956878335048736,
+      "kl": 0.169921875,
+      "learning_rate": 9.947355499530683e-05,
+      "loss": 0.0049,
+      "num_tokens": 4113282.0,
+      "reward": 3.1798436641693115,
+      "reward_std": 0.06495096534490585,
+      "rewards/check_gptzero_func": 0.6785714030265808,
+      "rewards/check_perplexity_diff_func": 1.9642857313156128,
+      "rewards/check_winston_local_func": 0.5369864702224731,
+      "step": 405
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5678321678321678,
+      "grad_norm": 0.006640329404323574,
+      "kl": 0.1728515625,
+      "learning_rate": 9.946468589238021e-05,
+      "loss": 0.0049,
+      "step": 406
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 121.39286041259766,
+      "epoch": 0.5692307692307692,
+      "grad_norm": 1.5620797981026132e-06,
+      "kl": 0.1396484375,
+      "learning_rate": 9.945574310466159e-05,
+      "loss": 0.0001,
+      "num_tokens": 4135506.0,
+      "reward": 2.573127269744873,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.2857142984867096,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.3588416278362274,
+      "step": 407
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5706293706293706,
+      "grad_norm": 1.51593967204227e-06,
+      "kl": 0.1396484375,
+      "learning_rate": 9.944672664547252e-05,
+      "loss": 0.0001,
+      "step": 408
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 107.8214340209961,
+      "epoch": 0.5720279720279721,
+      "grad_norm": 0.0007775272826913423,
+      "kl": 0.17578125,
+      "learning_rate": 9.943763652824436e-05,
+      "loss": 0.0002,
+      "num_tokens": 4156060.0,
+      "reward": 2.55350661277771,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3571428656578064,
+      "rewards/check_perplexity_diff_func": 1.7142857313156128,
+      "rewards/check_winston_local_func": 0.48207801580429077,
+      "step": 409
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5734265734265734,
+      "grad_norm": 0.0008914625567854122,
+      "kl": 0.1767578125,
+      "learning_rate": 9.942847276651811e-05,
+      "loss": 0.0002,
+      "step": 410
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 101.62500762939453,
+      "epoch": 0.5748251748251748,
+      "grad_norm": 4.9539161428254635e-05,
+      "kl": 0.1474609375,
+      "learning_rate": 9.941923537394456e-05,
+      "loss": 0.0001,
+      "num_tokens": 4176019.0,
+      "reward": 2.8906424045562744,
+      "reward_std": 0.00012528452498372644,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.5334992408752441,
+      "step": 411
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5762237762237762,
+      "grad_norm": 5.844422779033556e-05,
+      "kl": 0.1474609375,
+      "learning_rate": 9.940992436428409e-05,
+      "loss": 0.0001,
+      "step": 412
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 105.28572082519531,
+      "epoch": 0.5776223776223777,
+      "grad_norm": 0.0006008878616346907,
+      "kl": 0.2216796875,
+      "learning_rate": 9.940053975140684e-05,
+      "loss": 0.0002,
+      "num_tokens": 4196667.0,
+      "reward": 2.943253755569458,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.5357142686843872,
+      "rewards/check_perplexity_diff_func": 1.8571428060531616,
+      "rewards/check_winston_local_func": 0.5503963828086853,
+      "step": 413
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.579020979020979,
+      "grad_norm": 0.0003365492175069851,
+      "kl": 0.220703125,
+      "learning_rate": 9.939108154929252e-05,
+      "loss": 0.0002,
+      "step": 414
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 114.60714721679688,
+      "epoch": 0.5804195804195804,
+      "grad_norm": 0.008680207230787525,
+      "kl": 0.171875,
+      "learning_rate": 9.938154977203049e-05,
+      "loss": -0.0005,
+      "num_tokens": 4217727.0,
+      "reward": 2.7773444652557373,
+      "reward_std": 0.051937274634838104,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 1.8214285373687744,
+      "rewards/check_winston_local_func": 0.5273441076278687,
+      "step": 415
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5818181818181818,
+      "grad_norm": 0.00033864647249595475,
+      "kl": 0.171875,
+      "learning_rate": 9.937194443381972e-05,
+      "loss": -0.0005,
+      "step": 416
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 100.60714721679688,
+      "epoch": 0.5832167832167832,
+      "grad_norm": 4.017405087628847e-06,
+      "kl": 0.193359375,
+      "learning_rate": 9.936226554896875e-05,
+      "loss": 0.0002,
+      "num_tokens": 4237189.0,
+      "reward": 2.7684807777404785,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.44705215096473694,
+      "step": 417
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5846153846153846,
+      "grad_norm": 3.880139616189675e-06,
+      "kl": 0.193359375,
+      "learning_rate": 9.935251313189564e-05,
+      "loss": 0.0002,
+      "step": 418
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 107.75000762939453,
+      "epoch": 0.586013986013986,
+      "grad_norm": 0.00032680437267053027,
+      "kl": 0.1826171875,
+      "learning_rate": 9.934268719712807e-05,
+      "loss": 0.0002,
+      "num_tokens": 4258213.0,
+      "reward": 2.577641487121582,
+      "reward_std": 0.0015505586052313447,
+      "rewards/check_gptzero_func": 0.2857142984867096,
+      "rewards/check_perplexity_diff_func": 1.8571428060531616,
+      "rewards/check_winston_local_func": 0.4347843825817108,
+      "step": 419
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5874125874125874,
+      "grad_norm": 0.0003238465648274268,
+      "kl": 0.1826171875,
+      "learning_rate": 9.933278775930317e-05,
+      "loss": 0.0002,
+      "step": 420
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 108.67857360839844,
+      "epoch": 0.5888111888111888,
+      "grad_norm": 0.015591092439371165,
+      "kl": 0.16015625,
+      "learning_rate": 9.932281483316758e-05,
+      "loss": 0.0004,
+      "num_tokens": 4278847.0,
+      "reward": 2.615447998046875,
+      "reward_std": 0.05428782477974892,
+      "rewards/check_gptzero_func": 0.3214285671710968,
+      "rewards/check_perplexity_diff_func": 1.8214285373687744,
+      "rewards/check_winston_local_func": 0.47259071469306946,
+      "step": 421
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5902097902097903,
+      "grad_norm": 0.00016911180587449477,
+      "kl": 0.16015625,
+      "learning_rate": 9.931276843357742e-05,
+      "loss": 0.0003,
+      "step": 422
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 111.16072082519531,
+      "epoch": 0.5916083916083916,
+      "grad_norm": 1.6847767080478394e-05,
+      "kl": 0.1630859375,
+      "learning_rate": 9.930264857549825e-05,
+      "loss": 0.0002,
+      "num_tokens": 4300096.0,
+      "reward": 2.857067823410034,
+      "reward_std": 0.00101040443405509,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 1.8571428060531616,
+      "rewards/check_winston_local_func": 0.6070676445960999,
+      "step": 423
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.593006993006993,
+      "grad_norm": 2.453025078669727e-05,
+      "kl": 0.1630859375,
+      "learning_rate": 9.929245527400503e-05,
+      "loss": 0.0002,
+      "step": 424
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 102.50000762939453,
+      "epoch": 0.5944055944055944,
+      "grad_norm": 0.0015183266942611472,
+      "kl": 0.173828125,
+      "learning_rate": 9.928218854428221e-05,
+      "loss": 0.0002,
+      "num_tokens": 4320064.0,
+      "reward": 2.6430578231811523,
+      "reward_std": 5.020291791879572e-05,
+      "rewards/check_gptzero_func": 0.3571428656578064,
+      "rewards/check_perplexity_diff_func": 1.7857142686843872,
+      "rewards/check_winston_local_func": 0.5002006888389587,
+      "step": 425
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5958041958041959,
+      "grad_norm": 0.0013596061771866538,
+      "kl": 0.1708984375,
+      "learning_rate": 9.927184840162354e-05,
+      "loss": 0.0002,
+      "step": 426
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 110.66072082519531,
+      "epoch": 0.5972027972027972,
+      "grad_norm": 0.12076121626694686,
+      "kl": 0.189453125,
+      "learning_rate": 9.926143486143214e-05,
+      "loss": -0.001,
+      "num_tokens": 4341145.0,
+      "reward": 2.6579177379608154,
+      "reward_std": 0.05154726281762123,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 1.8214285373687744,
+      "rewards/check_winston_local_func": 0.4079175889492035,
+      "step": 427
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5986013986013986,
+      "grad_norm": 0.0008289218843548021,
+      "kl": 0.19140625,
+      "learning_rate": 9.92509479392205e-05,
+      "loss": -0.0013,
+      "step": 428
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 89.67857360839844,
+      "epoch": 0.6,
+      "grad_norm": 0.005379412774279031,
+      "kl": 0.1787109375,
+      "learning_rate": 9.924038765061042e-05,
+      "loss": 0.0001,
+      "num_tokens": 4359277.0,
+      "reward": 2.633777141571045,
+      "reward_std": 0.05039219558238983,
+      "rewards/check_gptzero_func": 0.2857142984867096,
+      "rewards/check_perplexity_diff_func": 1.9642857313156128,
+      "rewards/check_winston_local_func": 0.38377735018730164,
+      "step": 429
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6013986013986014,
+      "grad_norm": 0.010439668945298157,
+      "kl": 0.1787109375,
+      "learning_rate": 9.922975401133293e-05,
+      "loss": 0.0002,
+      "step": 430
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 102.67857360839844,
+      "epoch": 0.6027972027972028,
+      "grad_norm": 0.06317955454404899,
+      "kl": 0.201171875,
+      "learning_rate": 9.92190470372284e-05,
+      "loss": 0.0068,
+      "num_tokens": 4379109.0,
+      "reward": 2.6650938987731934,
+      "reward_std": 0.05930115655064583,
+      "rewards/check_gptzero_func": 0.25,
+      "rewards/check_perplexity_diff_func": 1.9642857313156128,
+      "rewards/check_winston_local_func": 0.4508081078529358,
+      "step": 431
+    },
+    {
+      "clip_ratio": 0.0002587991766631603,
+      "epoch": 0.6041958041958042,
+      "grad_norm": 0.05122842630547611,
+      "kl": 0.1865234375,
+      "learning_rate": 9.920826674424642e-05,
+      "loss": 0.0063,
+      "step": 432
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 80.35714721679688,
+      "epoch": 0.6055944055944056,
+      "grad_norm": 7.162658137250354e-07,
+      "kl": 0.1875,
+      "learning_rate": 9.919741314844577e-05,
+      "loss": 0.0002,
+      "num_tokens": 4396693.0,
+      "reward": 3.0637807846069336,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.5357142686843872,
+      "rewards/check_perplexity_diff_func": 1.8571428060531616,
+      "rewards/check_winston_local_func": 0.6709235310554504,
+      "step": 433
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.606993006993007,
+      "grad_norm": 8.451796450955314e-07,
+      "kl": 0.1875,
+      "learning_rate": 9.918648626599447e-05,
+      "loss": 0.0002,
+      "step": 434
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 124.98214721679688,
+      "epoch": 0.6083916083916084,
+      "grad_norm": 0.016099687268680437,
+      "kl": 0.1455078125,
+      "learning_rate": 9.91754861131697e-05,
+      "loss": -0.0001,
+      "num_tokens": 4418950.0,
+      "reward": 2.94807505607605,
+      "reward_std": 0.050821539014577866,
+      "rewards/check_gptzero_func": 0.5357142686843872,
+      "rewards/check_perplexity_diff_func": 1.8928571939468384,
+      "rewards/check_winston_local_func": 0.5195035338401794,
+      "step": 435
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6097902097902098,
+      "grad_norm": 0.014390989758817311,
+      "kl": 0.1455078125,
+      "learning_rate": 9.916441270635772e-05,
+      "loss": -0.0001,
+      "step": 436
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 106.64286041259766,
+      "epoch": 0.6111888111888112,
+      "grad_norm": 1.968332582426962e-06,
+      "kl": 0.14453125,
+      "learning_rate": 9.915326606205404e-05,
+      "loss": 0.0001,
+      "num_tokens": 4439018.0,
+      "reward": 2.77024507522583,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.4131018817424774,
+      "step": 437
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6125874125874126,
+      "grad_norm": 1.8961430119120163e-06,
+      "kl": 0.14453125,
+      "learning_rate": 9.914204619686314e-05,
+      "loss": 0.0001,
+      "step": 438
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 102.21428680419922,
+      "epoch": 0.6139860139860139,
+      "grad_norm": 0.000668351376631818,
+      "kl": 0.1728515625,
+      "learning_rate": 9.913075312749866e-05,
+      "loss": 0.0002,
+      "num_tokens": 4458806.0,
+      "reward": 2.8182172775268555,
+      "reward_std": 0.0025242711417376995,
+      "rewards/check_gptzero_func": 0.5357142686843872,
+      "rewards/check_perplexity_diff_func": 1.8571428060531616,
+      "rewards/check_winston_local_func": 0.42535990476608276,
+      "step": 439
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6153846153846154,
+      "grad_norm": 0.0006751756209535994,
+      "kl": 0.1728515625,
+      "learning_rate": 9.911938687078324e-05,
+      "loss": 0.0002,
+      "step": 440
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 81.25,
+      "epoch": 0.6167832167832168,
+      "grad_norm": 0.007904133813646444,
+      "kl": 0.240234375,
+      "learning_rate": 9.910794744364857e-05,
+      "loss": -0.0001,
+      "num_tokens": 4475982.0,
+      "reward": 2.8269970417022705,
+      "reward_std": 0.05014092102646828,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 1.8928571939468384,
+      "rewards/check_winston_local_func": 0.5412828326225281,
+      "step": 441
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6181818181818182,
+      "grad_norm": 0.00827597712943806,
+      "kl": 0.240234375,
+      "learning_rate": 9.909643486313533e-05,
+      "loss": -0.0001,
+      "step": 442
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 113.00000762939453,
+      "epoch": 0.6195804195804195,
+      "grad_norm": 6.016032313066646e-06,
+      "kl": 0.16796875,
+      "learning_rate": 9.908484914639318e-05,
+      "loss": 0.0002,
+      "num_tokens": 4497200.0,
+      "reward": 2.9243876934051514,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.5315303206443787,
+      "step": 443
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.620979020979021,
+      "grad_norm": 7.198804696892823e-06,
+      "kl": 0.16796875,
+      "learning_rate": 9.90731903106807e-05,
+      "loss": 0.0002,
+      "step": 444
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 129.375,
+      "epoch": 0.6223776223776224,
+      "grad_norm": 0.02075011817403209,
+      "kl": 0.158203125,
+      "learning_rate": 9.90614583733654e-05,
+      "loss": -0.0001,
+      "num_tokens": 4519807.0,
+      "reward": 2.866011619567871,
+      "reward_std": 0.10101933032274246,
+      "rewards/check_gptzero_func": 0.4642857015132904,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.47315436601638794,
+      "step": 445
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6237762237762238,
+      "grad_norm": 0.0034619333956659956,
+      "kl": 0.1591796875,
+      "learning_rate": 9.904965335192373e-05,
+      "loss": -0.0002,
+      "step": 446
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 98.92857360839844,
+      "epoch": 0.6251748251748251,
+      "grad_norm": 0.010057352537761423,
+      "kl": 0.212890625,
+      "learning_rate": 9.903777526394094e-05,
+      "loss": -0.0086,
+      "num_tokens": 4539039.0,
+      "reward": 2.980358362197876,
+      "reward_std": 0.05100167542695999,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 1.9642857313156128,
+      "rewards/check_winston_local_func": 0.587501049041748,
+      "step": 447
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6265734265734266,
+      "grad_norm": 0.006958390301115274,
+      "kl": 0.21875,
+      "learning_rate": 9.90258241271112e-05,
+      "loss": -0.0087,
+      "step": 448
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 111.71429443359375,
+      "epoch": 0.627972027972028,
+      "grad_norm": 1.6536770511152556e-06,
+      "kl": 0.142578125,
+      "learning_rate": 9.901379995923738e-05,
+      "loss": 0.0001,
+      "num_tokens": 4559675.0,
+      "reward": 2.681791067123413,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.2857142984867096,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.4675052762031555,
+      "step": 449
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6293706293706294,
+      "grad_norm": 1.6773170531853214e-06,
+      "kl": 0.142578125,
+      "learning_rate": 9.900170277823129e-05,
+      "loss": 0.0001,
+      "step": 450
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 107.35714721679688,
+      "epoch": 0.6307692307692307,
+      "grad_norm": 2.0958023033337787e-06,
+      "kl": 0.2109375,
+      "learning_rate": 9.898953260211338e-05,
+      "loss": 0.0002,
+      "num_tokens": 4579995.0,
+      "reward": 2.7400293350219727,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.4186006188392639,
+      "step": 451
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6321678321678321,
+      "grad_norm": 1.982971590769392e-06,
+      "kl": 0.2109375,
+      "learning_rate": 9.897728944901292e-05,
+      "loss": 0.0002,
+      "step": 452
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 118.26786041259766,
+      "epoch": 0.6335664335664336,
+      "grad_norm": 0.009450375112303724,
+      "kl": 0.1416015625,
+      "learning_rate": 9.896497333716783e-05,
+      "loss": -0.0051,
+      "num_tokens": 4602048.0,
+      "reward": 2.640944004058838,
+      "reward_std": 0.043768420815467834,
+      "rewards/check_gptzero_func": 0.5357142686843872,
+      "rewards/check_perplexity_diff_func": 1.5357142686843872,
+      "rewards/check_winston_local_func": 0.5695151090621948,
+      "step": 453
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.634965034965035,
+      "grad_norm": 0.0017713963386990425,
+      "kl": 0.1416015625,
+      "learning_rate": 9.895258428492475e-05,
+      "loss": -0.0051,
+      "step": 454
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 98.42857360839844,
+      "epoch": 0.6363636363636364,
+      "grad_norm": 0.014431904681184264,
+      "kl": 0.1943359375,
+      "learning_rate": 9.894012231073894e-05,
+      "loss": 0.0002,
+      "num_tokens": 4621654.0,
+      "reward": 2.7700040340423584,
+      "reward_std": 0.03728308901190758,
+      "rewards/check_gptzero_func": 0.4107142984867096,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.43071839213371277,
+      "step": 455
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6377622377622377,
+      "grad_norm": 0.008239966094546339,
+      "kl": 0.1943359375,
+      "learning_rate": 9.892758743317434e-05,
+      "loss": 0.0002,
+      "step": 456
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 90.50000762939453,
+      "epoch": 0.6391608391608392,
+      "grad_norm": 2.579298856684113e-06,
+      "kl": 0.1923828125,
+      "learning_rate": 9.891497967090344e-05,
+      "loss": 0.0002,
+      "num_tokens": 4640140.0,
+      "reward": 2.6383018493652344,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.25,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.4597306251525879,
+      "step": 457
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6405594405594406,
+      "grad_norm": 2.6625863420088633e-06,
+      "kl": 0.1923828125,
+      "learning_rate": 9.890229904270731e-05,
+      "loss": 0.0002,
+      "step": 458
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 113.03572082519531,
+      "epoch": 0.641958041958042,
+      "grad_norm": 1.7932858894376123e-05,
+      "kl": 0.1748046875,
+      "learning_rate": 9.888954556747563e-05,
+      "loss": 0.0002,
+      "num_tokens": 4661276.0,
+      "reward": 2.74210524559021,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3571428656578064,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.3849623203277588,
+      "step": 459
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6433566433566433,
+      "grad_norm": 1.7905922488331998e-05,
+      "kl": 0.1748046875,
+      "learning_rate": 9.887671926420648e-05,
+      "loss": 0.0002,
+      "step": 460
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 112.42857360839844,
+      "epoch": 0.6447552447552447,
+      "grad_norm": 0.014077582219597583,
+      "kl": 0.173828125,
+      "learning_rate": 9.886382015200652e-05,
+      "loss": 0.0002,
+      "num_tokens": 4682450.0,
+      "reward": 2.8680295944213867,
+      "reward_std": 0.05049533396959305,
+      "rewards/check_gptzero_func": 0.4642857015132904,
+      "rewards/check_perplexity_diff_func": 1.9642857313156128,
+      "rewards/check_winston_local_func": 0.43945807218551636,
+      "step": 461
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6461538461538462,
+      "grad_norm": 0.010796297977350395,
+      "kl": 0.173828125,
+      "learning_rate": 9.885084825009086e-05,
+      "loss": 0.0001,
+      "step": 462
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 105.14286041259766,
+      "epoch": 0.6475524475524476,
+      "grad_norm": 0.02303930816984521,
+      "kl": 0.17578125,
+      "learning_rate": 9.883780357778299e-05,
+      "loss": 0.0003,
+      "num_tokens": 4702710.0,
+      "reward": 2.885434150695801,
+      "reward_std": 0.06019994616508484,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 1.75,
+      "rewards/check_winston_local_func": 0.7068625092506409,
+      "step": 463
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6489510489510489,
+      "grad_norm": 0.0015050223064125838,
+      "kl": 0.1767578125,
+      "learning_rate": 9.882468615451484e-05,
+      "loss": 0.0002,
+      "step": 464
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 127.5714340209961,
+      "epoch": 0.6503496503496503,
+      "grad_norm": 3.0460982652821775e-06,
+      "kl": 0.1650390625,
+      "learning_rate": 9.881149599982671e-05,
+      "loss": 0.0002,
+      "num_tokens": 4725654.0,
+      "reward": 2.407341957092285,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.2857142984867096,
+      "rewards/check_perplexity_diff_func": 1.8571428060531616,
+      "rewards/check_winston_local_func": 0.2644847333431244,
+      "step": 465
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6517482517482518,
+      "grad_norm": 2.8166666383288673e-06,
+      "kl": 0.1650390625,
+      "learning_rate": 9.879823313336722e-05,
+      "loss": 0.0002,
+      "step": 466
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 122.16072082519531,
+      "epoch": 0.6531468531468532,
+      "grad_norm": 0.0001021700279542065,
+      "kl": 0.140625,
+      "learning_rate": 9.878489757489337e-05,
+      "loss": 0.0001,
+      "num_tokens": 4748375.0,
+      "reward": 2.6119117736816406,
+      "reward_std": 2.7804879209725186e-05,
+      "rewards/check_gptzero_func": 0.5357142686843872,
+      "rewards/check_perplexity_diff_func": 1.5714285373687744,
+      "rewards/check_winston_local_func": 0.5047687888145447,
+      "step": 467
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6545454545454545,
+      "grad_norm": 9.33433928584176e-05,
+      "kl": 0.140625,
+      "learning_rate": 9.877148934427037e-05,
+      "loss": 0.0001,
+      "step": 468
+    },
+    {
+      "clip_ratio": 0.0006711409660056233,
+      "completion_length": 95.64286041259766,
+      "epoch": 0.6559440559440559,
+      "grad_norm": 0.0004684591425632982,
+      "kl": 0.171875,
+      "learning_rate": 9.87580084614717e-05,
+      "loss": 0.0002,
+      "num_tokens": 4767439.0,
+      "reward": 2.3873257637023926,
+      "reward_std": 0.0017833748133853078,
+      "rewards/check_gptzero_func": 0.25,
+      "rewards/check_perplexity_diff_func": 1.7142857313156128,
+      "rewards/check_winston_local_func": 0.42303988337516785,
+      "step": 469
+    },
+    {
+      "clip_ratio": 0.0004793864209204912,
+      "epoch": 0.6573426573426573,
+      "grad_norm": 0.00047799981287926325,
+      "kl": 0.171875,
+      "learning_rate": 9.874445494657911e-05,
+      "loss": 0.0002,
+      "step": 470
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 79.39286041259766,
+      "epoch": 0.6587412587412588,
+      "grad_norm": 0.011441113949782547,
+      "kl": 0.216796875,
+      "learning_rate": 9.873082881978251e-05,
+      "loss": -0.0064,
+      "num_tokens": 4784715.0,
+      "reward": 3.037661075592041,
+      "reward_std": 0.023328183218836784,
+      "rewards/check_gptzero_func": 0.4821428656578064,
+      "rewards/check_perplexity_diff_func": 1.9642857313156128,
+      "rewards/check_winston_local_func": 0.5912323594093323,
+      "step": 471
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6601398601398601,
+      "grad_norm": 0.008433986682608557,
+      "kl": 0.216796875,
+      "learning_rate": 9.871713010137997e-05,
+      "loss": -0.0064,
+      "step": 472
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 119.42857360839844,
+      "epoch": 0.6615384615384615,
+      "grad_norm": 7.638914800437301e-06,
+      "kl": 0.1435546875,
+      "learning_rate": 9.870335881177774e-05,
+      "loss": 0.0001,
+      "num_tokens": 4806207.0,
+      "reward": 2.9706830978393555,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4642857015132904,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.5063972473144531,
+      "step": 473
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6629370629370629,
+      "grad_norm": 7.505737589947172e-06,
+      "kl": 0.1435546875,
+      "learning_rate": 9.868951497149011e-05,
+      "loss": 0.0001,
+      "step": 474
+    },
+    {
+      "clip_ratio": 0.00019073051225859672,
+      "completion_length": 89.87500762939453,
+      "epoch": 0.6643356643356644,
+      "grad_norm": 0.03662860311910791,
+      "kl": 0.2060546875,
+      "learning_rate": 9.86755986011395e-05,
+      "loss": 0.0065,
+      "num_tokens": 4824332.0,
+      "reward": 2.643022060394287,
+      "reward_std": 0.04776443541049957,
+      "rewards/check_gptzero_func": 0.3214285671710968,
+      "rewards/check_perplexity_diff_func": 1.8928571939468384,
+      "rewards/check_winston_local_func": 0.42873620986938477,
+      "step": 475
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6657342657342658,
+      "grad_norm": 0.03126231173733144,
+      "kl": 0.2060546875,
+      "learning_rate": 9.866160972145634e-05,
+      "loss": 0.0062,
+      "step": 476
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 123.10714721679688,
+      "epoch": 0.6671328671328671,
+      "grad_norm": 0.0002559996362709361,
+      "kl": 0.1357421875,
+      "learning_rate": 9.864754835327909e-05,
+      "loss": 0.0001,
+      "num_tokens": 4847094.0,
+      "reward": 2.767104387283325,
+      "reward_std": 0.0007265785825438797,
+      "rewards/check_gptzero_func": 0.5,
+      "rewards/check_perplexity_diff_func": 1.7142857313156128,
+      "rewards/check_winston_local_func": 0.5528185963630676,
+      "step": 477
+    },
+    {
+      "clip_ratio": 8.97343925316818e-05,
+      "epoch": 0.6685314685314685,
+      "grad_norm": 0.00042633622080268514,
+      "kl": 0.1357421875,
+      "learning_rate": 9.86334145175542e-05,
+      "loss": 0.0001,
+      "step": 478
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 96.67857360839844,
+      "epoch": 0.66993006993007,
+      "grad_norm": 0.0008693689003033054,
+      "kl": 0.16015625,
+      "learning_rate": 9.861920823533606e-05,
+      "loss": 0.0002,
+      "num_tokens": 4866394.0,
+      "reward": 2.5761330127716064,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3571428656578064,
+      "rewards/check_perplexity_diff_func": 1.7142857313156128,
+      "rewards/check_winston_local_func": 0.504704475402832,
+      "step": 479
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6713286713286714,
+      "grad_norm": 0.0008435837049153919,
+      "kl": 0.16015625,
+      "learning_rate": 9.860492952778696e-05,
+      "loss": 0.0002,
+      "step": 480
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 94.39286041259766,
+      "epoch": 0.6727272727272727,
+      "grad_norm": 3.349433978828943e-06,
+      "kl": 0.177734375,
+      "learning_rate": 9.859057841617709e-05,
+      "loss": 0.0002,
+      "num_tokens": 4885318.0,
+      "reward": 2.8612334728240967,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3571428656578064,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.5755191445350647,
+      "step": 481
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6741258741258741,
+      "grad_norm": 3.5095106255593373e-06,
+      "kl": 0.177734375,
+      "learning_rate": 9.857615492188452e-05,
+      "loss": 0.0002,
+      "step": 482
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 87.98214721679688,
+      "epoch": 0.6755244755244755,
+      "grad_norm": 0.00016582191324414645,
+      "kl": 0.208984375,
+      "learning_rate": 9.856165906639513e-05,
+      "loss": 0.0002,
+      "num_tokens": 4903401.0,
+      "reward": 2.797102928161621,
+      "reward_std": 0.0008326892857439816,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.4756740629673004,
+      "step": 483
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.676923076923077,
+      "grad_norm": 0.00017592290286556547,
+      "kl": 0.208984375,
+      "learning_rate": 9.85470908713026e-05,
+      "loss": 0.0002,
+      "step": 484
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 122.14286041259766,
+      "epoch": 0.6783216783216783,
+      "grad_norm": 3.840262302649335e-05,
+      "kl": 0.140625,
+      "learning_rate": 9.853245035830834e-05,
+      "loss": 0.0001,
+      "num_tokens": 4925777.0,
+      "reward": 2.80380916595459,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4642857015132904,
+      "rewards/check_perplexity_diff_func": 1.8571428060531616,
+      "rewards/check_winston_local_func": 0.4823804199695587,
+      "step": 485
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6797202797202797,
+      "grad_norm": 3.855750270963149e-05,
+      "kl": 0.140625,
+      "learning_rate": 9.851773754922152e-05,
+      "loss": 0.0001,
+      "step": 486
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 110.5714340209961,
+      "epoch": 0.6811188811188811,
+      "grad_norm": 1.0000521495087345e-06,
+      "kl": 0.1572265625,
+      "learning_rate": 9.850295246595898e-05,
+      "loss": 0.0002,
+      "num_tokens": 4946515.0,
+      "reward": 2.9591448307037354,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.5,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.5305731296539307,
+      "step": 487
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6825174825174826,
+      "grad_norm": 9.579988148024946e-07,
+      "kl": 0.1572265625,
+      "learning_rate": 9.848809513054523e-05,
+      "loss": 0.0002,
+      "step": 488
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 104.67857360839844,
+      "epoch": 0.6839160839160839,
+      "grad_norm": 4.4290499511336513e-07,
+      "kl": 0.1728515625,
+      "learning_rate": 9.847316556511245e-05,
+      "loss": 0.0002,
+      "num_tokens": 4966537.0,
+      "reward": 3.0352413654327393,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4642857015132904,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.5709553956985474,
+      "step": 489
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6853146853146853,
+      "grad_norm": 4.228282290104698e-07,
+      "kl": 0.1728515625,
+      "learning_rate": 9.845816379190036e-05,
+      "loss": 0.0002,
+      "step": 490
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 106.92857360839844,
+      "epoch": 0.6867132867132867,
+      "grad_norm": 7.161264654013381e-07,
+      "kl": 0.1865234375,
+      "learning_rate": 9.844308983325625e-05,
+      "loss": 0.0002,
+      "num_tokens": 4986837.0,
+      "reward": 2.8011527061462402,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4642857015132904,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.40829533338546753,
+      "step": 491
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6881118881118881,
+      "grad_norm": 6.606409864369399e-07,
+      "kl": 0.1865234375,
+      "learning_rate": 9.842794371163501e-05,
+      "loss": 0.0002,
+      "step": 492
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 127.3214340209961,
+      "epoch": 0.6895104895104895,
+      "grad_norm": 3.5056088482557826e-05,
+      "kl": 0.11767578125,
+      "learning_rate": 9.841272544959892e-05,
+      "loss": 0.0001,
+      "num_tokens": 5009783.0,
+      "reward": 2.6788315773010254,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.5,
+      "rewards/check_perplexity_diff_func": 1.7857142686843872,
+      "rewards/check_winston_local_func": 0.393117219209671,
+      "step": 493
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6909090909090909,
+      "grad_norm": 3.5363245519599165e-05,
+      "kl": 0.11767578125,
+      "learning_rate": 9.839743506981782e-05,
+      "loss": 0.0001,
+      "step": 494
+    },
+    {
+      "clip_ratio": 0.0004093328316230327,
+      "completion_length": 99.08928680419922,
+      "epoch": 0.6923076923076923,
+      "grad_norm": 0.0028759704706919223,
+      "kl": 0.1650390625,
+      "learning_rate": 9.838207259506891e-05,
+      "loss": 0.0002,
+      "num_tokens": 5029238.0,
+      "reward": 2.5216493606567383,
+      "reward_std": 0.0011310166446492076,
+      "rewards/check_gptzero_func": 0.3571428656578064,
+      "rewards/check_perplexity_diff_func": 1.7142857313156128,
+      "rewards/check_winston_local_func": 0.45022064447402954,
+      "step": 495
+    },
+    {
+      "clip_ratio": 0.0004093328316230327,
+      "epoch": 0.6937062937062937,
+      "grad_norm": 0.0035050811312865673,
+      "kl": 0.1650390625,
+      "learning_rate": 9.836663804823683e-05,
+      "loss": 0.0002,
+      "step": 496
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 124.85714721679688,
+      "epoch": 0.6951048951048951,
+      "grad_norm": 3.8191875557546916e-05,
+      "kl": 0.158203125,
+      "learning_rate": 9.835113145231356e-05,
+      "loss": 0.0002,
+      "num_tokens": 5051408.0,
+      "reward": 2.8089945316314697,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4642857015132904,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.4161372184753418,
+      "step": 497
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6965034965034965,
+      "grad_norm": 3.1854181850026e-05,
+      "kl": 0.158203125,
+      "learning_rate": 9.833555283039842e-05,
+      "loss": 0.0002,
+      "step": 498
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 115.17857360839844,
+      "epoch": 0.6979020979020979,
+      "grad_norm": 3.735234931957505e-07,
+      "kl": 0.15234375,
+      "learning_rate": 9.831990220569801e-05,
+      "loss": 0.0002,
+      "num_tokens": 5072486.0,
+      "reward": 2.7535111904144287,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3571428656578064,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.46779707074165344,
+      "step": 499
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6993006993006993,
+      "grad_norm": 3.749561048042854e-07,
+      "kl": 0.15234375,
+      "learning_rate": 9.83041796015262e-05,
+      "loss": 0.0002,
+      "step": 500
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 137.85714721679688,
+      "epoch": 0.7006993006993008,
+      "grad_norm": 0.004581651252064326,
+      "kl": 0.1357421875,
+      "learning_rate": 9.828838504130406e-05,
+      "loss": 0.0001,
+      "num_tokens": 24128.0,
+      "reward": 2.5751242637634277,
+      "reward_std": 0.050465863198041916,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 1.8214285373687744,
+      "rewards/check_winston_local_func": 0.36083847284317017,
+      "step": 501
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7020979020979021,
+      "grad_norm": 0.004623783518082634,
+      "kl": 0.1357421875,
+      "learning_rate": 9.827251854855991e-05,
+      "loss": 0.0001,
+      "step": 502
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 84.0714340209961,
+      "epoch": 0.7034965034965035,
+      "grad_norm": 5.4224474596128455e-06,
+      "kl": 0.1962890625,
+      "learning_rate": 9.825658014692914e-05,
+      "loss": 0.0002,
+      "num_tokens": 41660.0,
+      "reward": 2.881941556930542,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.48908427357673645,
+      "step": 503
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7048951048951049,
+      "grad_norm": 5.774418529053615e-06,
+      "kl": 0.1962890625,
+      "learning_rate": 9.824056986015433e-05,
+      "loss": 0.0002,
+      "step": 504
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 125.76786041259766,
+      "epoch": 0.7062937062937062,
+      "grad_norm": 0.012190639094338367,
+      "kl": 0.158203125,
+      "learning_rate": 9.82244877120851e-05,
+      "loss": -0.0,
+      "num_tokens": 63809.0,
+      "reward": 2.9153220653533936,
+      "reward_std": 0.05151599273085594,
+      "rewards/check_gptzero_func": 0.5,
+      "rewards/check_perplexity_diff_func": 1.9642857313156128,
+      "rewards/check_winston_local_func": 0.45103612542152405,
+      "step": 505
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7076923076923077,
+      "grad_norm": 0.0008706312902749691,
+      "kl": 0.1572265625,
+      "learning_rate": 9.820833372667812e-05,
+      "loss": -0.0001,
+      "step": 506
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 106.67857360839844,
+      "epoch": 0.7090909090909091,
+      "grad_norm": 7.4245877265908374e-06,
+      "kl": 0.171875,
+      "learning_rate": 9.819210792799712e-05,
+      "loss": 0.0002,
+      "num_tokens": 83877.0,
+      "reward": 2.742607831954956,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3571428656578064,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.3854646682739258,
+      "step": 507
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7104895104895105,
+      "grad_norm": 5.9863315776137925e-06,
+      "kl": 0.171875,
+      "learning_rate": 9.817581034021272e-05,
+      "loss": 0.0002,
+      "step": 508
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 112.21429443359375,
+      "epoch": 0.7118881118881119,
+      "grad_norm": 1.5219106062929897e-05,
+      "kl": 0.154296875,
+      "learning_rate": 9.815944098760257e-05,
+      "loss": 0.0002,
+      "num_tokens": 105225.0,
+      "reward": 2.886364698410034,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.5649359226226807,
+      "step": 509
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7132867132867133,
+      "grad_norm": 3.826854189322478e-05,
+      "kl": 0.154296875,
+      "learning_rate": 9.814299989455117e-05,
+      "loss": 0.0002,
+      "step": 510
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 93.50000762939453,
+      "epoch": 0.7146853146853147,
+      "grad_norm": 3.6634275067756628e-06,
+      "kl": 0.18359375,
+      "learning_rate": 9.81264870855499e-05,
+      "loss": 0.0002,
+      "num_tokens": 124237.0,
+      "reward": 2.9750092029571533,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.5,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.5464377403259277,
+      "step": 511
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7160839160839161,
+      "grad_norm": 3.3120330925968346e-06,
+      "kl": 0.18359375,
+      "learning_rate": 9.810990258519699e-05,
+      "loss": 0.0002,
+      "step": 512
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 114.5714340209961,
+      "epoch": 0.7174825174825175,
+      "grad_norm": 1.5887563588226216e-07,
+      "kl": 0.1474609375,
+      "learning_rate": 9.809324641819741e-05,
+      "loss": 0.0001,
+      "num_tokens": 145557.0,
+      "reward": 2.767850637435913,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3571428656578064,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.4107076823711395,
+      "step": 513
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7188811188811188,
+      "grad_norm": 1.5359302478120885e-07,
+      "kl": 0.1474609375,
+      "learning_rate": 9.807651860936297e-05,
+      "loss": 0.0001,
+      "step": 514
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 110.3214340209961,
+      "epoch": 0.7202797202797203,
+      "grad_norm": 1.620447571811333e-07,
+      "kl": 0.1630859375,
+      "learning_rate": 9.805971918361214e-05,
+      "loss": 0.0002,
+      "num_tokens": 165997.0,
+      "reward": 2.9055941104888916,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.5127367377281189,
+      "step": 515
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7216783216783217,
+      "grad_norm": 1.472040658632873e-07,
+      "kl": 0.1630859375,
+      "learning_rate": 9.804284816597008e-05,
+      "loss": 0.0002,
+      "step": 516
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 91.67857360839844,
+      "epoch": 0.7230769230769231,
+      "grad_norm": 2.4623250489986615e-06,
+      "kl": 0.142578125,
+      "learning_rate": 9.802590558156862e-05,
+      "loss": 0.0001,
+      "num_tokens": 184617.0,
+      "reward": 2.853912830352783,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.46105554699897766,
+      "step": 517
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7244755244755244,
+      "grad_norm": 2.0559549189734124e-06,
+      "kl": 0.142578125,
+      "learning_rate": 9.800889145564617e-05,
+      "loss": 0.0001,
+      "step": 518
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 127.21429443359375,
+      "epoch": 0.7258741258741259,
+      "grad_norm": 2.3530034340292295e-07,
+      "kl": 0.1552734375,
+      "learning_rate": 9.799180581354774e-05,
+      "loss": 0.0002,
+      "num_tokens": 207223.0,
+      "reward": 2.7364511489868164,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3571428656578064,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.45073673129081726,
+      "step": 519
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7272727272727273,
+      "grad_norm": 2.2322343800833895e-07,
+      "kl": 0.1552734375,
+      "learning_rate": 9.797464868072488e-05,
+      "loss": 0.0002,
+      "step": 520
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 104.5714340209961,
+      "epoch": 0.7286713286713287,
+      "grad_norm": 2.968346809067501e-07,
+      "kl": 0.177734375,
+      "learning_rate": 9.795742008273558e-05,
+      "loss": 0.0002,
+      "num_tokens": 227059.0,
+      "reward": 2.6932995319366455,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3571428656578064,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.3361565172672272,
+      "step": 521
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.73006993006993,
+      "grad_norm": 2.852980195095417e-07,
+      "kl": 0.177734375,
+      "learning_rate": 9.794012004524434e-05,
+      "loss": 0.0002,
+      "step": 522
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 105.42857360839844,
+      "epoch": 0.7314685314685314,
+      "grad_norm": 3.3796985539550523e-06,
+      "kl": 0.1640625,
+      "learning_rate": 9.792274859402205e-05,
+      "loss": 0.0002,
+      "num_tokens": 246949.0,
+      "reward": 2.6483211517333984,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3214285671710968,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.3983212113380432,
+      "step": 523
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7328671328671329,
+      "grad_norm": 3.1476330793325015e-06,
+      "kl": 0.1640625,
+      "learning_rate": 9.790530575494603e-05,
+      "loss": 0.0002,
+      "step": 524
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 125.03572082519531,
+      "epoch": 0.7342657342657343,
+      "grad_norm": 0.0060337949589422705,
+      "kl": 0.16796875,
+      "learning_rate": 9.788779155399987e-05,
+      "loss": -0.0001,
+      "num_tokens": 269441.0,
+      "reward": 2.768662214279175,
+      "reward_std": 0.049707408994436264,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 1.9642857313156128,
+      "rewards/check_winston_local_func": 0.4115191400051117,
+      "step": 525
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7356643356643356,
+      "grad_norm": 0.0068588182961818464,
+      "kl": 0.16796875,
+      "learning_rate": 9.787020601727352e-05,
+      "loss": -0.0,
+      "step": 526
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 78.39286041259766,
+      "epoch": 0.737062937062937,
+      "grad_norm": 1.3153821115529096e-05,
+      "kl": 0.2353515625,
+      "learning_rate": 9.785254917096318e-05,
+      "loss": 0.0002,
+      "num_tokens": 286345.0,
+      "reward": 3.074557304382324,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.5,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.5745573043823242,
+      "step": 527
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7384615384615385,
+      "grad_norm": 1.256427796379324e-05,
+      "kl": 0.2353515625,
+      "learning_rate": 9.783482104137127e-05,
+      "loss": 0.0002,
+      "step": 528
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 124.39286041259766,
+      "epoch": 0.7398601398601399,
+      "grad_norm": 1.3517164643713064e-05,
+      "kl": 0.12353515625,
+      "learning_rate": 9.781702165490639e-05,
+      "loss": 0.0001,
+      "num_tokens": 309195.0,
+      "reward": 2.4804210662841797,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 1.6428571939468384,
+      "rewards/check_winston_local_func": 0.44470664858818054,
+      "step": 529
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7412587412587412,
+      "grad_norm": 1.3131407312669822e-05,
+      "kl": 0.12353515625,
+      "learning_rate": 9.779915103808328e-05,
+      "loss": 0.0001,
+      "step": 530
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 111.53572082519531,
+      "epoch": 0.7426573426573426,
+      "grad_norm": 7.637416736333039e-07,
+      "kl": 0.15625,
+      "learning_rate": 9.778120921752285e-05,
+      "loss": 0.0002,
+      "num_tokens": 330505.0,
+      "reward": 3.070021390914917,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.5357142686843872,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.5343068838119507,
+      "step": 531
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7440559440559441,
+      "grad_norm": 7.316834780823122e-07,
+      "kl": 0.15625,
+      "learning_rate": 9.776319621995201e-05,
+      "loss": 0.0002,
+      "step": 532
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 106.35714721679688,
+      "epoch": 0.7454545454545455,
+      "grad_norm": 3.663080215251399e-07,
+      "kl": 0.193359375,
+      "learning_rate": 9.77451120722037e-05,
+      "loss": 0.0002,
+      "num_tokens": 350813.0,
+      "reward": 2.63775634765625,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3214285671710968,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.3877563774585724,
+      "step": 533
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7468531468531469,
+      "grad_norm": 3.8678973306591443e-07,
+      "kl": 0.193359375,
+      "learning_rate": 9.77269568012169e-05,
+      "loss": 0.0002,
+      "step": 534
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 106.67857360839844,
+      "epoch": 0.7482517482517482,
+      "grad_norm": 6.123173544504598e-06,
+      "kl": 0.2216796875,
+      "learning_rate": 9.770873043403648e-05,
+      "loss": 0.0002,
+      "num_tokens": 371585.0,
+      "reward": 2.874614953994751,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.5174719095230103,
+      "step": 535
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7496503496503496,
+      "grad_norm": 6.141198966103098e-06,
+      "kl": 0.2216796875,
+      "learning_rate": 9.769043299781327e-05,
+      "loss": 0.0002,
+      "step": 536
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 104.71428680419922,
+      "epoch": 0.7510489510489511,
+      "grad_norm": 6.024782598667822e-07,
+      "kl": 0.212890625,
+      "learning_rate": 9.767206451980394e-05,
+      "loss": 0.0002,
+      "num_tokens": 391201.0,
+      "reward": 2.666369676589966,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.25,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.4163695275783539,
+      "step": 537
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7524475524475525,
+      "grad_norm": 5.651815000751003e-07,
+      "kl": 0.212890625,
+      "learning_rate": 9.765362502737097e-05,
+      "loss": 0.0002,
+      "step": 538
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 107.21428680419922,
+      "epoch": 0.7538461538461538,
+      "grad_norm": 4.5359076717402366e-07,
+      "kl": 0.19140625,
+      "learning_rate": 9.763511454798268e-05,
+      "loss": 0.0002,
+      "num_tokens": 412087.0,
+      "reward": 2.660142660140991,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.25,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.4815710484981537,
+      "step": 539
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7552447552447552,
+      "grad_norm": 7.72545386297213e-07,
+      "kl": 0.19140625,
+      "learning_rate": 9.761653310921307e-05,
+      "loss": 0.0002,
+      "step": 540
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 133.85714721679688,
+      "epoch": 0.7566433566433567,
+      "grad_norm": 8.430058973039463e-08,
+      "kl": 0.146484375,
+      "learning_rate": 9.759788073874189e-05,
+      "loss": 0.0001,
+      "num_tokens": 435535.0,
+      "reward": 2.720244884490967,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.3631021976470947,
+      "step": 541
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7580419580419581,
+      "grad_norm": 8.327469845321351e-08,
+      "kl": 0.146484375,
+      "learning_rate": 9.757915746435453e-05,
+      "loss": 0.0001,
+      "step": 542
+    },
+    {
+      "clip_ratio": 0.0002040816325461492,
+      "completion_length": 102.89286041259766,
+      "epoch": 0.7594405594405594,
+      "grad_norm": 0.000356312443076848,
+      "kl": 0.203125,
+      "learning_rate": 9.756036331394202e-05,
+      "loss": 0.0002,
+      "num_tokens": 455137.0,
+      "reward": 2.6731011867523193,
+      "reward_std": 1.646135569899343e-05,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.351672500371933,
+      "step": 543
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7608391608391608,
+      "grad_norm": 0.00031516713281171804,
+      "kl": 0.203125,
+      "learning_rate": 9.754149831550098e-05,
+      "loss": 0.0002,
+      "step": 544
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 89.35714721679688,
+      "epoch": 0.7622377622377622,
+      "grad_norm": 2.70311309473727e-07,
+      "kl": 0.2080078125,
+      "learning_rate": 9.752256249713351e-05,
+      "loss": 0.0002,
+      "num_tokens": 473297.0,
+      "reward": 2.869070291519165,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3571428656578064,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.5119272470474243,
+      "step": 545
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7636363636363637,
+      "grad_norm": 2.830814999627113e-07,
+      "kl": 0.2080078125,
+      "learning_rate": 9.750355588704727e-05,
+      "loss": 0.0002,
+      "step": 546
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 132.44644165039062,
+      "epoch": 0.765034965034965,
+      "grad_norm": 0.0009688545197921594,
+      "kl": 0.1806640625,
+      "learning_rate": 9.748447851355535e-05,
+      "loss": 0.0002,
+      "num_tokens": 496926.0,
+      "reward": 2.7388832569122314,
+      "reward_std": 0.0023702967446297407,
+      "rewards/check_gptzero_func": 0.5357142686843872,
+      "rewards/check_perplexity_diff_func": 1.7142857313156128,
+      "rewards/check_winston_local_func": 0.4888834059238434,
+      "step": 547
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7664335664335664,
+      "grad_norm": 0.0016734864168510978,
+      "kl": 0.1806640625,
+      "learning_rate": 9.746533040507624e-05,
+      "loss": 0.0002,
+      "step": 548
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 97.03572082519531,
+      "epoch": 0.7678321678321678,
+      "grad_norm": 1.9163803223436476e-06,
+      "kl": 0.177734375,
+      "learning_rate": 9.744611159013381e-05,
+      "loss": 0.0002,
+      "num_tokens": 516080.0,
+      "reward": 2.9889118671417236,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.5,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.48891177773475647,
+      "step": 549
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7692307692307693,
+      "grad_norm": 1.2761616262245509e-06,
+      "kl": 0.177734375,
+      "learning_rate": 9.742682209735727e-05,
+      "loss": 0.0002,
+      "step": 550
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 92.39286041259766,
+      "epoch": 0.7706293706293706,
+      "grad_norm": 1.896094459987994e-07,
+      "kl": 0.1923828125,
+      "learning_rate": 9.740746195548112e-05,
+      "loss": 0.0002,
+      "num_tokens": 534554.0,
+      "reward": 2.943678855895996,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.5151072144508362,
+      "step": 551
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.772027972027972,
+      "grad_norm": 2.1137044966091938e-07,
+      "kl": 0.1923828125,
+      "learning_rate": 9.738803119334506e-05,
+      "loss": 0.0002,
+      "step": 552
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 103.21428680419922,
+      "epoch": 0.7734265734265734,
+      "grad_norm": 4.705318347406964e-07,
+      "kl": 0.1826171875,
+      "learning_rate": 9.736852983989404e-05,
+      "loss": 0.0002,
+      "num_tokens": 554220.0,
+      "reward": 2.9958415031433105,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.5,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.4958415627479553,
+      "step": 553
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7748251748251749,
+      "grad_norm": 5.24917119837944e-07,
+      "kl": 0.1826171875,
+      "learning_rate": 9.734895792417811e-05,
+      "loss": 0.0002,
+      "step": 554
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 102.53572082519531,
+      "epoch": 0.7762237762237763,
+      "grad_norm": 0.0005595837692618245,
+      "kl": 0.1611328125,
+      "learning_rate": 9.73293154753525e-05,
+      "loss": -0.0001,
+      "num_tokens": 574520.0,
+      "reward": 2.660358190536499,
+      "reward_std": 0.05061452463269234,
+      "rewards/check_gptzero_func": 0.2857142984867096,
+      "rewards/check_perplexity_diff_func": 1.75,
+      "rewards/check_winston_local_func": 0.6246438026428223,
+      "step": 555
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7776223776223776,
+      "grad_norm": 0.0005476655329061088,
+      "kl": 0.1611328125,
+      "learning_rate": 9.730960252267743e-05,
+      "loss": -0.0001,
+      "step": 556
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 110.28572082519531,
+      "epoch": 0.779020979020979,
+      "grad_norm": 1.9994241055330475e-07,
+      "kl": 0.162109375,
+      "learning_rate": 9.728981909551824e-05,
+      "loss": 0.0002,
+      "num_tokens": 595056.0,
+      "reward": 2.7659287452697754,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3571428656578064,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.40878555178642273,
+      "step": 557
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7804195804195804,
+      "grad_norm": 2.102720872162732e-07,
+      "kl": 0.162109375,
+      "learning_rate": 9.726996522334516e-05,
+      "loss": 0.0002,
+      "step": 558
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 116.3214340209961,
+      "epoch": 0.7818181818181819,
+      "grad_norm": 1.4418577967265551e-06,
+      "kl": 0.197265625,
+      "learning_rate": 9.725004093573342e-05,
+      "loss": 0.0002,
+      "num_tokens": 616218.0,
+      "reward": 2.90523099899292,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.4766596257686615,
+      "step": 559
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7832167832167832,
+      "grad_norm": 1.4761501812795077e-06,
+      "kl": 0.197265625,
+      "learning_rate": 9.723004626236314e-05,
+      "loss": 0.0002,
+      "step": 560
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 92.28572082519531,
+      "epoch": 0.7846153846153846,
+      "grad_norm": 6.640556191873379e-07,
+      "kl": 0.1904296875,
+      "learning_rate": 9.720998123301923e-05,
+      "loss": 0.0002,
+      "num_tokens": 634450.0,
+      "reward": 2.910466432571411,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4642857015132904,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.4461804926395416,
+      "step": 561
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.786013986013986,
+      "grad_norm": 6.785626739636823e-07,
+      "kl": 0.1904296875,
+      "learning_rate": 9.718984587759148e-05,
+      "loss": 0.0002,
+      "step": 562
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 97.67857360839844,
+      "epoch": 0.7874125874125875,
+      "grad_norm": 0.0002364011375932314,
+      "kl": 0.1953125,
+      "learning_rate": 9.71696402260744e-05,
+      "loss": 0.0002,
+      "num_tokens": 653730.0,
+      "reward": 3.09153151512146,
+      "reward_std": 4.5663102355320007e-05,
+      "rewards/check_gptzero_func": 0.5714285969734192,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.5201030373573303,
+      "step": 563
+    },
+    {
+      "clip_ratio": 0.00014228800137061626,
+      "epoch": 0.7888111888111888,
+      "grad_norm": 0.00020388090420570156,
+      "kl": 0.1953125,
+      "learning_rate": 9.714936430856723e-05,
+      "loss": 0.0002,
+      "step": 564
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 120.66072082519531,
+      "epoch": 0.7902097902097902,
+      "grad_norm": 8.089432221384178e-05,
+      "kl": 0.1357421875,
+      "learning_rate": 9.712901815527386e-05,
+      "loss": 0.0001,
+      "num_tokens": 675657.0,
+      "reward": 2.674098253250122,
+      "reward_std": 0.0001108883589040488,
+      "rewards/check_gptzero_func": 0.3571428656578064,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.38838380575180054,
+      "step": 565
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7916083916083916,
+      "grad_norm": 8.889658368646956e-05,
+      "kl": 0.1357421875,
+      "learning_rate": 9.710860179650287e-05,
+      "loss": 0.0001,
+      "step": 566
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 107.3214340209961,
+      "epoch": 0.793006993006993,
+      "grad_norm": 0.0001037228042730343,
+      "kl": 0.17578125,
+      "learning_rate": 9.70881152626673e-05,
+      "loss": 0.0002,
+      "num_tokens": 695807.0,
+      "reward": 2.615325927734375,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.25,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.4367544949054718,
+      "step": 567
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7944055944055944,
+      "grad_norm": 9.62410190276379e-05,
+      "kl": 0.17578125,
+      "learning_rate": 9.706755858428486e-05,
+      "loss": 0.0002,
+      "step": 568
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 128.3928680419922,
+      "epoch": 0.7958041958041958,
+      "grad_norm": 3.2052372287941215e-05,
+      "kl": 0.12890625,
+      "learning_rate": 9.704693179197767e-05,
+      "loss": 0.0001,
+      "num_tokens": 719201.0,
+      "reward": 2.616921901702881,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 1.7857142686843872,
+      "rewards/check_winston_local_func": 0.40263620018959045,
+      "step": 569
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7972027972027972,
+      "grad_norm": 3.1589569913845654e-05,
+      "kl": 0.12890625,
+      "learning_rate": 9.702623491647233e-05,
+      "loss": 0.0001,
+      "step": 570
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 97.3214340209961,
+      "epoch": 0.7986013986013986,
+      "grad_norm": 2.966365002312992e-07,
+      "kl": 0.1572265625,
+      "learning_rate": 9.70054679885998e-05,
+      "loss": 0.0002,
+      "num_tokens": 738947.0,
+      "reward": 2.835333824157715,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.47819074988365173,
+      "step": 571
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8,
+      "grad_norm": 4.1573650424054196e-07,
+      "kl": 0.1572265625,
+      "learning_rate": 9.698463103929542e-05,
+      "loss": 0.0002,
+      "step": 572
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 105.78572082519531,
+      "epoch": 0.8013986013986014,
+      "grad_norm": 0.000588551803293507,
+      "kl": 0.1611328125,
+      "learning_rate": 9.696372409959886e-05,
+      "loss": 0.0002,
+      "num_tokens": 759359.0,
+      "reward": 2.967468738555908,
+      "reward_std": 0.0020603849552571774,
+      "rewards/check_gptzero_func": 0.5,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.5388973355293274,
+      "step": 573
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8027972027972028,
+      "grad_norm": 0.0006215594211925717,
+      "kl": 0.1611328125,
+      "learning_rate": 9.694274720065399e-05,
+      "loss": 0.0002,
+      "step": 574
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 97.14286041259766,
+      "epoch": 0.8041958041958042,
+      "grad_norm": 3.3967306751168834e-06,
+      "kl": 0.17578125,
+      "learning_rate": 9.692170037370898e-05,
+      "loss": 0.0002,
+      "num_tokens": 778637.0,
+      "reward": 2.9381115436553955,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.509539783000946,
+      "step": 575
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8055944055944056,
+      "grad_norm": 3.2036118908588705e-06,
+      "kl": 0.17578125,
+      "learning_rate": 9.690058365011607e-05,
+      "loss": 0.0002,
+      "step": 576
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 102.64286041259766,
+      "epoch": 0.806993006993007,
+      "grad_norm": 8.125129903430204e-06,
+      "kl": 0.1630859375,
+      "learning_rate": 9.687939706133168e-05,
+      "loss": 0.0002,
+      "num_tokens": 798473.0,
+      "reward": 2.781018018722534,
+      "reward_std": 0.025253813713788986,
+      "rewards/check_gptzero_func": 0.4107142984867096,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.4417320787906647,
+      "step": 577
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8083916083916084,
+      "grad_norm": 7.98774787404577e-06,
+      "kl": 0.1630859375,
+      "learning_rate": 9.685814063891631e-05,
+      "loss": 0.0002,
+      "step": 578
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 126.85714721679688,
+      "epoch": 0.8097902097902098,
+      "grad_norm": 3.679599498892351e-07,
+      "kl": 0.138671875,
+      "learning_rate": 9.683681441453445e-05,
+      "loss": 0.0001,
+      "num_tokens": 821081.0,
+      "reward": 2.8629186153411865,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.43434715270996094,
+      "step": 579
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8111888111888111,
+      "grad_norm": 3.93663211553172e-07,
+      "kl": 0.138671875,
+      "learning_rate": 9.681541841995461e-05,
+      "loss": 0.0001,
+      "step": 580
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 106.00000762939453,
+      "epoch": 0.8125874125874126,
+      "grad_norm": 8.022029708518222e-07,
+      "kl": 0.162109375,
+      "learning_rate": 9.67939526870492e-05,
+      "loss": 0.0002,
+      "num_tokens": 841529.0,
+      "reward": 2.8752760887145996,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.55384761095047,
+      "step": 581
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.813986013986014,
+      "grad_norm": 8.501682817147847e-07,
+      "kl": 0.162109375,
+      "learning_rate": 9.677241724779453e-05,
+      "loss": 0.0002,
+      "step": 582
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 109.85714721679688,
+      "epoch": 0.8153846153846154,
+      "grad_norm": 1.7930012533014129e-06,
+      "kl": 0.1689453125,
+      "learning_rate": 9.675081213427076e-05,
+      "loss": 0.0002,
+      "num_tokens": 862213.0,
+      "reward": 2.9139628410339355,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.48539119958877563,
+      "step": 583
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8167832167832167,
+      "grad_norm": 1.6519465345892093e-06,
+      "kl": 0.1689453125,
+      "learning_rate": 9.672913737866179e-05,
+      "loss": 0.0002,
+      "step": 584
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 98.66072082519531,
+      "epoch": 0.8181818181818182,
+      "grad_norm": 0.002787236207669909,
+      "kl": 0.25,
+      "learning_rate": 9.670739301325534e-05,
+      "loss": 0.0002,
+      "num_tokens": 881842.0,
+      "reward": 2.6230576038360596,
+      "reward_std": 0.006216124631464481,
+      "rewards/check_gptzero_func": 0.3571428656578064,
+      "rewards/check_perplexity_diff_func": 1.7857142686843872,
+      "rewards/check_winston_local_func": 0.48020049929618835,
+      "step": 585
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8195804195804196,
+      "grad_norm": 0.002885109231591189,
+      "kl": 0.2490234375,
+      "learning_rate": 9.668557907044276e-05,
+      "loss": 0.0002,
+      "step": 586
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 79.14286041259766,
+      "epoch": 0.820979020979021,
+      "grad_norm": 4.462250819797621e-06,
+      "kl": 0.1923828125,
+      "learning_rate": 9.666369558271909e-05,
+      "loss": 0.0002,
+      "num_tokens": 898716.0,
+      "reward": 2.900649309158325,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.5077921152114868,
+      "step": 587
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8223776223776224,
+      "grad_norm": 4.488075066816524e-06,
+      "kl": 0.1923828125,
+      "learning_rate": 9.66417425826829e-05,
+      "loss": 0.0002,
+      "step": 588
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 112.03572082519531,
+      "epoch": 0.8237762237762237,
+      "grad_norm": 1.3821264354883407e-07,
+      "kl": 0.14453125,
+      "learning_rate": 9.661972010303641e-05,
+      "loss": 0.0001,
+      "num_tokens": 919636.0,
+      "reward": 2.7668612003326416,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4642857015132904,
+      "rewards/check_perplexity_diff_func": 1.8571428060531616,
+      "rewards/check_winston_local_func": 0.4454323649406433,
+      "step": 589
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8251748251748252,
+      "grad_norm": 1.4174011116816794e-07,
+      "kl": 0.14453125,
+      "learning_rate": 9.659762817658524e-05,
+      "loss": 0.0001,
+      "step": 590
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 107.92857360839844,
+      "epoch": 0.8265734265734266,
+      "grad_norm": 8.773734381605369e-07,
+      "kl": 0.1572265625,
+      "learning_rate": 9.65754668362385e-05,
+      "loss": 0.0002,
+      "num_tokens": 940096.0,
+      "reward": 2.8289589881896973,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4642857015132904,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.4361015856266022,
+      "step": 591
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.827972027972028,
+      "grad_norm": 1.0023469469060766e-06,
+      "kl": 0.1572265625,
+      "learning_rate": 9.655323611500875e-05,
+      "loss": 0.0002,
+      "step": 592
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 113.85714721679688,
+      "epoch": 0.8293706293706293,
+      "grad_norm": 6.1541718025779055e-06,
+      "kl": 0.166015625,
+      "learning_rate": 9.653093604601183e-05,
+      "loss": 0.0002,
+      "num_tokens": 961340.0,
+      "reward": 2.9606730937957764,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.5357142686843872,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.4249587059020996,
+      "step": 593
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8307692307692308,
+      "grad_norm": 6.512813773208274e-06,
+      "kl": 0.166015625,
+      "learning_rate": 9.650856666246693e-05,
+      "loss": 0.0002,
+      "step": 594
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 91.60714721679688,
+      "epoch": 0.8321678321678322,
+      "grad_norm": 3.2190340652283214e-06,
+      "kl": 0.1767578125,
+      "learning_rate": 9.648612799769644e-05,
+      "loss": 0.0002,
+      "num_tokens": 979916.0,
+      "reward": 3.0096070766448975,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.5810357332229614,
+      "step": 595
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8335664335664336,
+      "grad_norm": 3.1137752081162388e-06,
+      "kl": 0.1767578125,
+      "learning_rate": 9.646362008512602e-05,
+      "loss": 0.0002,
+      "step": 596
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 106.53572082519531,
+      "epoch": 0.8349650349650349,
+      "grad_norm": 6.812051267845749e-06,
+      "kl": 0.185546875,
+      "learning_rate": 9.644104295828447e-05,
+      "loss": 0.0002,
+      "num_tokens": 1000300.0,
+      "reward": 2.7739078998565674,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.4167649447917938,
+      "step": 597
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8363636363636363,
+      "grad_norm": 2.398683493432653e-06,
+      "kl": 0.185546875,
+      "learning_rate": 9.641839665080363e-05,
+      "loss": 0.0002,
+      "step": 598
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 107.41072082519531,
+      "epoch": 0.8377622377622378,
+      "grad_norm": 0.0008785473557125962,
+      "kl": 0.1796875,
+      "learning_rate": 9.63956811964185e-05,
+      "loss": 0.0002,
+      "num_tokens": 1021105.0,
+      "reward": 2.91171932220459,
+      "reward_std": 0.00400555832311511,
+      "rewards/check_gptzero_func": 0.5357142686843872,
+      "rewards/check_perplexity_diff_func": 1.8571428060531616,
+      "rewards/check_winston_local_func": 0.518862247467041,
+      "step": 599
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8391608391608392,
+      "grad_norm": 0.000892784560673435,
+      "kl": 0.1796875,
+      "learning_rate": 9.6372896628967e-05,
+      "loss": 0.0002,
+      "step": 600
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 78.8214340209961,
+      "epoch": 0.8405594405594405,
+      "grad_norm": 3.255738764439449e-05,
+      "kl": 0.2236328125,
+      "learning_rate": 9.635004298239004e-05,
+      "loss": 0.0002,
+      "num_tokens": 16776.0,
+      "reward": 2.546457290649414,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 1.0714285373687744,
+      "rewards/check_perplexity_diff_func": 0.9285714030265808,
+      "rewards/check_winston_local_func": 0.5464571118354797,
+      "step": 601
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8419580419580419,
+      "grad_norm": 2.996409313034994e-05,
+      "kl": 0.2236328125,
+      "learning_rate": 9.632712029073141e-05,
+      "loss": 0.0002,
+      "step": 602
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 98.50000762939453,
+      "epoch": 0.8433566433566434,
+      "grad_norm": 6.836714560556308e-07,
+      "kl": 0.1796875,
+      "learning_rate": 9.63041285881378e-05,
+      "loss": 0.0002,
+      "num_tokens": 36316.0,
+      "reward": 2.2952983379364014,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.8571428656578064,
+      "rewards/check_perplexity_diff_func": 0.9642857313156128,
+      "rewards/check_winston_local_func": 0.4738696217536926,
+      "step": 603
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8447552447552448,
+      "grad_norm": 6.51699821327856e-07,
+      "kl": 0.1796875,
+      "learning_rate": 9.628106790885865e-05,
+      "loss": 0.0002,
+      "step": 604
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 108.67857360839844,
+      "epoch": 0.8461538461538461,
+      "grad_norm": 1.0500525475797962e-05,
+      "kl": 0.220703125,
+      "learning_rate": 9.625793828724618e-05,
+      "loss": 0.0002,
+      "num_tokens": 56798.0,
+      "reward": 2.2515501976013184,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.7857142686843872,
+      "rewards/check_perplexity_diff_func": 0.9642857313156128,
+      "rewards/check_winston_local_func": 0.5015498399734497,
+      "step": 605
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8475524475524475,
+      "grad_norm": 9.962219746884584e-06,
+      "kl": 0.220703125,
+      "learning_rate": 9.62347397577553e-05,
+      "loss": 0.0002,
+      "step": 606
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 123.64286041259766,
+      "epoch": 0.848951048951049,
+      "grad_norm": 0.00014103568607896937,
+      "kl": 0.1728515625,
+      "learning_rate": 9.621147235494356e-05,
+      "loss": 0.0002,
+      "num_tokens": 79358.0,
+      "reward": 2.151411771774292,
+      "reward_std": 0.00029218001873232424,
+      "rewards/check_gptzero_func": 0.7142857313156128,
+      "rewards/check_perplexity_diff_func": 0.9285714030265808,
+      "rewards/check_winston_local_func": 0.5085545778274536,
+      "step": 607
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8503496503496504,
+      "grad_norm": 0.00012130440152598759,
+      "kl": 0.171875,
+      "learning_rate": 9.618813611347113e-05,
+      "loss": 0.0002,
+      "step": 608
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 118.71429443359375,
+      "epoch": 0.8517482517482518,
+      "grad_norm": 1.8287469066668684e-07,
+      "kl": 0.1494140625,
+      "learning_rate": 9.61647310681007e-05,
+      "loss": 0.0001,
+      "num_tokens": 101324.0,
+      "reward": 2.4129910469055176,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.9285714030265808,
+      "rewards/check_perplexity_diff_func": 0.9285714030265808,
+      "rewards/check_winston_local_func": 0.5558480024337769,
+      "step": 609
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8531468531468531,
+      "grad_norm": 1.9169052102503996e-07,
+      "kl": 0.1494140625,
+      "learning_rate": 9.614125725369747e-05,
+      "loss": 0.0001,
+      "step": 610
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 84.30357360839844,
+      "epoch": 0.8545454545454545,
+      "grad_norm": 0.0035849846804439827,
+      "kl": 0.2314453125,
+      "learning_rate": 9.611771470522908e-05,
+      "loss": 0.0002,
+      "num_tokens": 118947.0,
+      "reward": 2.3909339904785156,
+      "reward_std": 0.0009617910836823285,
+      "rewards/check_gptzero_func": 0.8571428656578064,
+      "rewards/check_perplexity_diff_func": 1.0,
+      "rewards/check_winston_local_func": 0.533791184425354,
+      "step": 611
+    },
+    {
+      "clip_ratio": 0.00020614307140931487,
+      "epoch": 0.855944055944056,
+      "grad_norm": 0.0024010748168965985,
+      "kl": 0.22265625,
+      "learning_rate": 9.609410345776554e-05,
+      "loss": 0.0002,
+      "step": 612
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 97.92857360839844,
+      "epoch": 0.8573426573426574,
+      "grad_norm": 0.0165782655855821,
+      "kl": 0.220703125,
+      "learning_rate": 9.607042354647924e-05,
+      "loss": -0.0006,
+      "num_tokens": 138117.0,
+      "reward": 2.496706485748291,
+      "reward_std": 0.025315813720226288,
+      "rewards/check_gptzero_func": 0.9642857313156128,
+      "rewards/check_perplexity_diff_func": 0.9464285969734192,
+      "rewards/check_winston_local_func": 0.585992157459259,
+      "step": 613
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8587412587412587,
+      "grad_norm": 0.005231318150303852,
+      "kl": 0.220703125,
+      "learning_rate": 9.60466750066448e-05,
+      "loss": -0.0006,
+      "step": 614
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 116.71429443359375,
+      "epoch": 0.8601398601398601,
+      "grad_norm": 1.4205544083992568e-05,
+      "kl": 0.1533203125,
+      "learning_rate": 9.60228578736391e-05,
+      "loss": 0.0002,
+      "num_tokens": 159543.0,
+      "reward": 2.470890998840332,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 1.0,
+      "rewards/check_perplexity_diff_func": 1.0,
+      "rewards/check_winston_local_func": 0.47089076042175293,
+      "step": 615
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8615384615384616,
+      "grad_norm": 1.6303392641980344e-05,
+      "kl": 0.1533203125,
+      "learning_rate": 9.599897218294122e-05,
+      "loss": 0.0002,
+      "step": 616
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 112.50000762939453,
+      "epoch": 0.862937062937063,
+      "grad_norm": 1.6234044022820597e-07,
+      "kl": 0.2021484375,
+      "learning_rate": 9.597501797013233e-05,
+      "loss": 0.0002,
+      "num_tokens": 180563.0,
+      "reward": 2.09031081199646,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.6428571343421936,
+      "rewards/check_perplexity_diff_func": 0.9285714030265808,
+      "rewards/check_winston_local_func": 0.5188822746276855,
+      "step": 617
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8643356643356643,
+      "grad_norm": 1.666191215120993e-07,
+      "kl": 0.2021484375,
+      "learning_rate": 9.595099527089569e-05,
+      "loss": 0.0002,
+      "step": 618
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 108.78572082519531,
+      "epoch": 0.8657342657342657,
+      "grad_norm": 8.531816841675701e-07,
+      "kl": 0.193359375,
+      "learning_rate": 9.592690412101658e-05,
+      "loss": 0.0002,
+      "num_tokens": 201177.0,
+      "reward": 1.9570515155792236,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.5714285969734192,
+      "rewards/check_perplexity_diff_func": 1.0,
+      "rewards/check_winston_local_func": 0.38562270998954773,
+      "step": 619
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8671328671328671,
+      "grad_norm": 8.842725127740978e-07,
+      "kl": 0.193359375,
+      "learning_rate": 9.590274455638225e-05,
+      "loss": 0.0002,
+      "step": 620
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 138.07144165039062,
+      "epoch": 0.8685314685314686,
+      "grad_norm": 5.1580717328778724e-05,
+      "kl": 0.1640625,
+      "learning_rate": 9.587851661298186e-05,
+      "loss": 0.0002,
+      "num_tokens": 225661.0,
+      "reward": 2.487313747406006,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 1.0714285373687744,
+      "rewards/check_perplexity_diff_func": 0.9642857313156128,
+      "rewards/check_winston_local_func": 0.4515993893146515,
+      "step": 621
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8699300699300699,
+      "grad_norm": 5.097158350926591e-05,
+      "kl": 0.1640625,
+      "learning_rate": 9.585422032690643e-05,
+      "loss": 0.0002,
+      "step": 622
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 126.30357360839844,
+      "epoch": 0.8713286713286713,
+      "grad_norm": 1.5371764428374234e-05,
+      "kl": 0.1494140625,
+      "learning_rate": 9.582985573434877e-05,
+      "loss": 0.0002,
+      "num_tokens": 248756.0,
+      "reward": 2.386916160583496,
+      "reward_std": 0.0005263118073344231,
+      "rewards/check_gptzero_func": 0.8571428656578064,
+      "rewards/check_perplexity_diff_func": 1.0,
+      "rewards/check_winston_local_func": 0.5297732949256897,
+      "step": 623
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8727272727272727,
+      "grad_norm": 1.5410923964835553e-05,
+      "kl": 0.1494140625,
+      "learning_rate": 9.580542287160348e-05,
+      "loss": 0.0002,
+      "step": 624
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 121.85714721679688,
+      "epoch": 0.8741258741258742,
+      "grad_norm": 0.0002966067751134756,
+      "kl": 0.173828125,
+      "learning_rate": 9.578092177506683e-05,
+      "loss": 0.0002,
+      "num_tokens": 270514.0,
+      "reward": 2.7340447902679443,
+      "reward_std": 6.457499694079161e-05,
+      "rewards/check_gptzero_func": 1.2142857313156128,
+      "rewards/check_perplexity_diff_func": 0.9642857313156128,
+      "rewards/check_winston_local_func": 0.555473268032074,
+      "step": 625
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8755244755244755,
+      "grad_norm": 0.0002795708879844016,
+      "kl": 0.173828125,
+      "learning_rate": 9.575635248123675e-05,
+      "loss": 0.0002,
+      "step": 626
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 97.89286041259766,
+      "epoch": 0.8769230769230769,
+      "grad_norm": 1.0675950775787285e-07,
+      "kl": 0.1904296875,
+      "learning_rate": 9.573171502671273e-05,
+      "loss": 0.0002,
+      "num_tokens": 289588.0,
+      "reward": 2.467723846435547,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.9285714030265808,
+      "rewards/check_perplexity_diff_func": 1.0,
+      "rewards/check_winston_local_func": 0.5391523838043213,
+      "step": 627
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8783216783216783,
+      "grad_norm": 1.0758265762547126e-07,
+      "kl": 0.1904296875,
+      "learning_rate": 9.570700944819584e-05,
+      "loss": 0.0002,
+      "step": 628
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 126.96429443359375,
+      "epoch": 0.8797202797202798,
+      "grad_norm": 1.8189249055345895e-06,
+      "kl": 0.150390625,
+      "learning_rate": 9.568223578248858e-05,
+      "loss": 0.0002,
+      "num_tokens": 312792.0,
+      "reward": 2.4109601974487305,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.9285714030265808,
+      "rewards/check_perplexity_diff_func": 0.9285714030265808,
+      "rewards/check_winston_local_func": 0.5538173913955688,
+      "step": 629
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8811188811188811,
+      "grad_norm": 1.817429271106412e-06,
+      "kl": 0.150390625,
+      "learning_rate": 9.565739406649493e-05,
+      "loss": 0.0002,
+      "step": 630
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 109.64286041259766,
+      "epoch": 0.8825174825174825,
+      "grad_norm": 7.209254474831735e-05,
+      "kl": 0.169921875,
+      "learning_rate": 9.563248433722019e-05,
+      "loss": 0.0002,
+      "num_tokens": 333908.0,
+      "reward": 2.1215410232543945,
+      "reward_std": 0.0032710533123463392,
+      "rewards/check_gptzero_func": 0.7142857313156128,
+      "rewards/check_perplexity_diff_func": 0.9285714030265808,
+      "rewards/check_winston_local_func": 0.4786837100982666,
+      "step": 631
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8839160839160839,
+      "grad_norm": 0.005778816018850288,
+      "kl": 0.169921875,
+      "learning_rate": 9.560750663177101e-05,
+      "loss": 0.0002,
+      "step": 632
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 101.0714340209961,
+      "epoch": 0.8853146853146853,
+      "grad_norm": 5.141706041264495e-06,
+      "kl": 0.1689453125,
+      "learning_rate": 9.558246098735528e-05,
+      "loss": 0.0002,
+      "num_tokens": 353808.0,
+      "reward": 2.3030316829681396,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.7857142686843872,
+      "rewards/check_perplexity_diff_func": 0.9642857313156128,
+      "rewards/check_winston_local_func": 0.5530317425727844,
+      "step": 633
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8867132867132868,
+      "grad_norm": 5.171400787940402e-06,
+      "kl": 0.1689453125,
+      "learning_rate": 9.55573474412821e-05,
+      "loss": 0.0002,
+      "step": 634
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 112.0714340209961,
+      "epoch": 0.8881118881118881,
+      "grad_norm": 2.5579574578411833e-07,
+      "kl": 0.16015625,
+      "learning_rate": 9.553216603096176e-05,
+      "loss": 0.0002,
+      "num_tokens": 375024.0,
+      "reward": 2.2100911140441895,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.7857142686843872,
+      "rewards/check_perplexity_diff_func": 1.0,
+      "rewards/check_winston_local_func": 0.42437654733657837,
+      "step": 635
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8895104895104895,
+      "grad_norm": 3.163234894184058e-07,
+      "kl": 0.16015625,
+      "learning_rate": 9.550691679390558e-05,
+      "loss": 0.0002,
+      "step": 636
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 78.39286041259766,
+      "epoch": 0.8909090909090909,
+      "grad_norm": 2.7422578147300963e-06,
+      "kl": 0.19921875,
+      "learning_rate": 9.548159976772592e-05,
+      "loss": 0.0002,
+      "num_tokens": 391944.0,
+      "reward": 2.234971523284912,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.7142857313156128,
+      "rewards/check_perplexity_diff_func": 0.9285714030265808,
+      "rewards/check_winston_local_func": 0.5921144485473633,
+      "step": 637
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8923076923076924,
+      "grad_norm": 2.3669860027714333e-06,
+      "kl": 0.19921875,
+      "learning_rate": 9.54562149901362e-05,
+      "loss": 0.0002,
+      "step": 638
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 120.75000762939453,
+      "epoch": 0.8937062937062937,
+      "grad_norm": 2.3171625543487874e-07,
+      "kl": 0.158203125,
+      "learning_rate": 9.543076249895066e-05,
+      "loss": 0.0002,
+      "num_tokens": 413760.0,
+      "reward": 2.152951955795288,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.7142857313156128,
+      "rewards/check_perplexity_diff_func": 1.0,
+      "rewards/check_winston_local_func": 0.43866607546806335,
+      "step": 639
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8951048951048951,
+      "grad_norm": 2.4228268027328343e-07,
+      "kl": 0.158203125,
+      "learning_rate": 9.540524233208448e-05,
+      "loss": 0.0002,
+      "step": 640
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 87.05357360839844,
+      "epoch": 0.8965034965034965,
+      "grad_norm": 0.008398279317441937,
+      "kl": 0.181640625,
+      "learning_rate": 9.537965452755365e-05,
+      "loss": 0.0094,
+      "num_tokens": 431959.0,
+      "reward": 2.0755865573883057,
+      "reward_std": 0.019971024245023727,
+      "rewards/check_gptzero_func": 0.7142857313156128,
+      "rewards/check_perplexity_diff_func": 0.8571428656578064,
+      "rewards/check_winston_local_func": 0.5041577219963074,
+      "step": 641
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8979020979020979,
+      "grad_norm": 0.0032568280015785474,
+      "kl": 0.181640625,
+      "learning_rate": 9.535399912347489e-05,
+      "loss": 0.0094,
+      "step": 642
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 107.28572082519531,
+      "epoch": 0.8993006993006993,
+      "grad_norm": 4.5580158500525314e-05,
+      "kl": 0.150390625,
+      "learning_rate": 9.53282761580656e-05,
+      "loss": 0.0001,
+      "num_tokens": 452913.0,
+      "reward": 2.6801095008850098,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 1.1428571939468384,
+      "rewards/check_perplexity_diff_func": 0.9285714030265808,
+      "rewards/check_winston_local_func": 0.6086806058883667,
+      "step": 643
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.9006993006993007,
+      "grad_norm": 4.426477450510755e-05,
+      "kl": 0.150390625,
+      "learning_rate": 9.530248566964391e-05,
+      "loss": 0.0001,
+      "step": 644
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 105.25000762939453,
+      "epoch": 0.9020979020979021,
+      "grad_norm": 2.371230122251656e-05,
+      "kl": 0.1513671875,
+      "learning_rate": 9.527662769662841e-05,
+      "loss": 0.0002,
+      "num_tokens": 473141.0,
+      "reward": 2.1837356090545654,
+      "reward_std": 0.00018372414342593402,
+      "rewards/check_gptzero_func": 0.7142857313156128,
+      "rewards/check_perplexity_diff_func": 1.0,
+      "rewards/check_winston_local_func": 0.4694497585296631,
+      "step": 645
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.9034965034965035,
+      "grad_norm": 2.391891150952087e-05,
+      "kl": 0.1513671875,
+      "learning_rate": 9.525070227753834e-05,
+      "loss": 0.0002,
+      "step": 646
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 117.64286041259766,
+      "epoch": 0.9048951048951049,
+      "grad_norm": 1.3380347420537266e-07,
+      "kl": 0.150390625,
+      "learning_rate": 9.522470945099334e-05,
+      "loss": 0.0002,
+      "num_tokens": 494483.0,
+      "reward": 1.8219071626663208,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.5,
+      "rewards/check_perplexity_diff_func": 1.0,
+      "rewards/check_winston_local_func": 0.3219069838523865,
+      "step": 647
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.9062937062937063,
+      "grad_norm": 1.336307086326504e-07,
+      "kl": 0.150390625,
+      "learning_rate": 9.519864925571345e-05,
+      "loss": 0.0002,
+      "step": 648
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 131.0,
+      "epoch": 0.9076923076923077,
+      "grad_norm": 0.00012151914677644382,
+      "kl": 0.11474609375,
+      "learning_rate": 9.517252173051911e-05,
+      "loss": 0.0001,
+      "num_tokens": 518551.0,
+      "reward": 2.6524739265441895,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 1.2142857313156128,
+      "rewards/check_perplexity_diff_func": 0.8928571343421936,
+      "rewards/check_winston_local_func": 0.545330822467804,
+      "step": 649
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.9090909090909091,
+      "grad_norm": 5.8305084675906205e-05,
+      "kl": 0.11474609375,
+      "learning_rate": 9.514632691433107e-05,
+      "loss": 0.0001,
+      "step": 650
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 100.89286041259766,
+      "epoch": 0.9104895104895104,
+      "grad_norm": 5.230987317211898e-07,
+      "kl": 0.2119140625,
+      "learning_rate": 9.512006484617025e-05,
+      "loss": 0.0002,
+      "num_tokens": 538141.0,
+      "reward": 2.5491878986358643,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 1.0,
+      "rewards/check_perplexity_diff_func": 0.9642857313156128,
+      "rewards/check_winston_local_func": 0.5849019885063171,
+      "step": 651
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.9118881118881119,
+      "grad_norm": 5.241995408876847e-07,
+      "kl": 0.2119140625,
+      "learning_rate": 9.509373556515782e-05,
+      "loss": 0.0002,
+      "step": 652
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 87.41072082519531,
+      "epoch": 0.9132867132867133,
+      "grad_norm": 0.00010906936400705836,
+      "kl": 0.201171875,
+      "learning_rate": 9.506733911051502e-05,
+      "loss": 0.0002,
+      "num_tokens": 555810.0,
+      "reward": 2.1223483085632324,
+      "reward_std": 0.00045456583029590547,
+      "rewards/check_gptzero_func": 0.6428571343421936,
+      "rewards/check_perplexity_diff_func": 1.0,
+      "rewards/check_winston_local_func": 0.47949108481407166,
+      "step": 653
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.9146853146853147,
+      "grad_norm": 0.0001088298231648261,
+      "kl": 0.201171875,
+      "learning_rate": 9.50408755215632e-05,
+      "loss": 0.0002,
+      "step": 654
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 108.28572082519531,
+      "epoch": 0.916083916083916,
+      "grad_norm": 4.051397464310879e-07,
+      "kl": 0.16015625,
+      "learning_rate": 9.50143448377237e-05,
+      "loss": 0.0002,
+      "num_tokens": 576240.0,
+      "reward": 2.4261841773986816,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.9285714030265808,
+      "rewards/check_perplexity_diff_func": 0.9642857313156128,
+      "rewards/check_winston_local_func": 0.5333269238471985,
+      "step": 655
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.9174825174825175,
+      "grad_norm": 4.914401353891805e-07,
+      "kl": 0.16015625,
+      "learning_rate": 9.498774709851779e-05,
+      "loss": 0.0002,
+      "step": 656
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 103.35714721679688,
+      "epoch": 0.9188811188811189,
+      "grad_norm": 2.415453772573951e-07,
+      "kl": 0.1845703125,
+      "learning_rate": 9.496108234356666e-05,
+      "loss": 0.0002,
+      "num_tokens": 596032.0,
+      "reward": 2.0929291248321533,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.6428571343421936,
+      "rewards/check_perplexity_diff_func": 1.0,
+      "rewards/check_winston_local_func": 0.4500720500946045,
+      "step": 657
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.9202797202797203,
+      "grad_norm": 2.538209980883487e-07,
+      "kl": 0.1845703125,
+      "learning_rate": 9.493435061259131e-05,
+      "loss": 0.0002,
+      "step": 658
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 108.5714340209961,
+      "epoch": 0.9216783216783216,
+      "grad_norm": 0.0005472818191454855,
+      "kl": 0.15234375,
+      "learning_rate": 9.49075519454125e-05,
+      "loss": 0.0002,
+      "num_tokens": 616806.0,
+      "reward": 2.419811964035034,
+      "reward_std": 0.0021216755267232656,
+      "rewards/check_gptzero_func": 1.0,
+      "rewards/check_perplexity_diff_func": 0.9642857313156128,
+      "rewards/check_winston_local_func": 0.4555261731147766,
+      "step": 659
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.9230769230769231,
+      "grad_norm": 0.000555498769506101,
+      "kl": 0.15234375,
+      "learning_rate": 9.48806863819507e-05,
+      "loss": 0.0002,
+      "step": 660
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 99.50000762939453,
+      "epoch": 0.9244755244755245,
+      "grad_norm": 1.210593242712218e-05,
+      "kl": 0.185546875,
+      "learning_rate": 9.485375396222609e-05,
+      "loss": 0.0002,
+      "num_tokens": 636302.0,
+      "reward": 2.2581381797790527,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.7857142686843872,
+      "rewards/check_perplexity_diff_func": 0.9642857313156128,
+      "rewards/check_winston_local_func": 0.5081380605697632,
+      "step": 661
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.9258741258741259,
+      "grad_norm": 1.2079740491548147e-05,
+      "kl": 0.185546875,
+      "learning_rate": 9.482675472635837e-05,
+      "loss": 0.0002,
+      "step": 662
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 130.8928680419922,
+      "epoch": 0.9272727272727272,
+      "grad_norm": 6.087538896625874e-06,
+      "kl": 0.1640625,
+      "learning_rate": 9.479968871456679e-05,
+      "loss": 0.0002,
+      "num_tokens": 659494.0,
+      "reward": 2.4411208629608154,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 1.0,
+      "rewards/check_perplexity_diff_func": 0.9285714030265808,
+      "rewards/check_winston_local_func": 0.5125490427017212,
+      "step": 663
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.9286713286713286,
+      "grad_norm": 5.91002527093218e-06,
+      "kl": 0.1640625,
+      "learning_rate": 9.477255596717012e-05,
+      "loss": 0.0002,
+      "step": 664
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 106.78572082519531,
+      "epoch": 0.9300699300699301,
+      "grad_norm": 0.0006001697890594295,
+      "kl": 0.1572265625,
+      "learning_rate": 9.474535652458647e-05,
+      "loss": 0.0002,
+      "num_tokens": 679806.0,
+      "reward": 1.9322819709777832,
+      "reward_std": 0.0051432885229587555,
+      "rewards/check_gptzero_func": 0.5714285969734192,
+      "rewards/check_perplexity_diff_func": 0.9642857313156128,
+      "rewards/check_winston_local_func": 0.39656758308410645,
+      "step": 665
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.9314685314685315,
+      "grad_norm": 0.0006059012217339499,
+      "kl": 0.1572265625,
+      "learning_rate": 9.471809042733336e-05,
+      "loss": 0.0002,
+      "step": 666
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 93.46428680419922,
+      "epoch": 0.9328671328671329,
+      "grad_norm": 2.199744146458813e-06,
+      "kl": 0.197265625,
+      "learning_rate": 9.469075771602755e-05,
+      "loss": 0.0002,
+      "num_tokens": 698264.0,
+      "reward": 2.203263282775879,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.7142857313156128,
+      "rewards/check_perplexity_diff_func": 0.9642857313156128,
+      "rewards/check_winston_local_func": 0.524691641330719,
+      "step": 667
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.9342657342657342,
+      "grad_norm": 2.363657066653465e-06,
+      "kl": 0.197265625,
+      "learning_rate": 9.46633584313851e-05,
+      "loss": 0.0002,
+      "step": 668
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 121.28572082519531,
+      "epoch": 0.9356643356643357,
+      "grad_norm": 1.1463012847374795e-06,
+      "kl": 0.138671875,
+      "learning_rate": 9.463589261422116e-05,
+      "loss": 0.0001,
+      "num_tokens": 720488.0,
+      "reward": 2.178800582885742,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.8571428656578064,
+      "rewards/check_perplexity_diff_func": 0.9285714030265808,
+      "rewards/check_winston_local_func": 0.3930862247943878,
+      "step": 669
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.9370629370629371,
+      "grad_norm": 1.142739230420817e-06,
+      "kl": 0.138671875,
+      "learning_rate": 9.460836030545007e-05,
+      "loss": 0.0001,
+      "step": 670
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 110.00000762939453,
+      "epoch": 0.9384615384615385,
+      "grad_norm": 1.5883057563648785e-06,
+      "kl": 0.1748046875,
+      "learning_rate": 9.458076154608515e-05,
+      "loss": 0.0002,
+      "num_tokens": 741268.0,
+      "reward": 2.4054763317108154,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 1.0714285373687744,
+      "rewards/check_perplexity_diff_func": 0.8928571343421936,
+      "rewards/check_winston_local_func": 0.44119033217430115,
+      "step": 671
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.9398601398601398,
+      "grad_norm": 1.485387924830463e-06,
+      "kl": 0.1748046875,
+      "learning_rate": 9.455309637723875e-05,
+      "loss": 0.0002,
+      "step": 672
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 76.0,
+      "epoch": 0.9412587412587412,
+      "grad_norm": 5.476513075735238e-07,
+      "kl": 0.275390625,
+      "learning_rate": 9.452536484012212e-05,
+      "loss": 0.0003,
+      "num_tokens": 757900.0,
+      "reward": 2.1300675868988037,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.5714285969734192,
+      "rewards/check_perplexity_diff_func": 1.0,
+      "rewards/check_winston_local_func": 0.558638870716095,
+      "step": 673
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.9426573426573427,
+      "grad_norm": 5.390848738144119e-07,
+      "kl": 0.275390625,
+      "learning_rate": 9.44975669760454e-05,
+      "loss": 0.0003,
+      "step": 674
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 85.42857360839844,
+      "epoch": 0.9440559440559441,
+      "grad_norm": 0.09102872158611763,
+      "kl": 0.2197265625,
+      "learning_rate": 9.446970282641754e-05,
+      "loss": 0.0018,
+      "num_tokens": 775790.0,
+      "reward": 2.0888733863830566,
+      "reward_std": 0.06202990189194679,
+      "rewards/check_gptzero_func": 0.5357142686843872,
+      "rewards/check_perplexity_diff_func": 1.0,
+      "rewards/check_winston_local_func": 0.5531590580940247,
+      "step": 675
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.9454545454545454,
+      "grad_norm": 0.05645096724429603,
+      "kl": 0.220703125,
+      "learning_rate": 9.444177243274618e-05,
+      "loss": 0.0003,
+      "step": 676
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 109.25000762939453,
+      "epoch": 0.9468531468531468,
+      "grad_norm": 8.293674316919555e-06,
+      "kl": 0.1435546875,
+      "learning_rate": 9.441377583663768e-05,
+      "loss": 0.0001,
+      "num_tokens": 796550.0,
+      "reward": 2.231931209564209,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.8571428656578064,
+      "rewards/check_perplexity_diff_func": 0.9285714030265808,
+      "rewards/check_winston_local_func": 0.44621697068214417,
+      "step": 677
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.9482517482517483,
+      "grad_norm": 1.3063887978182609e-05,
+      "kl": 0.1435546875,
+      "learning_rate": 9.438571307979704e-05,
+      "loss": 0.0001,
+      "step": 678
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 116.10714721679688,
+      "epoch": 0.9496503496503497,
+      "grad_norm": 8.55167943307393e-05,
+      "kl": 0.171875,
+      "learning_rate": 9.435758420402778e-05,
+      "loss": 0.0002,
+      "num_tokens": 818208.0,
+      "reward": 2.5719223022460938,
+      "reward_std": 0.00022740935673937201,
+      "rewards/check_gptzero_func": 1.0714285373687744,
+      "rewards/check_perplexity_diff_func": 0.9642857313156128,
+      "rewards/check_winston_local_func": 0.5362080335617065,
+      "step": 679
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.951048951048951,
+      "grad_norm": 7.003771025309948e-05,
+      "kl": 0.171875,
+      "learning_rate": 9.43293892512319e-05,
+      "loss": 0.0002,
+      "step": 680
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 134.75,
+      "epoch": 0.9524475524475524,
+      "grad_norm": 0.000302463889717664,
+      "kl": 0.1435546875,
+      "learning_rate": 9.430112826340987e-05,
+      "loss": 0.0001,
+      "num_tokens": 841930.0,
+      "reward": 2.399691581726074,
+      "reward_std": 0.0017504910938441753,
+      "rewards/check_gptzero_func": 1.0714285373687744,
+      "rewards/check_perplexity_diff_func": 0.8928571343421936,
+      "rewards/check_winston_local_func": 0.4354057312011719,
+      "step": 681
+    },
+    {
+      "clip_ratio": 9.007386688608676e-05,
+      "epoch": 0.9538461538461539,
+      "grad_norm": 0.00031004535575776196,
+      "kl": 0.1435546875,
+      "learning_rate": 9.42728012826605e-05,
+      "loss": 0.0001,
+      "step": 682
+    },
+    {
+      "clip_ratio": 0.00011586143227759749,
+      "completion_length": 117.26786041259766,
+      "epoch": 0.9552447552447553,
+      "grad_norm": 0.0023255881114160465,
+      "kl": 0.146484375,
+      "learning_rate": 9.424440835118094e-05,
+      "loss": 0.0003,
+      "num_tokens": 863539.0,
+      "reward": 2.1542186737060547,
+      "reward_std": 0.02600925788283348,
+      "rewards/check_gptzero_func": 0.7857142686843872,
+      "rewards/check_perplexity_diff_func": 0.9464285969734192,
+      "rewards/check_winston_local_func": 0.42207565903663635,
+      "step": 683
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.9566433566433566,
+      "grad_norm": 0.003353044551999646,
+      "kl": 0.146484375,
+      "learning_rate": 9.421594951126653e-05,
+      "loss": 0.0003,
+      "step": 684
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 130.75,
+      "epoch": 0.958041958041958,
+      "grad_norm": 3.4853626956300013e-07,
+      "kl": 0.1376953125,
+      "learning_rate": 9.418742480531085e-05,
+      "loss": 0.0001,
+      "num_tokens": 886503.0,
+      "reward": 2.1976943016052246,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.7857142686843872,
+      "rewards/check_perplexity_diff_func": 0.9285714030265808,
+      "rewards/check_winston_local_func": 0.4834086000919342,
+      "step": 685
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.9594405594405594,
+      "grad_norm": 4.041143328184125e-07,
+      "kl": 0.1376953125,
+      "learning_rate": 9.415883427580556e-05,
+      "loss": 0.0001,
+      "step": 686
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 88.14286041259766,
+      "epoch": 0.9608391608391609,
+      "grad_norm": 0.0002925937022236145,
+      "kl": 0.1884765625,
+      "learning_rate": 9.413017796534038e-05,
+      "loss": 0.0002,
+      "num_tokens": 904263.0,
+      "reward": 2.0918164253234863,
+      "reward_std": 0.0009377764072269201,
+      "rewards/check_gptzero_func": 0.6428571343421936,
+      "rewards/check_perplexity_diff_func": 1.0,
+      "rewards/check_winston_local_func": 0.448959082365036,
+      "step": 687
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.9622377622377623,
+      "grad_norm": 0.00027650769429382713,
+      "kl": 0.1884765625,
+      "learning_rate": 9.410145591660301e-05,
+      "loss": 0.0002,
+      "step": 688
+    },
+    {
+      "clip_ratio": 0.00036945813917554915,
+      "completion_length": 111.60714721679688,
+      "epoch": 0.9636363636363636,
+      "grad_norm": 0.00019032746086675872,
+      "kl": 0.154296875,
+      "learning_rate": 9.407266817237911e-05,
+      "loss": 0.0002,
+      "num_tokens": 925343.0,
+      "reward": 2.2334210872650146,
+      "reward_std": 0.00015559412713628262,
+      "rewards/check_gptzero_func": 0.7142857313156128,
+      "rewards/check_perplexity_diff_func": 0.9642857313156128,
+      "rewards/check_winston_local_func": 0.5548495650291443,
+      "step": 689
+    },
+    {
+      "clip_ratio": 0.00024630542611703277,
+      "epoch": 0.965034965034965,
+      "grad_norm": 0.00013578009050933864,
+      "kl": 0.1533203125,
+      "learning_rate": 9.404381477555217e-05,
+      "loss": 0.0002,
+      "step": 690
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 113.14286041259766,
+      "epoch": 0.9664335664335665,
+      "grad_norm": 1.7240378740659336e-05,
+      "kl": 0.16015625,
+      "learning_rate": 9.401489576910349e-05,
+      "loss": 0.0002,
+      "num_tokens": 946805.0,
+      "reward": 2.417461395263672,
+      "reward_std": 0.00010864253272302449,
+      "rewards/check_gptzero_func": 0.9285714030265808,
+      "rewards/check_perplexity_diff_func": 0.9642857313156128,
+      "rewards/check_winston_local_func": 0.5246042609214783,
+      "step": 691
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.9678321678321679,
+      "grad_norm": 1.6290593947397482e-05,
+      "kl": 0.16015625,
+      "learning_rate": 9.398591119611211e-05,
+      "loss": 0.0002,
+      "step": 692
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 89.4464340209961,
+      "epoch": 0.9692307692307692,
+      "grad_norm": 0.0003060987829802528,
+      "kl": 0.19140625,
+      "learning_rate": 9.395686109975474e-05,
+      "loss": 0.0002,
+      "num_tokens": 965064.0,
+      "reward": 2.075838804244995,
+      "reward_std": 0.0029447702690958977,
+      "rewards/check_gptzero_func": 0.6428571343421936,
+      "rewards/check_perplexity_diff_func": 1.0,
+      "rewards/check_winston_local_func": 0.43298134207725525,
+      "step": 693
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.9706293706293706,
+      "grad_norm": 0.001874059602241207,
+      "kl": 0.19140625,
+      "learning_rate": 9.392774552330567e-05,
+      "loss": 0.0002,
+      "step": 694
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 126.10714721679688,
+      "epoch": 0.972027972027972,
+      "grad_norm": 8.509194894061606e-06,
+      "kl": 0.1640625,
+      "learning_rate": 9.38985645101368e-05,
+      "loss": 0.0002,
+      "num_tokens": 987280.0,
+      "reward": 2.1274285316467285,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.7857142686843872,
+      "rewards/check_perplexity_diff_func": 0.9285714030265808,
+      "rewards/check_winston_local_func": 0.41314274072647095,
+      "step": 695
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.9734265734265735,
+      "grad_norm": 8.720610779885554e-06,
+      "kl": 0.1640625,
+      "learning_rate": 9.386931810371742e-05,
+      "loss": 0.0002,
+      "step": 696
+    },
+    {
+      "clip_ratio": 0.00018198363250121474,
+      "completion_length": 104.37500762939453,
+      "epoch": 0.9748251748251748,
+      "grad_norm": 0.006044271030051565,
+      "kl": 0.162109375,
+      "learning_rate": 9.38400063476143e-05,
+      "loss": 0.0001,
+      "num_tokens": 1007329.0,
+      "reward": 2.1394481658935547,
+      "reward_std": 0.008249713107943535,
+      "rewards/check_gptzero_func": 0.7857142686843872,
+      "rewards/check_perplexity_diff_func": 0.9642857313156128,
+      "rewards/check_winston_local_func": 0.38944822549819946,
+      "step": 697
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.9762237762237762,
+      "grad_norm": 0.0001835404667097808,
+      "kl": 0.162109375,
+      "learning_rate": 9.381062928549153e-05,
+      "loss": 0.0001,
+      "step": 698
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 121.28572082519531,
+      "epoch": 0.9776223776223776,
+      "grad_norm": 2.5914502760266237e-06,
+      "kl": 0.1494140625,
+      "learning_rate": 9.378118696111047e-05,
+      "loss": 0.0001,
+      "num_tokens": 1029293.0,
+      "reward": 2.0580966472625732,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.6428571343421936,
+      "rewards/check_perplexity_diff_func": 1.0,
+      "rewards/check_winston_local_func": 0.4152393937110901,
+      "step": 699
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.9790209790209791,
+      "grad_norm": 2.773516504511405e-06,
+      "kl": 0.1494140625,
+      "learning_rate": 9.375167941832973e-05,
+      "loss": 0.0001,
+      "step": 700
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 2860,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 4,
+  "save_steps": 100,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 0.0,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/checkpoint-700/training_args.bin b/checkpoint-700/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..14e2d3ed154ee910c6c1698e855fc17a04eb6193
--- /dev/null
+++ b/checkpoint-700/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2c2b399ccd68443f851f396da81b1c9d32709e5be46a59f8f9eb00238e3b493d
+size 7480
diff --git a/checkpoint-700/zero_to_fp32.py b/checkpoint-700/zero_to_fp32.py
new file mode 100644
index 0000000000000000000000000000000000000000..0e759146cadd92ddfefab3680146c2bd6a2b5c04
--- /dev/null
+++ b/checkpoint-700/zero_to_fp32.py
@@ -0,0 +1,760 @@
+#!/usr/bin/env python
+
+# Copyright (c) Microsoft Corporation.
+# SPDX-License-Identifier: Apache-2.0
+
+# DeepSpeed Team
+
+# This script extracts fp32 consolidated weights from a zero 1, 2 and 3 DeepSpeed checkpoints. It gets
+# copied into the top level checkpoint dir, so the user can easily do the conversion at any point in
+# the future. Once extracted, the weights don't require DeepSpeed and can be used in any
+# application.
+#
+# example:
+#   python zero_to_fp32.py . output_dir/
+#   or
+#   python zero_to_fp32.py . output_dir/ --safe_serialization
+
+import argparse
+import torch
+import glob
+import math
+import os
+import re
+import gc
+import json
+import numpy as np
+from tqdm import tqdm
+from collections import OrderedDict
+from dataclasses import dataclass
+
+# while this script doesn't use deepspeed to recover data, since the checkpoints are pickled with
+# DeepSpeed data structures it has to be available in the current python environment.
+from deepspeed.utils import logger
+from deepspeed.checkpoint.constants import (DS_VERSION, OPTIMIZER_STATE_DICT, SINGLE_PARTITION_OF_FP32_GROUPS,
+                                            FP32_FLAT_GROUPS, ZERO_STAGE, PARTITION_COUNT, PARAM_SHAPES, BUFFER_NAMES,
+                                            FROZEN_PARAM_SHAPES, FROZEN_PARAM_FRAGMENTS)
+
+
+@dataclass
+class zero_model_state:
+    buffers: dict()
+    param_shapes: dict()
+    shared_params: list
+    ds_version: int
+    frozen_param_shapes: dict()
+    frozen_param_fragments: dict()
+
+
+debug = 0
+
+# load to cpu
+device = torch.device('cpu')
+
+
+def atoi(text):
+    return int(text) if text.isdigit() else text
+
+
+def natural_keys(text):
+    '''
+    alist.sort(key=natural_keys) sorts in human order
+    http://nedbatchelder.com/blog/200712/human_sorting.html
+    (See Toothy's implementation in the comments)
+    '''
+    return [atoi(c) for c in re.split(r'(\d+)', text)]
+
+
+def get_model_state_file(checkpoint_dir, zero_stage):
+    if not os.path.isdir(checkpoint_dir):
+        raise FileNotFoundError(f"Directory '{checkpoint_dir}' doesn't exist")
+
+    # there should be only one file
+    if zero_stage <= 2:
+        file = os.path.join(checkpoint_dir, "mp_rank_00_model_states.pt")
+    elif zero_stage == 3:
+        file = os.path.join(checkpoint_dir, "zero_pp_rank_0_mp_rank_00_model_states.pt")
+
+    if not os.path.exists(file):
+        raise FileNotFoundError(f"can't find model states file at '{file}'")
+
+    return file
+
+
+def get_checkpoint_files(checkpoint_dir, glob_pattern):
+    # XXX: need to test that this simple glob rule works for multi-node setup too
+    ckpt_files = sorted(glob.glob(os.path.join(checkpoint_dir, glob_pattern)), key=natural_keys)
+
+    if len(ckpt_files) == 0:
+        raise FileNotFoundError(f"can't find {glob_pattern} files in directory '{checkpoint_dir}'")
+
+    return ckpt_files
+
+
+def get_optim_files(checkpoint_dir):
+    return get_checkpoint_files(checkpoint_dir, "*_optim_states.pt")
+
+
+def get_model_state_files(checkpoint_dir):
+    return get_checkpoint_files(checkpoint_dir, "*_model_states.pt")
+
+
+def parse_model_states(files):
+    zero_model_states = []
+    for file in files:
+        state_dict = torch.load(file, map_location=device, weights_only=False)
+
+        if BUFFER_NAMES not in state_dict:
+            raise ValueError(f"{file} is not a model state checkpoint")
+        buffer_names = state_dict[BUFFER_NAMES]
+        if debug:
+            print("Found buffers:", buffer_names)
+
+        # recover just the buffers while restoring them to fp32 if they were saved in fp16
+        buffers = {k: v.float() for k, v in state_dict["module"].items() if k in buffer_names}
+        param_shapes = state_dict[PARAM_SHAPES]
+
+        # collect parameters that are included in param_shapes
+        param_names = []
+        for s in param_shapes:
+            for name in s.keys():
+                param_names.append(name)
+
+        # update with frozen parameters
+        frozen_param_shapes = state_dict.get(FROZEN_PARAM_SHAPES, None)
+        if frozen_param_shapes is not None:
+            if debug:
+                print(f"Found frozen_param_shapes: {frozen_param_shapes}")
+            param_names += list(frozen_param_shapes.keys())
+
+        # handle shared params
+        shared_params = [[k, v] for k, v in state_dict["shared_params"].items()]
+
+        ds_version = state_dict.get(DS_VERSION, None)
+
+        frozen_param_fragments = state_dict.get(FROZEN_PARAM_FRAGMENTS, None)
+
+        z_model_state = zero_model_state(buffers=buffers,
+                                         param_shapes=param_shapes,
+                                         shared_params=shared_params,
+                                         ds_version=ds_version,
+                                         frozen_param_shapes=frozen_param_shapes,
+                                         frozen_param_fragments=frozen_param_fragments)
+        zero_model_states.append(z_model_state)
+
+    return zero_model_states
+
+
+def parse_optim_states(files, ds_checkpoint_dir):
+    total_files = len(files)
+    state_dicts = []
+    for f in tqdm(files, desc='Loading checkpoint shards'):
+        state_dict = torch.load(f, map_location=device, mmap=True, weights_only=False)
+        # immediately discard the potentially huge 2 optimizer states as we only care for fp32 master weights
+        # and also handle the case where it was already removed by another helper script
+        state_dict["optimizer_state_dict"].pop("optimizer_state_dict", None)
+        state_dicts.append(state_dict)
+
+    if not ZERO_STAGE in state_dicts[0][OPTIMIZER_STATE_DICT]:
+        raise ValueError(f"{files[0]} is not a zero checkpoint")
+    zero_stage = state_dicts[0][OPTIMIZER_STATE_DICT][ZERO_STAGE]
+    world_size = state_dicts[0][OPTIMIZER_STATE_DICT][PARTITION_COUNT]
+
+    # For ZeRO-2 each param group can have different partition_count as data parallelism for expert
+    # parameters can be different from data parallelism for non-expert parameters. So we can just
+    # use the max of the partition_count to get the dp world_size.
+
+    if type(world_size) is list:
+        world_size = max(world_size)
+
+    if world_size != total_files:
+        raise ValueError(
+            f"Expected {world_size} of '*_optim_states.pt' under '{ds_checkpoint_dir}' but found {total_files} files. "
+            "Possibly due to an overwrite of an old checkpoint, or a checkpoint didn't get saved by one or more processes."
+        )
+
+    # the groups are named differently in each stage
+    if zero_stage <= 2:
+        fp32_groups_key = SINGLE_PARTITION_OF_FP32_GROUPS
+    elif zero_stage == 3:
+        fp32_groups_key = FP32_FLAT_GROUPS
+    else:
+        raise ValueError(f"unknown zero stage {zero_stage}")
+
+    fp32_flat_groups = [state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key] for i in range(len(state_dicts))]
+    return zero_stage, world_size, fp32_flat_groups
+
+
+def _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters):
+    """
+    Returns fp32 state_dict reconstructed from ds checkpoint
+
+    Args:
+        - ``ds_checkpoint_dir``: path to the deepspeed checkpoint folder (where the optimizer files are)
+
+    """
+    print(f"Processing zero checkpoint '{ds_checkpoint_dir}'")
+
+    optim_files = get_optim_files(ds_checkpoint_dir)
+    zero_stage, world_size, fp32_flat_groups = parse_optim_states(optim_files, ds_checkpoint_dir)
+    print(f"Detected checkpoint of type zero stage {zero_stage}, world_size: {world_size}")
+
+    model_files = get_model_state_files(ds_checkpoint_dir)
+
+    zero_model_states = parse_model_states(model_files)
+    print(f'Parsing checkpoint created by deepspeed=={zero_model_states[0].ds_version}')
+
+    if zero_stage <= 2:
+        return _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states,
+                                                          exclude_frozen_parameters)
+    elif zero_stage == 3:
+        return _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states,
+                                                          exclude_frozen_parameters)
+
+
+def _zero2_merge_frozen_params(state_dict, zero_model_states):
+    if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0:
+        return
+
+    frozen_param_shapes = zero_model_states[0].frozen_param_shapes
+    frozen_param_fragments = zero_model_states[0].frozen_param_fragments
+
+    if debug:
+        num_elem = sum(s.numel() for s in frozen_param_shapes.values())
+        print(f'rank 0: {FROZEN_PARAM_SHAPES}.numel = {num_elem}')
+
+        wanted_params = len(frozen_param_shapes)
+        wanted_numel = sum(s.numel() for s in frozen_param_shapes.values())
+        avail_numel = sum([p.numel() for p in frozen_param_fragments.values()])
+        print(f'Frozen params: Have {avail_numel} numels to process.')
+        print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params')
+
+    total_params = 0
+    total_numel = 0
+    for name, shape in frozen_param_shapes.items():
+        total_params += 1
+        unpartitioned_numel = shape.numel()
+        total_numel += unpartitioned_numel
+
+        state_dict[name] = frozen_param_fragments[name]
+
+        if debug:
+            print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ")
+
+    print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements")
+
+
+def _has_callable(obj, fn):
+    attr = getattr(obj, fn, None)
+    return callable(attr)
+
+
+def _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states):
+    param_shapes = zero_model_states[0].param_shapes
+
+    # Reconstruction protocol:
+    #
+    # XXX: document this
+
+    if debug:
+        for i in range(world_size):
+            for j in range(len(fp32_flat_groups[0])):
+                print(f"{FP32_FLAT_GROUPS}[{i}][{j}].shape={fp32_flat_groups[i][j].shape}")
+
+    # XXX: memory usage doubles here (zero2)
+    num_param_groups = len(fp32_flat_groups[0])
+    merged_single_partition_of_fp32_groups = []
+    for i in range(num_param_groups):
+        merged_partitions = [sd[i] for sd in fp32_flat_groups]
+        full_single_fp32_vector = torch.cat(merged_partitions, 0)
+        merged_single_partition_of_fp32_groups.append(full_single_fp32_vector)
+    avail_numel = sum(
+        [full_single_fp32_vector.numel() for full_single_fp32_vector in merged_single_partition_of_fp32_groups])
+
+    if debug:
+        wanted_params = sum([len(shapes) for shapes in param_shapes])
+        wanted_numel = sum([sum(shape.numel() for shape in shapes.values()) for shapes in param_shapes])
+        # not asserting if there is a mismatch due to possible padding
+        print(f"Have {avail_numel} numels to process.")
+        print(f"Need {wanted_numel} numels in {wanted_params} params.")
+
+    # params
+    # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support
+    # out-of-core computing solution
+    total_numel = 0
+    total_params = 0
+    for shapes, full_single_fp32_vector in zip(param_shapes, merged_single_partition_of_fp32_groups):
+        offset = 0
+        avail_numel = full_single_fp32_vector.numel()
+        for name, shape in shapes.items():
+
+            unpartitioned_numel = shape.numel() if _has_callable(shape, 'numel') else math.prod(shape)
+            total_numel += unpartitioned_numel
+            total_params += 1
+
+            if debug:
+                print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ")
+            state_dict[name] = full_single_fp32_vector.narrow(0, offset, unpartitioned_numel).view(shape)
+            offset += unpartitioned_numel
+
+        # Z2 started to align to 2*world_size to improve nccl performance. Therefore both offset and
+        # avail_numel can differ by anywhere between 0..2*world_size. Due to two unrelated complex
+        # paddings performed in the code it's almost impossible to predict the exact numbers w/o the
+        # live optimizer object, so we are checking that the numbers are within the right range
+        align_to = 2 * world_size
+
+        def zero2_align(x):
+            return align_to * math.ceil(x / align_to)
+
+        if debug:
+            print(f"original offset={offset}, avail_numel={avail_numel}")
+
+        offset = zero2_align(offset)
+        avail_numel = zero2_align(avail_numel)
+
+        if debug:
+            print(f"aligned  offset={offset}, avail_numel={avail_numel}")
+
+        # Sanity check
+        if offset != avail_numel:
+            raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong")
+
+    print(f"Reconstructed fp32 state dict with {total_params} params {total_numel} elements")
+
+
+def _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states,
+                                               exclude_frozen_parameters):
+    state_dict = OrderedDict()
+
+    # buffers
+    buffers = zero_model_states[0].buffers
+    state_dict.update(buffers)
+    if debug:
+        print(f"added {len(buffers)} buffers")
+
+    if not exclude_frozen_parameters:
+        _zero2_merge_frozen_params(state_dict, zero_model_states)
+
+    _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states)
+
+    # recover shared parameters
+    for pair in zero_model_states[0].shared_params:
+        if pair[1] in state_dict:
+            state_dict[pair[0]] = state_dict[pair[1]]
+
+    return state_dict
+
+
+def zero3_partitioned_param_info(unpartitioned_numel, world_size):
+    remainder = unpartitioned_numel % world_size
+    padding_numel = (world_size - remainder) if remainder else 0
+    partitioned_numel = math.ceil(unpartitioned_numel / world_size)
+    return partitioned_numel, padding_numel
+
+
+def _zero3_merge_frozen_params(state_dict, world_size, zero_model_states):
+    if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0:
+        return
+
+    if debug:
+        for i in range(world_size):
+            num_elem = sum(s.numel() for s in zero_model_states[i].frozen_param_fragments.values())
+            print(f'rank {i}: {FROZEN_PARAM_SHAPES}.numel = {num_elem}')
+
+        frozen_param_shapes = zero_model_states[0].frozen_param_shapes
+        wanted_params = len(frozen_param_shapes)
+        wanted_numel = sum(s.numel() for s in frozen_param_shapes.values())
+        avail_numel = sum([p.numel() for p in zero_model_states[0].frozen_param_fragments.values()]) * world_size
+        print(f'Frozen params: Have {avail_numel} numels to process.')
+        print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params')
+
+    total_params = 0
+    total_numel = 0
+    for name, shape in zero_model_states[0].frozen_param_shapes.items():
+        total_params += 1
+        unpartitioned_numel = shape.numel()
+        total_numel += unpartitioned_numel
+
+        param_frags = tuple(model_state.frozen_param_fragments[name] for model_state in zero_model_states)
+        state_dict[name] = torch.cat(param_frags, 0).narrow(0, 0, unpartitioned_numel).view(shape)
+
+        partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size)
+
+        if debug:
+            print(
+                f"Frozen params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}"
+            )
+
+    print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements")
+
+
+class GatheredTensor:
+    """
+    A pseudo tensor that collects partitioned weights.
+    It is more memory efficient when there are multiple groups.
+    """
+
+    def __init__(self, flat_groups, flat_groups_offset, offset, partitioned_numel, shape):
+        self.flat_groups = flat_groups
+        self.flat_groups_offset = flat_groups_offset
+        self.offset = offset
+        self.partitioned_numel = partitioned_numel
+        self.shape = shape
+        self.dtype = self.flat_groups[0][0].dtype
+
+    def contiguous(self):
+        """
+        Merge partitioned weights from flat_groups into a single tensor.
+        """
+        end_idx = self.offset + self.partitioned_numel
+        world_size = len(self.flat_groups)
+        pad_flat_param_chunks = []
+
+        for rank_i in range(world_size):
+            # for each rank, we need to collect weights from related group/groups
+            flat_groups_at_rank_i = self.flat_groups[rank_i]
+            start_group_id = None
+            end_group_id = None
+            for group_id in range(len(self.flat_groups_offset)):
+                if self.flat_groups_offset[group_id] <= self.offset < self.flat_groups_offset[group_id + 1]:
+                    start_group_id = group_id
+                if self.flat_groups_offset[group_id] < end_idx <= self.flat_groups_offset[group_id + 1]:
+                    end_group_id = group_id
+                    break
+            # collect weights from related group/groups
+            for group_id in range(start_group_id, end_group_id + 1):
+                flat_tensor = flat_groups_at_rank_i[group_id]
+                start_offset = self.offset - self.flat_groups_offset[group_id]
+                end_offset = min(end_idx, self.flat_groups_offset[group_id + 1]) - self.flat_groups_offset[group_id]
+                pad_flat_param_chunks.append(flat_tensor[start_offset:end_offset])
+
+        # collect weights from all ranks
+        pad_flat_param = torch.cat(pad_flat_param_chunks, dim=0)
+        param = pad_flat_param[:self.shape.numel()].view(self.shape).contiguous()
+        return param
+
+
+def _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states):
+    param_shapes = zero_model_states[0].param_shapes
+    avail_numel = sum([flat_group.numel() for flat_group in fp32_flat_groups[0]]) * world_size
+
+    # Reconstruction protocol: For zero3 we need to zip the partitions together at boundary of each
+    # param, re-consolidating each param, while dealing with padding if any
+
+    # merge list of dicts, preserving order
+    param_shapes = {k: v for d in param_shapes for k, v in d.items()}
+
+    if debug:
+        for i in range(world_size):
+            print(f"{FP32_FLAT_GROUPS}[{i}].shape={fp32_flat_groups[i].shape}")
+
+        wanted_params = len(param_shapes)
+        wanted_numel = sum(shape.numel() for shape in param_shapes.values())
+        # not asserting if there is a mismatch due to possible padding
+        avail_numel = fp32_flat_groups[0].numel() * world_size
+        print(f"Trainable params: Have {avail_numel} numels to process.")
+        print(f"Trainable params: Need {wanted_numel} numels in {wanted_params} params.")
+
+    # params
+    # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support
+    # out-of-core computing solution
+    offset = 0
+    total_numel = 0
+    total_params = 0
+    flat_groups_offset = [0] + list(np.cumsum([flat_tensor.numel() for flat_tensor in fp32_flat_groups[0]]))
+    for name, shape in tqdm(param_shapes.items(), desc='Gathering sharded weights'):
+        unpartitioned_numel = shape.numel()
+        total_numel += unpartitioned_numel
+        total_params += 1
+        partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size)
+
+        if debug:
+            print(
+                f"Trainable params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}"
+            )
+
+        # memory efficient tensor
+        tensor = GatheredTensor(fp32_flat_groups, flat_groups_offset, offset, partitioned_numel, shape)
+        state_dict[name] = tensor
+        offset += partitioned_numel
+
+    offset *= world_size
+
+    # Sanity check
+    if offset != avail_numel:
+        raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong")
+
+    print(f"Reconstructed Trainable fp32 state dict with {total_params} params {total_numel} elements")
+
+
+def _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states,
+                                               exclude_frozen_parameters):
+    state_dict = OrderedDict()
+
+    # buffers
+    buffers = zero_model_states[0].buffers
+    state_dict.update(buffers)
+    if debug:
+        print(f"added {len(buffers)} buffers")
+
+    if not exclude_frozen_parameters:
+        _zero3_merge_frozen_params(state_dict, world_size, zero_model_states)
+
+    _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states)
+
+    # recover shared parameters
+    for pair in zero_model_states[0].shared_params:
+        if pair[1] in state_dict:
+            state_dict[pair[0]] = state_dict[pair[1]]
+
+    return state_dict
+
+
+def to_torch_tensor(state_dict, return_empty_tensor=False):
+    """
+    Convert state_dict of GatheredTensor to torch tensor
+    """
+    torch_state_dict = {}
+    converted_tensors = {}
+    for name, tensor in state_dict.items():
+        tensor_id = id(tensor)
+        if tensor_id in converted_tensors:  # shared tensors
+            shared_tensor = torch_state_dict[converted_tensors[tensor_id]]
+            torch_state_dict[name] = shared_tensor
+        else:
+            converted_tensors[tensor_id] = name
+            if return_empty_tensor:
+                torch_state_dict[name] = torch.empty(tensor.shape, dtype=tensor.dtype)
+            else:
+                torch_state_dict[name] = tensor.contiguous()
+    return torch_state_dict
+
+
+def get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir,
+                                             tag=None,
+                                             exclude_frozen_parameters=False,
+                                             lazy_mode=False):
+    """
+    Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated state_dict that can be loaded with
+    ``load_state_dict()`` and used for training without DeepSpeed or shared with others, for example
+    via a model hub.
+
+    Args:
+        - ``checkpoint_dir``: path to the desired checkpoint folder
+        - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in 'latest' file. e.g., ``global_step14``
+        - ``exclude_frozen_parameters``: exclude frozen parameters
+        - ``lazy_mode``: get state_dict in lazy mode. It returns a dict of pesduo tensor instead of torch tensor, which is more memory efficient.
+          Convert the pesduo tensor to torch tensor by ``.contiguous()``
+
+    Returns:
+        - pytorch ``state_dict``
+
+    A typical usage might be ::
+
+        from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint
+        # do the training and checkpoint saving
+        state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir) # already on cpu
+        model = model.cpu() # move to cpu
+        model.load_state_dict(state_dict)
+        # submit to model hub or save the model to share with others
+
+    In this example the ``model`` will no longer be usable in the deepspeed context of the same
+    application. i.e. you will need to re-initialize the deepspeed engine, since
+    ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it.
+
+    If you want it all done for you, use ``load_state_dict_from_zero_checkpoint`` instead.
+
+    Note: the above usage may not work if your application doesn't have sufficient free CPU memory.
+    You may need to use the offline approach using the ``zero_to_fp32.py`` script that is saved with
+    the checkpoint. Or you can load state_dict in lazy mode ::
+
+        from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint
+        state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, lazy_mode=True) # not on cpu
+        for name, lazy_tensor in state_dict.item():
+            tensor = lazy_tensor.contiguous()  # to cpu
+            print(name, tensor)
+            # del tensor to release memory if it no longer in use
+    """
+    if tag is None:
+        latest_path = os.path.join(checkpoint_dir, 'latest')
+        if os.path.isfile(latest_path):
+            with open(latest_path, 'r') as fd:
+                tag = fd.read().strip()
+        else:
+            raise ValueError(f"Unable to find 'latest' file at {latest_path}")
+
+    ds_checkpoint_dir = os.path.join(checkpoint_dir, tag)
+
+    if not os.path.isdir(ds_checkpoint_dir):
+        raise FileNotFoundError(f"Directory '{ds_checkpoint_dir}' doesn't exist")
+
+    state_dict = _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters)
+    if lazy_mode:
+        return state_dict
+    else:
+        return to_torch_tensor(state_dict)
+
+
+def convert_zero_checkpoint_to_fp32_state_dict(checkpoint_dir,
+                                               output_dir,
+                                               max_shard_size="5GB",
+                                               safe_serialization=False,
+                                               tag=None,
+                                               exclude_frozen_parameters=False):
+    """
+    Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` file that can be
+    loaded with ``torch.load(file)`` + ``load_state_dict()`` and used for training without DeepSpeed.
+
+    Args:
+        - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``)
+        - ``output_dir``: directory to the pytorch fp32 state_dict output files
+        - ``max_shard_size``: the maximum size for a checkpoint before being sharded, default value is 5GB
+        - ``safe_serialization``:  whether to save the model using `safetensors` or the traditional PyTorch way (that uses `pickle`).
+        - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14``
+        - ``exclude_frozen_parameters``: exclude frozen parameters
+    """
+
+    # Dependency pre-check
+    if safe_serialization:
+        try:
+            from safetensors.torch import save_file
+        except ImportError:
+            print('If you want to use `safe_serialization`, please `pip install safetensors`')
+            raise
+    if max_shard_size is not None:
+        try:
+            from huggingface_hub import split_torch_state_dict_into_shards
+        except ImportError:
+            print('If you want to use `max_shard_size`, please `pip install huggingface_hub`')
+            raise
+
+    # Convert zero checkpoint to state_dict
+    state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir,
+                                                          tag,
+                                                          exclude_frozen_parameters,
+                                                          lazy_mode=True)
+
+    # Shard the model if it is too big.
+    weights_name = "model.safetensors" if safe_serialization else "pytorch_model.bin"
+    if max_shard_size is not None:
+        filename_pattern = weights_name.replace(".bin", "{suffix}.bin").replace(".safetensors", "{suffix}.safetensors")
+        # an memory-efficient approach for sharding
+        empty_state_dict = to_torch_tensor(state_dict, return_empty_tensor=True)
+        state_dict_split = split_torch_state_dict_into_shards(empty_state_dict,
+                                                              filename_pattern=filename_pattern,
+                                                              max_shard_size=max_shard_size)
+    else:
+        from collections import namedtuple
+        StateDictSplit = namedtuple("StateDictSplit", ["is_sharded", "filename_to_tensors"])
+        state_dict_split = StateDictSplit(is_sharded=False,
+                                          filename_to_tensors={weights_name: list(state_dict.keys())})
+
+    # Save the model by shard
+    os.makedirs(output_dir, exist_ok=True)
+    filename_to_tensors = state_dict_split.filename_to_tensors.items()
+    for shard_file, tensors in tqdm(filename_to_tensors, desc="Saving checkpoint shards"):
+        shard_state_dict = {tensor_name: state_dict[tensor_name] for tensor_name in tensors}
+        shard_state_dict = to_torch_tensor(shard_state_dict)
+        output_path = os.path.join(output_dir, shard_file)
+        if safe_serialization:
+            save_file(shard_state_dict, output_path, metadata={"format": "pt"})
+        else:
+            torch.save(shard_state_dict, output_path)
+        # release the memory of current shard
+        for tensor_name in list(shard_state_dict.keys()):
+            del state_dict[tensor_name]
+            del shard_state_dict[tensor_name]
+        del shard_state_dict
+        gc.collect()
+
+    # Save index if sharded
+    if state_dict_split.is_sharded:
+        index = {
+            "metadata": state_dict_split.metadata,
+            "weight_map": state_dict_split.tensor_to_filename,
+        }
+        save_index_file = "model.safetensors.index.json" if safe_serialization else "pytorch_model.bin.index.json"
+        save_index_file = os.path.join(output_dir, save_index_file)
+        with open(save_index_file, "w", encoding="utf-8") as f:
+            content = json.dumps(index, indent=2, sort_keys=True) + "\n"
+            f.write(content)
+
+
+def load_state_dict_from_zero_checkpoint(model, checkpoint_dir, tag=None):
+    """
+    1. Put the provided model to cpu
+    2. Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict``
+    3. Load it into the provided model
+
+    Args:
+        - ``model``: the model object to update
+        - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``)
+        - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14``
+
+    Returns:
+        - ``model`: modified model
+
+    Make sure you have plenty of CPU memory available before you call this function. If you don't
+    have enough use the ``zero_to_fp32.py`` utility to do the conversion. You will find it
+    conveniently placed for you in the checkpoint folder.
+
+    A typical usage might be ::
+
+        from deepspeed.utils.zero_to_fp32 import load_state_dict_from_zero_checkpoint
+        model = load_state_dict_from_zero_checkpoint(trainer.model, checkpoint_dir)
+        # submit to model hub or save the model to share with others
+
+    Note, that once this was run, the ``model`` will no longer be usable in the deepspeed context
+    of the same application. i.e. you will need to re-initialize the deepspeed engine, since
+    ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it.
+
+    """
+    logger.info(f"Extracting fp32 weights")
+    state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag)
+
+    logger.info(f"Overwriting model with fp32 weights")
+    model = model.cpu()
+    model.load_state_dict(state_dict, strict=False)
+
+    return model
+
+
+if __name__ == "__main__":
+    parser = argparse.ArgumentParser()
+    parser.add_argument("checkpoint_dir",
+                        type=str,
+                        help="path to the desired checkpoint folder, e.g., path/checkpoint-12")
+    parser.add_argument("output_dir",
+                        type=str,
+                        help="directory to the pytorch fp32 state_dict output files"
+                        "(e.g. path/checkpoint-12-output/)")
+    parser.add_argument(
+        "--max_shard_size",
+        type=str,
+        default="5GB",
+        help="The maximum size for a checkpoint before being sharded. Checkpoints shard will then be each of size"
+        "lower than this size. If expressed as a string, needs to be digits followed by a unit (like `5MB`"
+        "We default it to 5GB in order for models to be able to run easily on free-tier google colab instances"
+        "without CPU OOM issues.")
+    parser.add_argument(
+        "--safe_serialization",
+        default=False,
+        action='store_true',
+        help="Whether to save the model using `safetensors` or the traditional PyTorch way (that uses `pickle`).")
+    parser.add_argument("-t",
+                        "--tag",
+                        type=str,
+                        default=None,
+                        help="checkpoint tag used as a unique identifier for checkpoint. e.g., global_step1")
+    parser.add_argument("--exclude_frozen_parameters", action='store_true', help="exclude frozen parameters")
+    parser.add_argument("-d", "--debug", action='store_true', help="enable debug")
+    args = parser.parse_args()
+
+    debug = args.debug
+
+    convert_zero_checkpoint_to_fp32_state_dict(args.checkpoint_dir,
+                                               args.output_dir,
+                                               max_shard_size=args.max_shard_size,
+                                               safe_serialization=args.safe_serialization,
+                                               tag=args.tag,
+                                               exclude_frozen_parameters=args.exclude_frozen_parameters)
diff --git a/checkpoint-800/README.md b/checkpoint-800/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..85f3785e3148e4fdfa69be9cd1528c3af1891702
--- /dev/null
+++ b/checkpoint-800/README.md
@@ -0,0 +1,202 @@
+---
+base_model: mistralai/Mistral-Nemo-Instruct-2407
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.14.0
\ No newline at end of file
diff --git a/checkpoint-800/adapter_config.json b/checkpoint-800/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..1a10cbd48cf122bc7508a7cd9c23007d932168d0
--- /dev/null
+++ b/checkpoint-800/adapter_config.json
@@ -0,0 +1,37 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "mistralai/Mistral-Nemo-Instruct-2407",
+  "bias": "none",
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 16,
+  "lora_bias": false,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 16,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "k_proj",
+    "down_proj",
+    "q_proj",
+    "v_proj",
+    "o_proj",
+    "up_proj",
+    "gate_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/checkpoint-800/adapter_model.safetensors b/checkpoint-800/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..cfb1b803873a8e711c98ac42ceb4a4801c9584a5
--- /dev/null
+++ b/checkpoint-800/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:73b00beb0a2a98a95c5a8b294da2966ce324997cdf97c70149685b572a90ac9b
+size 114107976
diff --git a/checkpoint-800/global_step800/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt b/checkpoint-800/global_step800/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..7a5e83aa77f4fa67b19dc066ba4f17092551bb08
--- /dev/null
+++ b/checkpoint-800/global_step800/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7b26046f07fde102454f4263c0b0a4f60fe671a53bbbf646cc207916d9910e95
+size 49133072
diff --git a/checkpoint-800/global_step800/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt b/checkpoint-800/global_step800/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..cb17f69259859ecb811a9e3289fc8023b0f0269f
--- /dev/null
+++ b/checkpoint-800/global_step800/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4a570e70daa42bb41ccd66dc3b492a660e0abc05e1bb547236e5d51ee5e7c419
+size 49133072
diff --git a/checkpoint-800/global_step800/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt b/checkpoint-800/global_step800/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..f0248eae0ebef8ba991cd9907725b32fa91b4a31
--- /dev/null
+++ b/checkpoint-800/global_step800/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3b4e20de0442ace9aa50073afa4baa19bf4fd4b78bc06369fe48882d3af7c6f1
+size 49133072
diff --git a/checkpoint-800/global_step800/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt b/checkpoint-800/global_step800/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..60c37d25f83ab552e4c4b7595efe2177bd13b30d
--- /dev/null
+++ b/checkpoint-800/global_step800/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7c2ceefb59874a6e2a48fc9a765d78508cf1869e6e26bc6f209c910ed59ceb28
+size 49133072
diff --git a/checkpoint-800/global_step800/bf16_zero_pp_rank_4_mp_rank_00_optim_states.pt b/checkpoint-800/global_step800/bf16_zero_pp_rank_4_mp_rank_00_optim_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..5bb44e906bb455c17115178b29f2679e40a32288
--- /dev/null
+++ b/checkpoint-800/global_step800/bf16_zero_pp_rank_4_mp_rank_00_optim_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9b9a025e7c5463d423e8883458b2b257bdc35d8af47dc77e638827e60c7d420c
+size 49133072
diff --git a/checkpoint-800/global_step800/bf16_zero_pp_rank_5_mp_rank_00_optim_states.pt b/checkpoint-800/global_step800/bf16_zero_pp_rank_5_mp_rank_00_optim_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..f305d593e0cbca5de98f0b25c4b6ce170cfb3cb4
--- /dev/null
+++ b/checkpoint-800/global_step800/bf16_zero_pp_rank_5_mp_rank_00_optim_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:16dc57d5a41d1ae5cec317d24131d5e55f4f2483f25985285a1ea727c5e0bc97
+size 49133072
diff --git a/checkpoint-800/global_step800/bf16_zero_pp_rank_6_mp_rank_00_optim_states.pt b/checkpoint-800/global_step800/bf16_zero_pp_rank_6_mp_rank_00_optim_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..760cc44d2d254c21cf6c0b2e4c235d5de4b0eeeb
--- /dev/null
+++ b/checkpoint-800/global_step800/bf16_zero_pp_rank_6_mp_rank_00_optim_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b7be91b8c3a9ca22aa663e112ca964fdee154ef3139ceecaa41fa540fe7ddf77
+size 49133072
diff --git a/checkpoint-800/global_step800/zero_pp_rank_0_mp_rank_00_model_states.pt b/checkpoint-800/global_step800/zero_pp_rank_0_mp_rank_00_model_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..5dc6f68df4571e39f741c3f727650f9352a84600
--- /dev/null
+++ b/checkpoint-800/global_step800/zero_pp_rank_0_mp_rank_00_model_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:96760936d38a88a62c8ad219c2bc7d81f193d7f50939661069693d59cecff5da
+size 553886
diff --git a/checkpoint-800/global_step800/zero_pp_rank_1_mp_rank_00_model_states.pt b/checkpoint-800/global_step800/zero_pp_rank_1_mp_rank_00_model_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..89e7fef2112c4a81db40cd9831dd049c7015bc0d
--- /dev/null
+++ b/checkpoint-800/global_step800/zero_pp_rank_1_mp_rank_00_model_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9246a2c101f91d745e8648e7f026f47fe8f4a3f50345ef4f688416287b7b94d2
+size 553886
diff --git a/checkpoint-800/global_step800/zero_pp_rank_2_mp_rank_00_model_states.pt b/checkpoint-800/global_step800/zero_pp_rank_2_mp_rank_00_model_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..f4ffc6642805178f2048f16cac7a1957b0243905
--- /dev/null
+++ b/checkpoint-800/global_step800/zero_pp_rank_2_mp_rank_00_model_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5d38d79b164fa84c7fda96d1ab82f3f0bcbe6ae5c7a3fd774fbeb8ca411065a6
+size 553886
diff --git a/checkpoint-800/global_step800/zero_pp_rank_3_mp_rank_00_model_states.pt b/checkpoint-800/global_step800/zero_pp_rank_3_mp_rank_00_model_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..15cec059e25ed1f8c32aafcc2c0a6c0065c66469
--- /dev/null
+++ b/checkpoint-800/global_step800/zero_pp_rank_3_mp_rank_00_model_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cad520959b7534f6d5dc2186afb3dd2327d5ed04a91d63f85880a2d91f28c963
+size 553886
diff --git a/checkpoint-800/global_step800/zero_pp_rank_4_mp_rank_00_model_states.pt b/checkpoint-800/global_step800/zero_pp_rank_4_mp_rank_00_model_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..28d6b4167e51d65c1141a0dc63e655cbf530f092
--- /dev/null
+++ b/checkpoint-800/global_step800/zero_pp_rank_4_mp_rank_00_model_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b4f7d9b218649bdf32423aa2c31863a47f59a3f46d46b09c5ed7cf5300238bad
+size 553886
diff --git a/checkpoint-800/global_step800/zero_pp_rank_5_mp_rank_00_model_states.pt b/checkpoint-800/global_step800/zero_pp_rank_5_mp_rank_00_model_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..a3d62863f8c632913264849a75096efc4ac9df88
--- /dev/null
+++ b/checkpoint-800/global_step800/zero_pp_rank_5_mp_rank_00_model_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:86f5618481ad6652077d250306711b2cd16dd6c4784e829138e968f711b4c84a
+size 553886
diff --git a/checkpoint-800/global_step800/zero_pp_rank_6_mp_rank_00_model_states.pt b/checkpoint-800/global_step800/zero_pp_rank_6_mp_rank_00_model_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..8764635f005edfc92ea6be6765c30de007ec33d1
--- /dev/null
+++ b/checkpoint-800/global_step800/zero_pp_rank_6_mp_rank_00_model_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d59630dfc6a6dfd8440e8cdbf16eec84ef368633f92fb763a3e1292b52d3986b
+size 553886
diff --git a/checkpoint-800/latest b/checkpoint-800/latest
new file mode 100644
index 0000000000000000000000000000000000000000..57729c0be88118cbd582c8c68b4149cee821f0b4
--- /dev/null
+++ b/checkpoint-800/latest
@@ -0,0 +1 @@
+global_step800
\ No newline at end of file
diff --git a/checkpoint-800/rng_state_0.pth b/checkpoint-800/rng_state_0.pth
new file mode 100644
index 0000000000000000000000000000000000000000..6b1f09307692c6e429c91cda1ba0ab10a002eac3
--- /dev/null
+++ b/checkpoint-800/rng_state_0.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:99f50929a61fb67423f55c0f33b6ab61d16618c7d9bb8d09b72649bbd1c3f6b0
+size 15728
diff --git a/checkpoint-800/rng_state_1.pth b/checkpoint-800/rng_state_1.pth
new file mode 100644
index 0000000000000000000000000000000000000000..65dd17a88557fcabe6d87078df0e570c117c82f7
--- /dev/null
+++ b/checkpoint-800/rng_state_1.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:84df6327598e45bac29f3d915ab5870f61d9660b559bad90dd6f629759d60c64
+size 15728
diff --git a/checkpoint-800/rng_state_2.pth b/checkpoint-800/rng_state_2.pth
new file mode 100644
index 0000000000000000000000000000000000000000..f855b21a9431d7188470bce7a2940762582804b0
--- /dev/null
+++ b/checkpoint-800/rng_state_2.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5e6cfab048ec259364b3c3b25baadc74ed32842f05ef4cfac30c820104ea16ae
+size 15728
diff --git a/checkpoint-800/rng_state_3.pth b/checkpoint-800/rng_state_3.pth
new file mode 100644
index 0000000000000000000000000000000000000000..762c6ce968a281a8a08c28f257cc3a89c76a8f0c
--- /dev/null
+++ b/checkpoint-800/rng_state_3.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ba78a3331f9065562d24d54bdc54d0927b279f1009d1c5a717152a0ee9a8d0b8
+size 15792
diff --git a/checkpoint-800/rng_state_4.pth b/checkpoint-800/rng_state_4.pth
new file mode 100644
index 0000000000000000000000000000000000000000..839f35acf9087ed704c8cf812fbf59b6dd37543e
--- /dev/null
+++ b/checkpoint-800/rng_state_4.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9580b65704ce4e0eaebaeaf73b48e92173ca8082f2bf3bf0e6c16181155e95c4
+size 15728
diff --git a/checkpoint-800/rng_state_5.pth b/checkpoint-800/rng_state_5.pth
new file mode 100644
index 0000000000000000000000000000000000000000..bf100a21f1d144b50bfa1f045e01b1845219beed
--- /dev/null
+++ b/checkpoint-800/rng_state_5.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ad691a03017a27194bdeaa22e4e8e565f25687f69ed6893da91d506a635a8004
+size 15728
diff --git a/checkpoint-800/rng_state_6.pth b/checkpoint-800/rng_state_6.pth
new file mode 100644
index 0000000000000000000000000000000000000000..5b88707b21d808a5697219b54da12402ab4080dd
--- /dev/null
+++ b/checkpoint-800/rng_state_6.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6f23e0de8a38857769ea5447ac0f8a1aa0b5652b2e1cfc185338d02ac45f7d99
+size 15728
diff --git a/checkpoint-800/scheduler.pt b/checkpoint-800/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..6abdf7fb839c9087d8bae3eb5a62b54d3843f2f2
--- /dev/null
+++ b/checkpoint-800/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e00d6619acd478188c962eaba50828f3299d65c5cfd2ee4d1f69924755771732
+size 1064
diff --git a/checkpoint-800/special_tokens_map.json b/checkpoint-800/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..72ecfeeb7e14d244c936169d2ed139eeae235ef1
--- /dev/null
+++ b/checkpoint-800/special_tokens_map.json
@@ -0,0 +1,24 @@
+{
+  "bos_token": {
+    "content": "<s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "</s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "</s>",
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}
diff --git a/checkpoint-800/tokenizer.json b/checkpoint-800/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..b35b1c0f0bd5007afc7ad5976eeaf44988bd2067
--- /dev/null
+++ b/checkpoint-800/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9a1c103d223ee5dc2dde8307635f7d12581b40855198a38efcfeb6db4e08da69
+size 17078445
diff --git a/checkpoint-800/tokenizer_config.json b/checkpoint-800/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..550f3f7fc9144301cd0f17eeb0ad8746a5fa5354
--- /dev/null
+++ b/checkpoint-800/tokenizer_config.json
@@ -0,0 +1,8020 @@
+{
+  "add_bos_token": true,
+  "add_eos_token": false,
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "[INST]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "4": {
+      "content": "[/INST]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "5": {
+      "content": "[AVAILABLE_TOOLS]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "6": {
+      "content": "[/AVAILABLE_TOOLS]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "7": {
+      "content": "[TOOL_RESULTS]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "8": {
+      "content": "[/TOOL_RESULTS]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "9": {
+      "content": "[TOOL_CALLS]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "10": {
+      "content": "<pad>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "11": {
+      "content": "[PREFIX]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "12": {
+      "content": "[MIDDLE]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "13": {
+      "content": "[SUFFIX]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "14": {
+      "content": "<SPECIAL_14>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "15": {
+      "content": "<SPECIAL_15>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "16": {
+      "content": "<SPECIAL_16>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "17": {
+      "content": "<SPECIAL_17>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "18": {
+      "content": "<SPECIAL_18>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "19": {
+      "content": "<SPECIAL_19>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "20": {
+      "content": "<SPECIAL_20>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "21": {
+      "content": "<SPECIAL_21>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "22": {
+      "content": "<SPECIAL_22>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "23": {
+      "content": "<SPECIAL_23>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "24": {
+      "content": "<SPECIAL_24>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "25": {
+      "content": "<SPECIAL_25>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "26": {
+      "content": "<SPECIAL_26>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "27": {
+      "content": "<SPECIAL_27>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "28": {
+      "content": "<SPECIAL_28>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "29": {
+      "content": "<SPECIAL_29>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "30": {
+      "content": "<SPECIAL_30>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "31": {
+      "content": "<SPECIAL_31>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32": {
+      "content": "<SPECIAL_32>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "33": {
+      "content": "<SPECIAL_33>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "34": {
+      "content": "<SPECIAL_34>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "35": {
+      "content": "<SPECIAL_35>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "36": {
+      "content": "<SPECIAL_36>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "37": {
+      "content": "<SPECIAL_37>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "38": {
+      "content": "<SPECIAL_38>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "39": {
+      "content": "<SPECIAL_39>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "40": {
+      "content": "<SPECIAL_40>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "41": {
+      "content": "<SPECIAL_41>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "42": {
+      "content": "<SPECIAL_42>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "43": {
+      "content": "<SPECIAL_43>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "44": {
+      "content": "<SPECIAL_44>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "45": {
+      "content": "<SPECIAL_45>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "46": {
+      "content": "<SPECIAL_46>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "47": {
+      "content": "<SPECIAL_47>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "48": {
+      "content": "<SPECIAL_48>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "49": {
+      "content": "<SPECIAL_49>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "50": {
+      "content": "<SPECIAL_50>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "51": {
+      "content": "<SPECIAL_51>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "52": {
+      "content": "<SPECIAL_52>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "53": {
+      "content": "<SPECIAL_53>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "54": {
+      "content": "<SPECIAL_54>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "55": {
+      "content": "<SPECIAL_55>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "56": {
+      "content": "<SPECIAL_56>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "57": {
+      "content": "<SPECIAL_57>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "58": {
+      "content": "<SPECIAL_58>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "59": {
+      "content": "<SPECIAL_59>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "60": {
+      "content": "<SPECIAL_60>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "61": {
+      "content": "<SPECIAL_61>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "62": {
+      "content": "<SPECIAL_62>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "63": {
+      "content": "<SPECIAL_63>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "64": {
+      "content": "<SPECIAL_64>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "65": {
+      "content": "<SPECIAL_65>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "66": {
+      "content": "<SPECIAL_66>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "67": {
+      "content": "<SPECIAL_67>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "68": {
+      "content": "<SPECIAL_68>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "69": {
+      "content": "<SPECIAL_69>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "70": {
+      "content": "<SPECIAL_70>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "71": {
+      "content": "<SPECIAL_71>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "72": {
+      "content": "<SPECIAL_72>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "73": {
+      "content": "<SPECIAL_73>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "74": {
+      "content": "<SPECIAL_74>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "75": {
+      "content": "<SPECIAL_75>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "76": {
+      "content": "<SPECIAL_76>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "77": {
+      "content": "<SPECIAL_77>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "78": {
+      "content": "<SPECIAL_78>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "79": {
+      "content": "<SPECIAL_79>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "80": {
+      "content": "<SPECIAL_80>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "81": {
+      "content": "<SPECIAL_81>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "82": {
+      "content": "<SPECIAL_82>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "83": {
+      "content": "<SPECIAL_83>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "84": {
+      "content": "<SPECIAL_84>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "85": {
+      "content": "<SPECIAL_85>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "86": {
+      "content": "<SPECIAL_86>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "87": {
+      "content": "<SPECIAL_87>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "88": {
+      "content": "<SPECIAL_88>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "89": {
+      "content": "<SPECIAL_89>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "90": {
+      "content": "<SPECIAL_90>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "91": {
+      "content": "<SPECIAL_91>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "92": {
+      "content": "<SPECIAL_92>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "93": {
+      "content": "<SPECIAL_93>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "94": {
+      "content": "<SPECIAL_94>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "95": {
+      "content": "<SPECIAL_95>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "96": {
+      "content": "<SPECIAL_96>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "97": {
+      "content": "<SPECIAL_97>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "98": {
+      "content": "<SPECIAL_98>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "99": {
+      "content": "<SPECIAL_99>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "100": {
+      "content": "<SPECIAL_100>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "101": {
+      "content": "<SPECIAL_101>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "102": {
+      "content": "<SPECIAL_102>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "103": {
+      "content": "<SPECIAL_103>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "104": {
+      "content": "<SPECIAL_104>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "105": {
+      "content": "<SPECIAL_105>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "106": {
+      "content": "<SPECIAL_106>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "107": {
+      "content": "<SPECIAL_107>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "108": {
+      "content": "<SPECIAL_108>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "109": {
+      "content": "<SPECIAL_109>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "110": {
+      "content": "<SPECIAL_110>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "111": {
+      "content": "<SPECIAL_111>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "112": {
+      "content": "<SPECIAL_112>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "113": {
+      "content": "<SPECIAL_113>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "114": {
+      "content": "<SPECIAL_114>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "115": {
+      "content": "<SPECIAL_115>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "116": {
+      "content": "<SPECIAL_116>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "117": {
+      "content": "<SPECIAL_117>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "118": {
+      "content": "<SPECIAL_118>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "119": {
+      "content": "<SPECIAL_119>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "120": {
+      "content": "<SPECIAL_120>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "121": {
+      "content": "<SPECIAL_121>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "122": {
+      "content": "<SPECIAL_122>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "123": {
+      "content": "<SPECIAL_123>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "124": {
+      "content": "<SPECIAL_124>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "125": {
+      "content": "<SPECIAL_125>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "126": {
+      "content": "<SPECIAL_126>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "127": {
+      "content": "<SPECIAL_127>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128": {
+      "content": "<SPECIAL_128>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "129": {
+      "content": "<SPECIAL_129>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "130": {
+      "content": "<SPECIAL_130>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "131": {
+      "content": "<SPECIAL_131>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "132": {
+      "content": "<SPECIAL_132>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "133": {
+      "content": "<SPECIAL_133>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "134": {
+      "content": "<SPECIAL_134>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "135": {
+      "content": "<SPECIAL_135>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "136": {
+      "content": "<SPECIAL_136>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "137": {
+      "content": "<SPECIAL_137>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "138": {
+      "content": "<SPECIAL_138>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "139": {
+      "content": "<SPECIAL_139>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "140": {
+      "content": "<SPECIAL_140>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "141": {
+      "content": "<SPECIAL_141>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "142": {
+      "content": "<SPECIAL_142>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "143": {
+      "content": "<SPECIAL_143>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "144": {
+      "content": "<SPECIAL_144>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "145": {
+      "content": "<SPECIAL_145>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "146": {
+      "content": "<SPECIAL_146>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "147": {
+      "content": "<SPECIAL_147>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "148": {
+      "content": "<SPECIAL_148>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "149": {
+      "content": "<SPECIAL_149>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "150": {
+      "content": "<SPECIAL_150>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151": {
+      "content": "<SPECIAL_151>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "152": {
+      "content": "<SPECIAL_152>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "153": {
+      "content": "<SPECIAL_153>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "154": {
+      "content": "<SPECIAL_154>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "155": {
+      "content": "<SPECIAL_155>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "156": {
+      "content": "<SPECIAL_156>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "157": {
+      "content": "<SPECIAL_157>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "158": {
+      "content": "<SPECIAL_158>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "159": {
+      "content": "<SPECIAL_159>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "160": {
+      "content": "<SPECIAL_160>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "161": {
+      "content": "<SPECIAL_161>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "162": {
+      "content": "<SPECIAL_162>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "163": {
+      "content": "<SPECIAL_163>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "164": {
+      "content": "<SPECIAL_164>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "165": {
+      "content": "<SPECIAL_165>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "166": {
+      "content": "<SPECIAL_166>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "167": {
+      "content": "<SPECIAL_167>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "168": {
+      "content": "<SPECIAL_168>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "169": {
+      "content": "<SPECIAL_169>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170": {
+      "content": "<SPECIAL_170>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "171": {
+      "content": "<SPECIAL_171>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "172": {
+      "content": "<SPECIAL_172>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "173": {
+      "content": "<SPECIAL_173>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "174": {
+      "content": "<SPECIAL_174>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "175": {
+      "content": "<SPECIAL_175>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "176": {
+      "content": "<SPECIAL_176>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "177": {
+      "content": "<SPECIAL_177>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "178": {
+      "content": "<SPECIAL_178>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "179": {
+      "content": "<SPECIAL_179>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "180": {
+      "content": "<SPECIAL_180>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "181": {
+      "content": "<SPECIAL_181>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "182": {
+      "content": "<SPECIAL_182>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "183": {
+      "content": "<SPECIAL_183>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "184": {
+      "content": "<SPECIAL_184>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "185": {
+      "content": "<SPECIAL_185>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "186": {
+      "content": "<SPECIAL_186>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "187": {
+      "content": "<SPECIAL_187>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "188": {
+      "content": "<SPECIAL_188>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "189": {
+      "content": "<SPECIAL_189>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "190": {
+      "content": "<SPECIAL_190>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "191": {
+      "content": "<SPECIAL_191>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "192": {
+      "content": "<SPECIAL_192>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "193": {
+      "content": "<SPECIAL_193>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "194": {
+      "content": "<SPECIAL_194>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "195": {
+      "content": "<SPECIAL_195>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "196": {
+      "content": "<SPECIAL_196>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "197": {
+      "content": "<SPECIAL_197>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "198": {
+      "content": "<SPECIAL_198>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "199": {
+      "content": "<SPECIAL_199>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "200": {
+      "content": "<SPECIAL_200>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "201": {
+      "content": "<SPECIAL_201>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "202": {
+      "content": "<SPECIAL_202>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "203": {
+      "content": "<SPECIAL_203>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "204": {
+      "content": "<SPECIAL_204>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "205": {
+      "content": "<SPECIAL_205>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "206": {
+      "content": "<SPECIAL_206>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "207": {
+      "content": "<SPECIAL_207>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "208": {
+      "content": "<SPECIAL_208>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "209": {
+      "content": "<SPECIAL_209>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "210": {
+      "content": "<SPECIAL_210>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "211": {
+      "content": "<SPECIAL_211>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "212": {
+      "content": "<SPECIAL_212>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "213": {
+      "content": "<SPECIAL_213>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "214": {
+      "content": "<SPECIAL_214>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "215": {
+      "content": "<SPECIAL_215>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "216": {
+      "content": "<SPECIAL_216>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "217": {
+      "content": "<SPECIAL_217>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "218": {
+      "content": "<SPECIAL_218>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "219": {
+      "content": "<SPECIAL_219>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "220": {
+      "content": "<SPECIAL_220>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "221": {
+      "content": "<SPECIAL_221>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "222": {
+      "content": "<SPECIAL_222>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "223": {
+      "content": "<SPECIAL_223>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "224": {
+      "content": "<SPECIAL_224>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "225": {
+      "content": "<SPECIAL_225>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "226": {
+      "content": "<SPECIAL_226>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "227": {
+      "content": "<SPECIAL_227>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "228": {
+      "content": "<SPECIAL_228>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "229": {
+      "content": "<SPECIAL_229>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "230": {
+      "content": "<SPECIAL_230>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "231": {
+      "content": "<SPECIAL_231>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "232": {
+      "content": "<SPECIAL_232>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "233": {
+      "content": "<SPECIAL_233>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "234": {
+      "content": "<SPECIAL_234>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "235": {
+      "content": "<SPECIAL_235>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "236": {
+      "content": "<SPECIAL_236>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "237": {
+      "content": "<SPECIAL_237>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "238": {
+      "content": "<SPECIAL_238>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "239": {
+      "content": "<SPECIAL_239>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "240": {
+      "content": "<SPECIAL_240>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "241": {
+      "content": "<SPECIAL_241>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "242": {
+      "content": "<SPECIAL_242>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "243": {
+      "content": "<SPECIAL_243>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "244": {
+      "content": "<SPECIAL_244>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "245": {
+      "content": "<SPECIAL_245>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "246": {
+      "content": "<SPECIAL_246>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "247": {
+      "content": "<SPECIAL_247>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "248": {
+      "content": "<SPECIAL_248>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "249": {
+      "content": "<SPECIAL_249>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250": {
+      "content": "<SPECIAL_250>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "251": {
+      "content": "<SPECIAL_251>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "252": {
+      "content": "<SPECIAL_252>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "253": {
+      "content": "<SPECIAL_253>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "254": {
+      "content": "<SPECIAL_254>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "255": {
+      "content": "<SPECIAL_255>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "256": {
+      "content": "<SPECIAL_256>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "257": {
+      "content": "<SPECIAL_257>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "258": {
+      "content": "<SPECIAL_258>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "259": {
+      "content": "<SPECIAL_259>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "260": {
+      "content": "<SPECIAL_260>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "261": {
+      "content": "<SPECIAL_261>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "262": {
+      "content": "<SPECIAL_262>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "263": {
+      "content": "<SPECIAL_263>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "264": {
+      "content": "<SPECIAL_264>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "265": {
+      "content": "<SPECIAL_265>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "266": {
+      "content": "<SPECIAL_266>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "267": {
+      "content": "<SPECIAL_267>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "268": {
+      "content": "<SPECIAL_268>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "269": {
+      "content": "<SPECIAL_269>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "270": {
+      "content": "<SPECIAL_270>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "271": {
+      "content": "<SPECIAL_271>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "272": {
+      "content": "<SPECIAL_272>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "273": {
+      "content": "<SPECIAL_273>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "274": {
+      "content": "<SPECIAL_274>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "275": {
+      "content": "<SPECIAL_275>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "276": {
+      "content": "<SPECIAL_276>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "277": {
+      "content": "<SPECIAL_277>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "278": {
+      "content": "<SPECIAL_278>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "279": {
+      "content": "<SPECIAL_279>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "280": {
+      "content": "<SPECIAL_280>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "281": {
+      "content": "<SPECIAL_281>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "282": {
+      "content": "<SPECIAL_282>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "283": {
+      "content": "<SPECIAL_283>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "284": {
+      "content": "<SPECIAL_284>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "285": {
+      "content": "<SPECIAL_285>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "286": {
+      "content": "<SPECIAL_286>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "287": {
+      "content": "<SPECIAL_287>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "288": {
+      "content": "<SPECIAL_288>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "289": {
+      "content": "<SPECIAL_289>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "290": {
+      "content": "<SPECIAL_290>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "291": {
+      "content": "<SPECIAL_291>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "292": {
+      "content": "<SPECIAL_292>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "293": {
+      "content": "<SPECIAL_293>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "294": {
+      "content": "<SPECIAL_294>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "295": {
+      "content": "<SPECIAL_295>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "296": {
+      "content": "<SPECIAL_296>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "297": {
+      "content": "<SPECIAL_297>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "298": {
+      "content": "<SPECIAL_298>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "299": {
+      "content": "<SPECIAL_299>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "300": {
+      "content": "<SPECIAL_300>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "301": {
+      "content": "<SPECIAL_301>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "302": {
+      "content": "<SPECIAL_302>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "303": {
+      "content": "<SPECIAL_303>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "304": {
+      "content": "<SPECIAL_304>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "305": {
+      "content": "<SPECIAL_305>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "306": {
+      "content": "<SPECIAL_306>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "307": {
+      "content": "<SPECIAL_307>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "308": {
+      "content": "<SPECIAL_308>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "309": {
+      "content": "<SPECIAL_309>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "310": {
+      "content": "<SPECIAL_310>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "311": {
+      "content": "<SPECIAL_311>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "312": {
+      "content": "<SPECIAL_312>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "313": {
+      "content": "<SPECIAL_313>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "314": {
+      "content": "<SPECIAL_314>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "315": {
+      "content": "<SPECIAL_315>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "316": {
+      "content": "<SPECIAL_316>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "317": {
+      "content": "<SPECIAL_317>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "318": {
+      "content": "<SPECIAL_318>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "319": {
+      "content": "<SPECIAL_319>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "320": {
+      "content": "<SPECIAL_320>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "321": {
+      "content": "<SPECIAL_321>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "322": {
+      "content": "<SPECIAL_322>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "323": {
+      "content": "<SPECIAL_323>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "324": {
+      "content": "<SPECIAL_324>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "325": {
+      "content": "<SPECIAL_325>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "326": {
+      "content": "<SPECIAL_326>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "327": {
+      "content": "<SPECIAL_327>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "328": {
+      "content": "<SPECIAL_328>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "329": {
+      "content": "<SPECIAL_329>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "330": {
+      "content": "<SPECIAL_330>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "331": {
+      "content": "<SPECIAL_331>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "332": {
+      "content": "<SPECIAL_332>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "333": {
+      "content": "<SPECIAL_333>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "334": {
+      "content": "<SPECIAL_334>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "335": {
+      "content": "<SPECIAL_335>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "336": {
+      "content": "<SPECIAL_336>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "337": {
+      "content": "<SPECIAL_337>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "338": {
+      "content": "<SPECIAL_338>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "339": {
+      "content": "<SPECIAL_339>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "340": {
+      "content": "<SPECIAL_340>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "341": {
+      "content": "<SPECIAL_341>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "342": {
+      "content": "<SPECIAL_342>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "343": {
+      "content": "<SPECIAL_343>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "344": {
+      "content": "<SPECIAL_344>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "345": {
+      "content": "<SPECIAL_345>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "346": {
+      "content": "<SPECIAL_346>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "347": {
+      "content": "<SPECIAL_347>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "348": {
+      "content": "<SPECIAL_348>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "349": {
+      "content": "<SPECIAL_349>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "350": {
+      "content": "<SPECIAL_350>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "351": {
+      "content": "<SPECIAL_351>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "352": {
+      "content": "<SPECIAL_352>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "353": {
+      "content": "<SPECIAL_353>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "354": {
+      "content": "<SPECIAL_354>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "355": {
+      "content": "<SPECIAL_355>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "356": {
+      "content": "<SPECIAL_356>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "357": {
+      "content": "<SPECIAL_357>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "358": {
+      "content": "<SPECIAL_358>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "359": {
+      "content": "<SPECIAL_359>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "360": {
+      "content": "<SPECIAL_360>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "361": {
+      "content": "<SPECIAL_361>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "362": {
+      "content": "<SPECIAL_362>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "363": {
+      "content": "<SPECIAL_363>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "364": {
+      "content": "<SPECIAL_364>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "365": {
+      "content": "<SPECIAL_365>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "366": {
+      "content": "<SPECIAL_366>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "367": {
+      "content": "<SPECIAL_367>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "368": {
+      "content": "<SPECIAL_368>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "369": {
+      "content": "<SPECIAL_369>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "370": {
+      "content": "<SPECIAL_370>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "371": {
+      "content": "<SPECIAL_371>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "372": {
+      "content": "<SPECIAL_372>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "373": {
+      "content": "<SPECIAL_373>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "374": {
+      "content": "<SPECIAL_374>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "375": {
+      "content": "<SPECIAL_375>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "376": {
+      "content": "<SPECIAL_376>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "377": {
+      "content": "<SPECIAL_377>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "378": {
+      "content": "<SPECIAL_378>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "379": {
+      "content": "<SPECIAL_379>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "380": {
+      "content": "<SPECIAL_380>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "381": {
+      "content": "<SPECIAL_381>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "382": {
+      "content": "<SPECIAL_382>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "383": {
+      "content": "<SPECIAL_383>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "384": {
+      "content": "<SPECIAL_384>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "385": {
+      "content": "<SPECIAL_385>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "386": {
+      "content": "<SPECIAL_386>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "387": {
+      "content": "<SPECIAL_387>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "388": {
+      "content": "<SPECIAL_388>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "389": {
+      "content": "<SPECIAL_389>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "390": {
+      "content": "<SPECIAL_390>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "391": {
+      "content": "<SPECIAL_391>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "392": {
+      "content": "<SPECIAL_392>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "393": {
+      "content": "<SPECIAL_393>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "394": {
+      "content": "<SPECIAL_394>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "395": {
+      "content": "<SPECIAL_395>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "396": {
+      "content": "<SPECIAL_396>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "397": {
+      "content": "<SPECIAL_397>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "398": {
+      "content": "<SPECIAL_398>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "399": {
+      "content": "<SPECIAL_399>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "400": {
+      "content": "<SPECIAL_400>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "401": {
+      "content": "<SPECIAL_401>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "402": {
+      "content": "<SPECIAL_402>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "403": {
+      "content": "<SPECIAL_403>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "404": {
+      "content": "<SPECIAL_404>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "405": {
+      "content": "<SPECIAL_405>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "406": {
+      "content": "<SPECIAL_406>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "407": {
+      "content": "<SPECIAL_407>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "408": {
+      "content": "<SPECIAL_408>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "409": {
+      "content": "<SPECIAL_409>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "410": {
+      "content": "<SPECIAL_410>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "411": {
+      "content": "<SPECIAL_411>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "412": {
+      "content": "<SPECIAL_412>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "413": {
+      "content": "<SPECIAL_413>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "414": {
+      "content": "<SPECIAL_414>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "415": {
+      "content": "<SPECIAL_415>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "416": {
+      "content": "<SPECIAL_416>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "417": {
+      "content": "<SPECIAL_417>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "418": {
+      "content": "<SPECIAL_418>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "419": {
+      "content": "<SPECIAL_419>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "420": {
+      "content": "<SPECIAL_420>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "421": {
+      "content": "<SPECIAL_421>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "422": {
+      "content": "<SPECIAL_422>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "423": {
+      "content": "<SPECIAL_423>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "424": {
+      "content": "<SPECIAL_424>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "425": {
+      "content": "<SPECIAL_425>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "426": {
+      "content": "<SPECIAL_426>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "427": {
+      "content": "<SPECIAL_427>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "428": {
+      "content": "<SPECIAL_428>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "429": {
+      "content": "<SPECIAL_429>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "430": {
+      "content": "<SPECIAL_430>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "431": {
+      "content": "<SPECIAL_431>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "432": {
+      "content": "<SPECIAL_432>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "433": {
+      "content": "<SPECIAL_433>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "434": {
+      "content": "<SPECIAL_434>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "435": {
+      "content": "<SPECIAL_435>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "436": {
+      "content": "<SPECIAL_436>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "437": {
+      "content": "<SPECIAL_437>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "438": {
+      "content": "<SPECIAL_438>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "439": {
+      "content": "<SPECIAL_439>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "440": {
+      "content": "<SPECIAL_440>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "441": {
+      "content": "<SPECIAL_441>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "442": {
+      "content": "<SPECIAL_442>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "443": {
+      "content": "<SPECIAL_443>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "444": {
+      "content": "<SPECIAL_444>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "445": {
+      "content": "<SPECIAL_445>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "446": {
+      "content": "<SPECIAL_446>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "447": {
+      "content": "<SPECIAL_447>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "448": {
+      "content": "<SPECIAL_448>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "449": {
+      "content": "<SPECIAL_449>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "450": {
+      "content": "<SPECIAL_450>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "451": {
+      "content": "<SPECIAL_451>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "452": {
+      "content": "<SPECIAL_452>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "453": {
+      "content": "<SPECIAL_453>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "454": {
+      "content": "<SPECIAL_454>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "455": {
+      "content": "<SPECIAL_455>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "456": {
+      "content": "<SPECIAL_456>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "457": {
+      "content": "<SPECIAL_457>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "458": {
+      "content": "<SPECIAL_458>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "459": {
+      "content": "<SPECIAL_459>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "460": {
+      "content": "<SPECIAL_460>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "461": {
+      "content": "<SPECIAL_461>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "462": {
+      "content": "<SPECIAL_462>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "463": {
+      "content": "<SPECIAL_463>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "464": {
+      "content": "<SPECIAL_464>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "465": {
+      "content": "<SPECIAL_465>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "466": {
+      "content": "<SPECIAL_466>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "467": {
+      "content": "<SPECIAL_467>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "468": {
+      "content": "<SPECIAL_468>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "469": {
+      "content": "<SPECIAL_469>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "470": {
+      "content": "<SPECIAL_470>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "471": {
+      "content": "<SPECIAL_471>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "472": {
+      "content": "<SPECIAL_472>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "473": {
+      "content": "<SPECIAL_473>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "474": {
+      "content": "<SPECIAL_474>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "475": {
+      "content": "<SPECIAL_475>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "476": {
+      "content": "<SPECIAL_476>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "477": {
+      "content": "<SPECIAL_477>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "478": {
+      "content": "<SPECIAL_478>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "479": {
+      "content": "<SPECIAL_479>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "480": {
+      "content": "<SPECIAL_480>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "481": {
+      "content": "<SPECIAL_481>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "482": {
+      "content": "<SPECIAL_482>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "483": {
+      "content": "<SPECIAL_483>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "484": {
+      "content": "<SPECIAL_484>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "485": {
+      "content": "<SPECIAL_485>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "486": {
+      "content": "<SPECIAL_486>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "487": {
+      "content": "<SPECIAL_487>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "488": {
+      "content": "<SPECIAL_488>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "489": {
+      "content": "<SPECIAL_489>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "490": {
+      "content": "<SPECIAL_490>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "491": {
+      "content": "<SPECIAL_491>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "492": {
+      "content": "<SPECIAL_492>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "493": {
+      "content": "<SPECIAL_493>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "494": {
+      "content": "<SPECIAL_494>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "495": {
+      "content": "<SPECIAL_495>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "496": {
+      "content": "<SPECIAL_496>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "497": {
+      "content": "<SPECIAL_497>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "498": {
+      "content": "<SPECIAL_498>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "499": {
+      "content": "<SPECIAL_499>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "500": {
+      "content": "<SPECIAL_500>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "501": {
+      "content": "<SPECIAL_501>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "502": {
+      "content": "<SPECIAL_502>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "503": {
+      "content": "<SPECIAL_503>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "504": {
+      "content": "<SPECIAL_504>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "505": {
+      "content": "<SPECIAL_505>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "506": {
+      "content": "<SPECIAL_506>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "507": {
+      "content": "<SPECIAL_507>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "508": {
+      "content": "<SPECIAL_508>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "509": {
+      "content": "<SPECIAL_509>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "510": {
+      "content": "<SPECIAL_510>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "511": {
+      "content": "<SPECIAL_511>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "512": {
+      "content": "<SPECIAL_512>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "513": {
+      "content": "<SPECIAL_513>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "514": {
+      "content": "<SPECIAL_514>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "515": {
+      "content": "<SPECIAL_515>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "516": {
+      "content": "<SPECIAL_516>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "517": {
+      "content": "<SPECIAL_517>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "518": {
+      "content": "<SPECIAL_518>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "519": {
+      "content": "<SPECIAL_519>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "520": {
+      "content": "<SPECIAL_520>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "521": {
+      "content": "<SPECIAL_521>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "522": {
+      "content": "<SPECIAL_522>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "523": {
+      "content": "<SPECIAL_523>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "524": {
+      "content": "<SPECIAL_524>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "525": {
+      "content": "<SPECIAL_525>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "526": {
+      "content": "<SPECIAL_526>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "527": {
+      "content": "<SPECIAL_527>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "528": {
+      "content": "<SPECIAL_528>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "529": {
+      "content": "<SPECIAL_529>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "530": {
+      "content": "<SPECIAL_530>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "531": {
+      "content": "<SPECIAL_531>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "532": {
+      "content": "<SPECIAL_532>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "533": {
+      "content": "<SPECIAL_533>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "534": {
+      "content": "<SPECIAL_534>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "535": {
+      "content": "<SPECIAL_535>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "536": {
+      "content": "<SPECIAL_536>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "537": {
+      "content": "<SPECIAL_537>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "538": {
+      "content": "<SPECIAL_538>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "539": {
+      "content": "<SPECIAL_539>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "540": {
+      "content": "<SPECIAL_540>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "541": {
+      "content": "<SPECIAL_541>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "542": {
+      "content": "<SPECIAL_542>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "543": {
+      "content": "<SPECIAL_543>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "544": {
+      "content": "<SPECIAL_544>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "545": {
+      "content": "<SPECIAL_545>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "546": {
+      "content": "<SPECIAL_546>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "547": {
+      "content": "<SPECIAL_547>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "548": {
+      "content": "<SPECIAL_548>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "549": {
+      "content": "<SPECIAL_549>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "550": {
+      "content": "<SPECIAL_550>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "551": {
+      "content": "<SPECIAL_551>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "552": {
+      "content": "<SPECIAL_552>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "553": {
+      "content": "<SPECIAL_553>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "554": {
+      "content": "<SPECIAL_554>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "555": {
+      "content": "<SPECIAL_555>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "556": {
+      "content": "<SPECIAL_556>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "557": {
+      "content": "<SPECIAL_557>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "558": {
+      "content": "<SPECIAL_558>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "559": {
+      "content": "<SPECIAL_559>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "560": {
+      "content": "<SPECIAL_560>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "561": {
+      "content": "<SPECIAL_561>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "562": {
+      "content": "<SPECIAL_562>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "563": {
+      "content": "<SPECIAL_563>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "564": {
+      "content": "<SPECIAL_564>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "565": {
+      "content": "<SPECIAL_565>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "566": {
+      "content": "<SPECIAL_566>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "567": {
+      "content": "<SPECIAL_567>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "568": {
+      "content": "<SPECIAL_568>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "569": {
+      "content": "<SPECIAL_569>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "570": {
+      "content": "<SPECIAL_570>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "571": {
+      "content": "<SPECIAL_571>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "572": {
+      "content": "<SPECIAL_572>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "573": {
+      "content": "<SPECIAL_573>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "574": {
+      "content": "<SPECIAL_574>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "575": {
+      "content": "<SPECIAL_575>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "576": {
+      "content": "<SPECIAL_576>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "577": {
+      "content": "<SPECIAL_577>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "578": {
+      "content": "<SPECIAL_578>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "579": {
+      "content": "<SPECIAL_579>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "580": {
+      "content": "<SPECIAL_580>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "581": {
+      "content": "<SPECIAL_581>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "582": {
+      "content": "<SPECIAL_582>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "583": {
+      "content": "<SPECIAL_583>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "584": {
+      "content": "<SPECIAL_584>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "585": {
+      "content": "<SPECIAL_585>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "586": {
+      "content": "<SPECIAL_586>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "587": {
+      "content": "<SPECIAL_587>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "588": {
+      "content": "<SPECIAL_588>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "589": {
+      "content": "<SPECIAL_589>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "590": {
+      "content": "<SPECIAL_590>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "591": {
+      "content": "<SPECIAL_591>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "592": {
+      "content": "<SPECIAL_592>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "593": {
+      "content": "<SPECIAL_593>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "594": {
+      "content": "<SPECIAL_594>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "595": {
+      "content": "<SPECIAL_595>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "596": {
+      "content": "<SPECIAL_596>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "597": {
+      "content": "<SPECIAL_597>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "598": {
+      "content": "<SPECIAL_598>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "599": {
+      "content": "<SPECIAL_599>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "600": {
+      "content": "<SPECIAL_600>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "601": {
+      "content": "<SPECIAL_601>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "602": {
+      "content": "<SPECIAL_602>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "603": {
+      "content": "<SPECIAL_603>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "604": {
+      "content": "<SPECIAL_604>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "605": {
+      "content": "<SPECIAL_605>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "606": {
+      "content": "<SPECIAL_606>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "607": {
+      "content": "<SPECIAL_607>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "608": {
+      "content": "<SPECIAL_608>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "609": {
+      "content": "<SPECIAL_609>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "610": {
+      "content": "<SPECIAL_610>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "611": {
+      "content": "<SPECIAL_611>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "612": {
+      "content": "<SPECIAL_612>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "613": {
+      "content": "<SPECIAL_613>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "614": {
+      "content": "<SPECIAL_614>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "615": {
+      "content": "<SPECIAL_615>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "616": {
+      "content": "<SPECIAL_616>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "617": {
+      "content": "<SPECIAL_617>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "618": {
+      "content": "<SPECIAL_618>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "619": {
+      "content": "<SPECIAL_619>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "620": {
+      "content": "<SPECIAL_620>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "621": {
+      "content": "<SPECIAL_621>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "622": {
+      "content": "<SPECIAL_622>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "623": {
+      "content": "<SPECIAL_623>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "624": {
+      "content": "<SPECIAL_624>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "625": {
+      "content": "<SPECIAL_625>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "626": {
+      "content": "<SPECIAL_626>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "627": {
+      "content": "<SPECIAL_627>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "628": {
+      "content": "<SPECIAL_628>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "629": {
+      "content": "<SPECIAL_629>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "630": {
+      "content": "<SPECIAL_630>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "631": {
+      "content": "<SPECIAL_631>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "632": {
+      "content": "<SPECIAL_632>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "633": {
+      "content": "<SPECIAL_633>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "634": {
+      "content": "<SPECIAL_634>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "635": {
+      "content": "<SPECIAL_635>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "636": {
+      "content": "<SPECIAL_636>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "637": {
+      "content": "<SPECIAL_637>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "638": {
+      "content": "<SPECIAL_638>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "639": {
+      "content": "<SPECIAL_639>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "640": {
+      "content": "<SPECIAL_640>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "641": {
+      "content": "<SPECIAL_641>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "642": {
+      "content": "<SPECIAL_642>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "643": {
+      "content": "<SPECIAL_643>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "644": {
+      "content": "<SPECIAL_644>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "645": {
+      "content": "<SPECIAL_645>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "646": {
+      "content": "<SPECIAL_646>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "647": {
+      "content": "<SPECIAL_647>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "648": {
+      "content": "<SPECIAL_648>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "649": {
+      "content": "<SPECIAL_649>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "650": {
+      "content": "<SPECIAL_650>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "651": {
+      "content": "<SPECIAL_651>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "652": {
+      "content": "<SPECIAL_652>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "653": {
+      "content": "<SPECIAL_653>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "654": {
+      "content": "<SPECIAL_654>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "655": {
+      "content": "<SPECIAL_655>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "656": {
+      "content": "<SPECIAL_656>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "657": {
+      "content": "<SPECIAL_657>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "658": {
+      "content": "<SPECIAL_658>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "659": {
+      "content": "<SPECIAL_659>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "660": {
+      "content": "<SPECIAL_660>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "661": {
+      "content": "<SPECIAL_661>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "662": {
+      "content": "<SPECIAL_662>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "663": {
+      "content": "<SPECIAL_663>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "664": {
+      "content": "<SPECIAL_664>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "665": {
+      "content": "<SPECIAL_665>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "666": {
+      "content": "<SPECIAL_666>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "667": {
+      "content": "<SPECIAL_667>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "668": {
+      "content": "<SPECIAL_668>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "669": {
+      "content": "<SPECIAL_669>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "670": {
+      "content": "<SPECIAL_670>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "671": {
+      "content": "<SPECIAL_671>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "672": {
+      "content": "<SPECIAL_672>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "673": {
+      "content": "<SPECIAL_673>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "674": {
+      "content": "<SPECIAL_674>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "675": {
+      "content": "<SPECIAL_675>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "676": {
+      "content": "<SPECIAL_676>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "677": {
+      "content": "<SPECIAL_677>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "678": {
+      "content": "<SPECIAL_678>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "679": {
+      "content": "<SPECIAL_679>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "680": {
+      "content": "<SPECIAL_680>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "681": {
+      "content": "<SPECIAL_681>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "682": {
+      "content": "<SPECIAL_682>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "683": {
+      "content": "<SPECIAL_683>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "684": {
+      "content": "<SPECIAL_684>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "685": {
+      "content": "<SPECIAL_685>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "686": {
+      "content": "<SPECIAL_686>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "687": {
+      "content": "<SPECIAL_687>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "688": {
+      "content": "<SPECIAL_688>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "689": {
+      "content": "<SPECIAL_689>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "690": {
+      "content": "<SPECIAL_690>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "691": {
+      "content": "<SPECIAL_691>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "692": {
+      "content": "<SPECIAL_692>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "693": {
+      "content": "<SPECIAL_693>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "694": {
+      "content": "<SPECIAL_694>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "695": {
+      "content": "<SPECIAL_695>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "696": {
+      "content": "<SPECIAL_696>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "697": {
+      "content": "<SPECIAL_697>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "698": {
+      "content": "<SPECIAL_698>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "699": {
+      "content": "<SPECIAL_699>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "700": {
+      "content": "<SPECIAL_700>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "701": {
+      "content": "<SPECIAL_701>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "702": {
+      "content": "<SPECIAL_702>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "703": {
+      "content": "<SPECIAL_703>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "704": {
+      "content": "<SPECIAL_704>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "705": {
+      "content": "<SPECIAL_705>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "706": {
+      "content": "<SPECIAL_706>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "707": {
+      "content": "<SPECIAL_707>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "708": {
+      "content": "<SPECIAL_708>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "709": {
+      "content": "<SPECIAL_709>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "710": {
+      "content": "<SPECIAL_710>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "711": {
+      "content": "<SPECIAL_711>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "712": {
+      "content": "<SPECIAL_712>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "713": {
+      "content": "<SPECIAL_713>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "714": {
+      "content": "<SPECIAL_714>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "715": {
+      "content": "<SPECIAL_715>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "716": {
+      "content": "<SPECIAL_716>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "717": {
+      "content": "<SPECIAL_717>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "718": {
+      "content": "<SPECIAL_718>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "719": {
+      "content": "<SPECIAL_719>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "720": {
+      "content": "<SPECIAL_720>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "721": {
+      "content": "<SPECIAL_721>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "722": {
+      "content": "<SPECIAL_722>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "723": {
+      "content": "<SPECIAL_723>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "724": {
+      "content": "<SPECIAL_724>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "725": {
+      "content": "<SPECIAL_725>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "726": {
+      "content": "<SPECIAL_726>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "727": {
+      "content": "<SPECIAL_727>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "728": {
+      "content": "<SPECIAL_728>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "729": {
+      "content": "<SPECIAL_729>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "730": {
+      "content": "<SPECIAL_730>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "731": {
+      "content": "<SPECIAL_731>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "732": {
+      "content": "<SPECIAL_732>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "733": {
+      "content": "<SPECIAL_733>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "734": {
+      "content": "<SPECIAL_734>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "735": {
+      "content": "<SPECIAL_735>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "736": {
+      "content": "<SPECIAL_736>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "737": {
+      "content": "<SPECIAL_737>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "738": {
+      "content": "<SPECIAL_738>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "739": {
+      "content": "<SPECIAL_739>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "740": {
+      "content": "<SPECIAL_740>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "741": {
+      "content": "<SPECIAL_741>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "742": {
+      "content": "<SPECIAL_742>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "743": {
+      "content": "<SPECIAL_743>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "744": {
+      "content": "<SPECIAL_744>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "745": {
+      "content": "<SPECIAL_745>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "746": {
+      "content": "<SPECIAL_746>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "747": {
+      "content": "<SPECIAL_747>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "748": {
+      "content": "<SPECIAL_748>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "749": {
+      "content": "<SPECIAL_749>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "750": {
+      "content": "<SPECIAL_750>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "751": {
+      "content": "<SPECIAL_751>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "752": {
+      "content": "<SPECIAL_752>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "753": {
+      "content": "<SPECIAL_753>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "754": {
+      "content": "<SPECIAL_754>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "755": {
+      "content": "<SPECIAL_755>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "756": {
+      "content": "<SPECIAL_756>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "757": {
+      "content": "<SPECIAL_757>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "758": {
+      "content": "<SPECIAL_758>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "759": {
+      "content": "<SPECIAL_759>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "760": {
+      "content": "<SPECIAL_760>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "761": {
+      "content": "<SPECIAL_761>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "762": {
+      "content": "<SPECIAL_762>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "763": {
+      "content": "<SPECIAL_763>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "764": {
+      "content": "<SPECIAL_764>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "765": {
+      "content": "<SPECIAL_765>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "766": {
+      "content": "<SPECIAL_766>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "767": {
+      "content": "<SPECIAL_767>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "768": {
+      "content": "<SPECIAL_768>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "769": {
+      "content": "<SPECIAL_769>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "770": {
+      "content": "<SPECIAL_770>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "771": {
+      "content": "<SPECIAL_771>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "772": {
+      "content": "<SPECIAL_772>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "773": {
+      "content": "<SPECIAL_773>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "774": {
+      "content": "<SPECIAL_774>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "775": {
+      "content": "<SPECIAL_775>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "776": {
+      "content": "<SPECIAL_776>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "777": {
+      "content": "<SPECIAL_777>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "778": {
+      "content": "<SPECIAL_778>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "779": {
+      "content": "<SPECIAL_779>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "780": {
+      "content": "<SPECIAL_780>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "781": {
+      "content": "<SPECIAL_781>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "782": {
+      "content": "<SPECIAL_782>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "783": {
+      "content": "<SPECIAL_783>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "784": {
+      "content": "<SPECIAL_784>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "785": {
+      "content": "<SPECIAL_785>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "786": {
+      "content": "<SPECIAL_786>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "787": {
+      "content": "<SPECIAL_787>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "788": {
+      "content": "<SPECIAL_788>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "789": {
+      "content": "<SPECIAL_789>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "790": {
+      "content": "<SPECIAL_790>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "791": {
+      "content": "<SPECIAL_791>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "792": {
+      "content": "<SPECIAL_792>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "793": {
+      "content": "<SPECIAL_793>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "794": {
+      "content": "<SPECIAL_794>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "795": {
+      "content": "<SPECIAL_795>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "796": {
+      "content": "<SPECIAL_796>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "797": {
+      "content": "<SPECIAL_797>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "798": {
+      "content": "<SPECIAL_798>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "799": {
+      "content": "<SPECIAL_799>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "800": {
+      "content": "<SPECIAL_800>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "801": {
+      "content": "<SPECIAL_801>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "802": {
+      "content": "<SPECIAL_802>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "803": {
+      "content": "<SPECIAL_803>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "804": {
+      "content": "<SPECIAL_804>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "805": {
+      "content": "<SPECIAL_805>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "806": {
+      "content": "<SPECIAL_806>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "807": {
+      "content": "<SPECIAL_807>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "808": {
+      "content": "<SPECIAL_808>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "809": {
+      "content": "<SPECIAL_809>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "810": {
+      "content": "<SPECIAL_810>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "811": {
+      "content": "<SPECIAL_811>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "812": {
+      "content": "<SPECIAL_812>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "813": {
+      "content": "<SPECIAL_813>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "814": {
+      "content": "<SPECIAL_814>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "815": {
+      "content": "<SPECIAL_815>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "816": {
+      "content": "<SPECIAL_816>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "817": {
+      "content": "<SPECIAL_817>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "818": {
+      "content": "<SPECIAL_818>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "819": {
+      "content": "<SPECIAL_819>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "820": {
+      "content": "<SPECIAL_820>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "821": {
+      "content": "<SPECIAL_821>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "822": {
+      "content": "<SPECIAL_822>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "823": {
+      "content": "<SPECIAL_823>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "824": {
+      "content": "<SPECIAL_824>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "825": {
+      "content": "<SPECIAL_825>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "826": {
+      "content": "<SPECIAL_826>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "827": {
+      "content": "<SPECIAL_827>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "828": {
+      "content": "<SPECIAL_828>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "829": {
+      "content": "<SPECIAL_829>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "830": {
+      "content": "<SPECIAL_830>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "831": {
+      "content": "<SPECIAL_831>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "832": {
+      "content": "<SPECIAL_832>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "833": {
+      "content": "<SPECIAL_833>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "834": {
+      "content": "<SPECIAL_834>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "835": {
+      "content": "<SPECIAL_835>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "836": {
+      "content": "<SPECIAL_836>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "837": {
+      "content": "<SPECIAL_837>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "838": {
+      "content": "<SPECIAL_838>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "839": {
+      "content": "<SPECIAL_839>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "840": {
+      "content": "<SPECIAL_840>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "841": {
+      "content": "<SPECIAL_841>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "842": {
+      "content": "<SPECIAL_842>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "843": {
+      "content": "<SPECIAL_843>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "844": {
+      "content": "<SPECIAL_844>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "845": {
+      "content": "<SPECIAL_845>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "846": {
+      "content": "<SPECIAL_846>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "847": {
+      "content": "<SPECIAL_847>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "848": {
+      "content": "<SPECIAL_848>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "849": {
+      "content": "<SPECIAL_849>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "850": {
+      "content": "<SPECIAL_850>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "851": {
+      "content": "<SPECIAL_851>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "852": {
+      "content": "<SPECIAL_852>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "853": {
+      "content": "<SPECIAL_853>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "854": {
+      "content": "<SPECIAL_854>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "855": {
+      "content": "<SPECIAL_855>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "856": {
+      "content": "<SPECIAL_856>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "857": {
+      "content": "<SPECIAL_857>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "858": {
+      "content": "<SPECIAL_858>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "859": {
+      "content": "<SPECIAL_859>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "860": {
+      "content": "<SPECIAL_860>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "861": {
+      "content": "<SPECIAL_861>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "862": {
+      "content": "<SPECIAL_862>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "863": {
+      "content": "<SPECIAL_863>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "864": {
+      "content": "<SPECIAL_864>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "865": {
+      "content": "<SPECIAL_865>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "866": {
+      "content": "<SPECIAL_866>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "867": {
+      "content": "<SPECIAL_867>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "868": {
+      "content": "<SPECIAL_868>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "869": {
+      "content": "<SPECIAL_869>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "870": {
+      "content": "<SPECIAL_870>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "871": {
+      "content": "<SPECIAL_871>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "872": {
+      "content": "<SPECIAL_872>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "873": {
+      "content": "<SPECIAL_873>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "874": {
+      "content": "<SPECIAL_874>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "875": {
+      "content": "<SPECIAL_875>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "876": {
+      "content": "<SPECIAL_876>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "877": {
+      "content": "<SPECIAL_877>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "878": {
+      "content": "<SPECIAL_878>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "879": {
+      "content": "<SPECIAL_879>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "880": {
+      "content": "<SPECIAL_880>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "881": {
+      "content": "<SPECIAL_881>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "882": {
+      "content": "<SPECIAL_882>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "883": {
+      "content": "<SPECIAL_883>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "884": {
+      "content": "<SPECIAL_884>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "885": {
+      "content": "<SPECIAL_885>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "886": {
+      "content": "<SPECIAL_886>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "887": {
+      "content": "<SPECIAL_887>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "888": {
+      "content": "<SPECIAL_888>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "889": {
+      "content": "<SPECIAL_889>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "890": {
+      "content": "<SPECIAL_890>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "891": {
+      "content": "<SPECIAL_891>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "892": {
+      "content": "<SPECIAL_892>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "893": {
+      "content": "<SPECIAL_893>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "894": {
+      "content": "<SPECIAL_894>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "895": {
+      "content": "<SPECIAL_895>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "896": {
+      "content": "<SPECIAL_896>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "897": {
+      "content": "<SPECIAL_897>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "898": {
+      "content": "<SPECIAL_898>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "899": {
+      "content": "<SPECIAL_899>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "900": {
+      "content": "<SPECIAL_900>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "901": {
+      "content": "<SPECIAL_901>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "902": {
+      "content": "<SPECIAL_902>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "903": {
+      "content": "<SPECIAL_903>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "904": {
+      "content": "<SPECIAL_904>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "905": {
+      "content": "<SPECIAL_905>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "906": {
+      "content": "<SPECIAL_906>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "907": {
+      "content": "<SPECIAL_907>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "908": {
+      "content": "<SPECIAL_908>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "909": {
+      "content": "<SPECIAL_909>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "910": {
+      "content": "<SPECIAL_910>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "911": {
+      "content": "<SPECIAL_911>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "912": {
+      "content": "<SPECIAL_912>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "913": {
+      "content": "<SPECIAL_913>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "914": {
+      "content": "<SPECIAL_914>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "915": {
+      "content": "<SPECIAL_915>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "916": {
+      "content": "<SPECIAL_916>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "917": {
+      "content": "<SPECIAL_917>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "918": {
+      "content": "<SPECIAL_918>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "919": {
+      "content": "<SPECIAL_919>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "920": {
+      "content": "<SPECIAL_920>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "921": {
+      "content": "<SPECIAL_921>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "922": {
+      "content": "<SPECIAL_922>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "923": {
+      "content": "<SPECIAL_923>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "924": {
+      "content": "<SPECIAL_924>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "925": {
+      "content": "<SPECIAL_925>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "926": {
+      "content": "<SPECIAL_926>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "927": {
+      "content": "<SPECIAL_927>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "928": {
+      "content": "<SPECIAL_928>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "929": {
+      "content": "<SPECIAL_929>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "930": {
+      "content": "<SPECIAL_930>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "931": {
+      "content": "<SPECIAL_931>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "932": {
+      "content": "<SPECIAL_932>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "933": {
+      "content": "<SPECIAL_933>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "934": {
+      "content": "<SPECIAL_934>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "935": {
+      "content": "<SPECIAL_935>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "936": {
+      "content": "<SPECIAL_936>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "937": {
+      "content": "<SPECIAL_937>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "938": {
+      "content": "<SPECIAL_938>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "939": {
+      "content": "<SPECIAL_939>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "940": {
+      "content": "<SPECIAL_940>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "941": {
+      "content": "<SPECIAL_941>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "942": {
+      "content": "<SPECIAL_942>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "943": {
+      "content": "<SPECIAL_943>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "944": {
+      "content": "<SPECIAL_944>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "945": {
+      "content": "<SPECIAL_945>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "946": {
+      "content": "<SPECIAL_946>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "947": {
+      "content": "<SPECIAL_947>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "948": {
+      "content": "<SPECIAL_948>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "949": {
+      "content": "<SPECIAL_949>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "950": {
+      "content": "<SPECIAL_950>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "951": {
+      "content": "<SPECIAL_951>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "952": {
+      "content": "<SPECIAL_952>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "953": {
+      "content": "<SPECIAL_953>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "954": {
+      "content": "<SPECIAL_954>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "955": {
+      "content": "<SPECIAL_955>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "956": {
+      "content": "<SPECIAL_956>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "957": {
+      "content": "<SPECIAL_957>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "958": {
+      "content": "<SPECIAL_958>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "959": {
+      "content": "<SPECIAL_959>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "960": {
+      "content": "<SPECIAL_960>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "961": {
+      "content": "<SPECIAL_961>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "962": {
+      "content": "<SPECIAL_962>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "963": {
+      "content": "<SPECIAL_963>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "964": {
+      "content": "<SPECIAL_964>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "965": {
+      "content": "<SPECIAL_965>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "966": {
+      "content": "<SPECIAL_966>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "967": {
+      "content": "<SPECIAL_967>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "968": {
+      "content": "<SPECIAL_968>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "969": {
+      "content": "<SPECIAL_969>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "970": {
+      "content": "<SPECIAL_970>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "971": {
+      "content": "<SPECIAL_971>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "972": {
+      "content": "<SPECIAL_972>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "973": {
+      "content": "<SPECIAL_973>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "974": {
+      "content": "<SPECIAL_974>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "975": {
+      "content": "<SPECIAL_975>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "976": {
+      "content": "<SPECIAL_976>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "977": {
+      "content": "<SPECIAL_977>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "978": {
+      "content": "<SPECIAL_978>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "979": {
+      "content": "<SPECIAL_979>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "980": {
+      "content": "<SPECIAL_980>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "981": {
+      "content": "<SPECIAL_981>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "982": {
+      "content": "<SPECIAL_982>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "983": {
+      "content": "<SPECIAL_983>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "984": {
+      "content": "<SPECIAL_984>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "985": {
+      "content": "<SPECIAL_985>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "986": {
+      "content": "<SPECIAL_986>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "987": {
+      "content": "<SPECIAL_987>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "988": {
+      "content": "<SPECIAL_988>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "989": {
+      "content": "<SPECIAL_989>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "990": {
+      "content": "<SPECIAL_990>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "991": {
+      "content": "<SPECIAL_991>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "992": {
+      "content": "<SPECIAL_992>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "993": {
+      "content": "<SPECIAL_993>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "994": {
+      "content": "<SPECIAL_994>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "995": {
+      "content": "<SPECIAL_995>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "996": {
+      "content": "<SPECIAL_996>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "997": {
+      "content": "<SPECIAL_997>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "998": {
+      "content": "<SPECIAL_998>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "999": {
+      "content": "<SPECIAL_999>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<s>",
+  "chat_template": "{%- if messages[0][\"role\"] == \"system\" %}\n    {%- set system_message = messages[0][\"content\"] %}\n    {%- set loop_messages = messages[1:] %}\n{%- else %}\n    {%- set loop_messages = messages %}\n{%- endif %}\n{%- if not tools is defined %}\n    {%- set tools = none %}\n{%- endif %}\n{%- set user_messages = loop_messages | selectattr(\"role\", \"equalto\", \"user\") | list %}\n\n{#- This block checks for alternating user/assistant messages, skipping tool calling messages #}\n{%- set ns = namespace() %}\n{%- set ns.index = 0 %}\n{%- for message in loop_messages %}\n    {%- if not (message.role == \"tool\" or message.role == \"tool_results\" or (message.tool_calls is defined and message.tool_calls is not none)) %}\n        {%- if (message[\"role\"] == \"user\") != (ns.index % 2 == 0) %}\n            {{- raise_exception(\"After the optional system message, conversation roles must alternate user/assistant/user/assistant/...\") }}\n        {%- endif %}\n        {%- set ns.index = ns.index + 1 %}\n    {%- endif %}\n{%- endfor %}\n\n{{- bos_token }}\n{%- for message in loop_messages %}\n    {%- if message[\"role\"] == \"user\" %}\n        {%- if tools is not none and (message == user_messages[-1]) %}\n            {{- \"[AVAILABLE_TOOLS][\" }}\n            {%- for tool in tools %}\n                {%- set tool = tool.function %}\n                {{- '{\"type\": \"function\", \"function\": {' }}\n                {%- for key, val in tool.items() if key != \"return\" %}\n                    {%- if val is string %}\n                        {{- '\"' + key + '\": \"' + val + '\"' }}\n                    {%- else %}\n                        {{- '\"' + key + '\": ' + val|tojson }}\n                    {%- endif %}\n                    {%- if not loop.last %}\n                        {{- \", \" }}\n                    {%- endif %}\n                {%- endfor %}\n                {{- \"}}\" }}\n                {%- if not loop.last %}\n                    {{- \", \" }}\n                {%- else %}\n                    {{- \"]\" }}\n                {%- endif %}\n            {%- endfor %}\n            {{- \"[/AVAILABLE_TOOLS]\" }}\n            {%- endif %}\n        {%- if loop.last and system_message is defined %}\n            {{- \"[INST]\" + system_message + \"\\n\\n\" + message[\"content\"] + \"[/INST]\" }}\n        {%- else %}\n            {{- \"[INST]\" + message[\"content\"] + \"[/INST]\" }}\n        {%- endif %}\n    {%- elif (message.tool_calls is defined and message.tool_calls is not none) %}\n        {{- \"[TOOL_CALLS][\" }}\n        {%- for tool_call in message.tool_calls %}\n            {%- set out = tool_call.function|tojson %}\n            {{- out[:-1] }}\n            {%- if not tool_call.id is defined or tool_call.id|length != 9 %}\n                {{- raise_exception(\"Tool call IDs should be alphanumeric strings with length 9!\") }}\n            {%- endif %}\n            {{- ', \"id\": \"' + tool_call.id + '\"}' }}\n            {%- if not loop.last %}\n                {{- \", \" }}\n            {%- else %}\n                {{- \"]\" + eos_token }}\n            {%- endif %}\n        {%- endfor %}\n    {%- elif message[\"role\"] == \"assistant\" %}\n        {{- message[\"content\"] + eos_token}}\n    {%- elif message[\"role\"] == \"tool_results\" or message[\"role\"] == \"tool\" %}\n        {%- if message.content is defined and message.content.content is defined %}\n            {%- set content = message.content.content %}\n        {%- else %}\n            {%- set content = message.content %}\n        {%- endif %}\n        {{- '[TOOL_RESULTS]{\"content\": ' + content|string + \", \" }}\n        {%- if not message.tool_call_id is defined or message.tool_call_id|length != 9 %}\n            {{- raise_exception(\"Tool call IDs should be alphanumeric strings with length 9!\") }}\n        {%- endif %}\n        {{- '\"call_id\": \"' + message.tool_call_id + '\"}[/TOOL_RESULTS]' }}\n    {%- else %}\n        {{- raise_exception(\"Only user and assistant roles are supported, with the exception of an initial optional system message!\") }}\n    {%- endif %}\n{%- endfor %}\n",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "</s>",
+  "extra_special_tokens": {},
+  "model_input_names": [
+    "input_ids",
+    "attention_mask"
+  ],
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "</s>",
+  "tokenizer_class": "PreTrainedTokenizer",
+  "unk_token": "<unk>"
+}
diff --git a/checkpoint-800/trainer_state.json b/checkpoint-800/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..abf61eb3c9f3c964d51c00b28a41d8bc30a58ca8
--- /dev/null
+++ b/checkpoint-800/trainer_state.json
@@ -0,0 +1,10034 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 1.1202797202797203,
+  "eval_steps": 500,
+  "global_step": 800,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 94.14286041259766,
+      "epoch": 0.0013986013986013986,
+      "grad_norm": 0.09098726750251449,
+      "kl": 0.0003147125244140625,
+      "learning_rate": 3.496503496503497e-07,
+      "loss": -0.0001,
+      "num_tokens": 19338.0,
+      "reward": 1.3853150606155396,
+      "reward_std": 0.4827312231063843,
+      "rewards/check_gptzero_func": 0.3214285671710968,
+      "rewards/check_perplexity_diff_func": 0.6428571343421936,
+      "rewards/check_winston_local_func": 0.42102929949760437,
+      "step": 1
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.002797202797202797,
+      "grad_norm": 0.09100260462019598,
+      "kl": 0.0003147125244140625,
+      "learning_rate": 6.993006993006994e-07,
+      "loss": -0.0001,
+      "step": 2
+    },
+    {
+      "clip_ratio": 0.0031990089919418097,
+      "completion_length": 97.73214721679688,
+      "epoch": 0.004195804195804196,
+      "grad_norm": 0.16978190242842756,
+      "kl": 0.0003662109375,
+      "learning_rate": 1.0489510489510491e-06,
+      "loss": -0.0017,
+      "num_tokens": 39487.0,
+      "reward": 0.9710169434547424,
+      "reward_std": 0.6376833319664001,
+      "rewards/check_gptzero_func": 0.125,
+      "rewards/check_perplexity_diff_func": 0.5,
+      "rewards/check_winston_local_func": 0.34601688385009766,
+      "step": 3
+    },
+    {
+      "clip_ratio": 0.0018373647471889853,
+      "epoch": 0.005594405594405594,
+      "grad_norm": 0.17663660241053314,
+      "kl": 0.000423431396484375,
+      "learning_rate": 1.3986013986013987e-06,
+      "loss": -0.0013,
+      "step": 4
+    },
+    {
+      "clip_ratio": 0.0016871786210685968,
+      "completion_length": 72.83928680419922,
+      "epoch": 0.006993006993006993,
+      "grad_norm": 0.09466520543424979,
+      "kl": 0.00034332275390625,
+      "learning_rate": 1.7482517482517485e-06,
+      "loss": 0.0008,
+      "num_tokens": 56082.0,
+      "reward": 1.0736479759216309,
+      "reward_std": 0.6862176656723022,
+      "rewards/check_gptzero_func": 0.2321428507566452,
+      "rewards/check_perplexity_diff_func": 0.3928571343421936,
+      "rewards/check_winston_local_func": 0.4486479163169861,
+      "step": 5
+    },
+    {
+      "clip_ratio": 0.0010098000057041645,
+      "epoch": 0.008391608391608392,
+      "grad_norm": 0.09539550614656178,
+      "kl": 0.0003376007080078125,
+      "learning_rate": 2.0979020979020983e-06,
+      "loss": 0.001,
+      "step": 6
+    },
+    {
+      "clip_ratio": 0.00184189947322011,
+      "completion_length": 95.92857360839844,
+      "epoch": 0.009790209790209791,
+      "grad_norm": 0.10573631695411624,
+      "kl": 0.0003757476806640625,
+      "learning_rate": 2.4475524475524477e-06,
+      "loss": 0.001,
+      "num_tokens": 75906.0,
+      "reward": 1.135968804359436,
+      "reward_std": 0.48974505066871643,
+      "rewards/check_gptzero_func": 0.2142857164144516,
+      "rewards/check_perplexity_diff_func": 0.5,
+      "rewards/check_winston_local_func": 0.42168304324150085,
+      "step": 7
+    },
+    {
+      "clip_ratio": 0.0018530809320509434,
+      "epoch": 0.011188811188811189,
+      "grad_norm": 0.10614532740192534,
+      "kl": 0.0004119873046875,
+      "learning_rate": 2.7972027972027974e-06,
+      "loss": 0.0011,
+      "step": 8
+    },
+    {
+      "clip_ratio": 0.0017129909247159958,
+      "completion_length": 106.76786041259766,
+      "epoch": 0.012587412587412588,
+      "grad_norm": 0.07916450383171113,
+      "kl": 0.000377655029296875,
+      "learning_rate": 3.1468531468531472e-06,
+      "loss": -0.0012,
+      "num_tokens": 96989.0,
+      "reward": 0.9590713977813721,
+      "reward_std": 0.5206417441368103,
+      "rewards/check_gptzero_func": 0.2321428507566452,
+      "rewards/check_perplexity_diff_func": 0.3928571343421936,
+      "rewards/check_winston_local_func": 0.3340713679790497,
+      "step": 9
+    },
+    {
+      "clip_ratio": 0.0034840807784348726,
+      "epoch": 0.013986013986013986,
+      "grad_norm": 0.08061029037170311,
+      "kl": 0.00037384033203125,
+      "learning_rate": 3.496503496503497e-06,
+      "loss": -0.0008,
+      "step": 10
+    },
+    {
+      "clip_ratio": 0.0017933619674295187,
+      "completion_length": 102.85714721679688,
+      "epoch": 0.015384615384615385,
+      "grad_norm": 0.07930375667041344,
+      "kl": 0.00030517578125,
+      "learning_rate": 3.846153846153847e-06,
+      "loss": -0.0038,
+      "num_tokens": 117271.0,
+      "reward": 1.1903148889541626,
+      "reward_std": 0.4722224473953247,
+      "rewards/check_gptzero_func": 0.2857142984867096,
+      "rewards/check_perplexity_diff_func": 0.5357142686843872,
+      "rewards/check_winston_local_func": 0.368886262178421,
+      "step": 11
+    },
+    {
+      "clip_ratio": 0.0015239757485687733,
+      "epoch": 0.016783216783216783,
+      "grad_norm": 0.07898151410343454,
+      "kl": 0.0003871917724609375,
+      "learning_rate": 4.195804195804197e-06,
+      "loss": -0.0041,
+      "step": 12
+    },
+    {
+      "clip_ratio": 0.0029670600779354572,
+      "completion_length": 104.37500762939453,
+      "epoch": 0.01818181818181818,
+      "grad_norm": 0.09247854491588844,
+      "kl": 0.0003986358642578125,
+      "learning_rate": 4.5454545454545455e-06,
+      "loss": -0.0049,
+      "num_tokens": 138174.0,
+      "reward": 0.9787324070930481,
+      "reward_std": 0.63657546043396,
+      "rewards/check_gptzero_func": 0.1964285671710968,
+      "rewards/check_perplexity_diff_func": 0.5714285969734192,
+      "rewards/check_winston_local_func": 0.21087531745433807,
+      "step": 13
+    },
+    {
+      "clip_ratio": 0.003217793768271804,
+      "epoch": 0.019580419580419582,
+      "grad_norm": 0.09293749933034297,
+      "kl": 0.0003204345703125,
+      "learning_rate": 4.895104895104895e-06,
+      "loss": -0.0045,
+      "step": 14
+    },
+    {
+      "clip_ratio": 0.002677519340068102,
+      "completion_length": 67.41072082519531,
+      "epoch": 0.02097902097902098,
+      "grad_norm": 0.14020673218962584,
+      "kl": 0.000499725341796875,
+      "learning_rate": 5.244755244755245e-06,
+      "loss": -0.008,
+      "num_tokens": 154385.0,
+      "reward": 1.0573337078094482,
+      "reward_std": 0.5388314127922058,
+      "rewards/check_gptzero_func": 0.2857142984867096,
+      "rewards/check_perplexity_diff_func": 0.3571428656578064,
+      "rewards/check_winston_local_func": 0.41447654366493225,
+      "step": 15
+    },
+    {
+      "clip_ratio": 0.0021408216562122107,
+      "epoch": 0.022377622377622378,
+      "grad_norm": 0.16110555979669544,
+      "kl": 0.000530242919921875,
+      "learning_rate": 5.594405594405595e-06,
+      "loss": -0.0081,
+      "step": 16
+    },
+    {
+      "clip_ratio": 0.002852825215086341,
+      "completion_length": 89.83928680419922,
+      "epoch": 0.023776223776223775,
+      "grad_norm": 0.1066932843002634,
+      "kl": 0.000415802001953125,
+      "learning_rate": 5.944055944055945e-06,
+      "loss": 0.0166,
+      "num_tokens": 173158.0,
+      "reward": 1.2226747274398804,
+      "reward_std": 0.45578521490097046,
+      "rewards/check_gptzero_func": 0.3392857015132904,
+      "rewards/check_perplexity_diff_func": 0.5,
+      "rewards/check_winston_local_func": 0.3833889663219452,
+      "step": 17
+    },
+    {
+      "clip_ratio": 0.002018069615587592,
+      "epoch": 0.025174825174825177,
+      "grad_norm": 0.10578254537451186,
+      "kl": 0.0003833770751953125,
+      "learning_rate": 6.2937062937062944e-06,
+      "loss": 0.0164,
+      "step": 18
+    },
+    {
+      "clip_ratio": 0.0032663617748767138,
+      "completion_length": 105.83928680419922,
+      "epoch": 0.026573426573426574,
+      "grad_norm": 0.11554707163597037,
+      "kl": 0.0003719329833984375,
+      "learning_rate": 6.643356643356643e-06,
+      "loss": -0.0005,
+      "num_tokens": 194649.0,
+      "reward": 0.8957912921905518,
+      "reward_std": 0.47702187299728394,
+      "rewards/check_gptzero_func": 0.1785714328289032,
+      "rewards/check_perplexity_diff_func": 0.3571428656578064,
+      "rewards/check_winston_local_func": 0.3600769340991974,
+      "step": 19
+    },
+    {
+      "clip_ratio": 0.0024422537535429,
+      "epoch": 0.027972027972027972,
+      "grad_norm": 0.11417666172406592,
+      "kl": 0.0003814697265625,
+      "learning_rate": 6.993006993006994e-06,
+      "loss": -0.0008,
+      "step": 20
+    },
+    {
+      "clip_ratio": 0.0013639701064676046,
+      "completion_length": 107.39286041259766,
+      "epoch": 0.02937062937062937,
+      "grad_norm": 0.09634196153318929,
+      "kl": 0.000396728515625,
+      "learning_rate": 7.342657342657343e-06,
+      "loss": -0.0011,
+      "num_tokens": 215987.0,
+      "reward": 0.9197577238082886,
+      "reward_std": 0.39100831747055054,
+      "rewards/check_gptzero_func": 0.2321428507566452,
+      "rewards/check_perplexity_diff_func": 0.3571428656578064,
+      "rewards/check_winston_local_func": 0.33047202229499817,
+      "step": 21
+    },
+    {
+      "clip_ratio": 0.003595889313146472,
+      "epoch": 0.03076923076923077,
+      "grad_norm": 0.09779857975187953,
+      "kl": 0.00052642822265625,
+      "learning_rate": 7.692307692307694e-06,
+      "loss": -0.0013,
+      "step": 22
+    },
+    {
+      "clip_ratio": 0.0024316231720149517,
+      "completion_length": 89.08928680419922,
+      "epoch": 0.032167832167832165,
+      "grad_norm": 0.13189477135999747,
+      "kl": 0.0003414154052734375,
+      "learning_rate": 8.041958041958042e-06,
+      "loss": -0.0009,
+      "num_tokens": 234678.0,
+      "reward": 0.9593304395675659,
+      "reward_std": 0.5972030758857727,
+      "rewards/check_gptzero_func": 0.1964285671710968,
+      "rewards/check_perplexity_diff_func": 0.4285714328289032,
+      "rewards/check_winston_local_func": 0.33433040976524353,
+      "step": 23
+    },
+    {
+      "clip_ratio": 0.000728283659555018,
+      "epoch": 0.033566433566433566,
+      "grad_norm": 0.12953291972672934,
+      "kl": 0.000335693359375,
+      "learning_rate": 8.391608391608393e-06,
+      "loss": -0.0011,
+      "step": 24
+    },
+    {
+      "clip_ratio": 0.0021546650677919388,
+      "completion_length": 90.98214721679688,
+      "epoch": 0.03496503496503497,
+      "grad_norm": 0.07682474826469807,
+      "kl": 0.0004558563232421875,
+      "learning_rate": 8.741258741258741e-06,
+      "loss": 0.0077,
+      "num_tokens": 253833.0,
+      "reward": 0.8255766034126282,
+      "reward_std": 0.3988816440105438,
+      "rewards/check_gptzero_func": 0.2678571343421936,
+      "rewards/check_perplexity_diff_func": 0.2857142984867096,
+      "rewards/check_winston_local_func": 0.272005170583725,
+      "step": 25
+    },
+    {
+      "clip_ratio": 0.0019250252516940236,
+      "epoch": 0.03636363636363636,
+      "grad_norm": 0.07873511079065725,
+      "kl": 0.0003490447998046875,
+      "learning_rate": 9.090909090909091e-06,
+      "loss": 0.0075,
+      "step": 26
+    },
+    {
+      "clip_ratio": 0.001886485842987895,
+      "completion_length": 74.4464340209961,
+      "epoch": 0.03776223776223776,
+      "grad_norm": 0.06765277577538054,
+      "kl": 0.000354766845703125,
+      "learning_rate": 9.44055944055944e-06,
+      "loss": -0.0003,
+      "num_tokens": 270908.0,
+      "reward": 1.3086735010147095,
+      "reward_std": 0.3881511986255646,
+      "rewards/check_gptzero_func": 0.4107142984867096,
+      "rewards/check_perplexity_diff_func": 0.4642857015132904,
+      "rewards/check_winston_local_func": 0.4336733818054199,
+      "step": 27
+    },
+    {
+      "clip_ratio": 0.0020898371003568172,
+      "epoch": 0.039160839160839164,
+      "grad_norm": 0.06722581226675682,
+      "kl": 0.000453948974609375,
+      "learning_rate": 9.79020979020979e-06,
+      "loss": -0.0006,
+      "step": 28
+    },
+    {
+      "clip_ratio": 0.002855924190953374,
+      "completion_length": 103.64286041259766,
+      "epoch": 0.04055944055944056,
+      "grad_norm": 0.16301390061045865,
+      "kl": 0.0003948211669921875,
+      "learning_rate": 1.013986013986014e-05,
+      "loss": 0.0063,
+      "num_tokens": 292060.0,
+      "reward": 1.22488272190094,
+      "reward_std": 0.8093323111534119,
+      "rewards/check_gptzero_func": 0.2678571343421936,
+      "rewards/check_perplexity_diff_func": 0.6071428656578064,
+      "rewards/check_winston_local_func": 0.34988275170326233,
+      "step": 29
+    },
+    {
+      "clip_ratio": 0.0016208746237680316,
+      "epoch": 0.04195804195804196,
+      "grad_norm": 0.17335125351231856,
+      "kl": 0.000507354736328125,
+      "learning_rate": 1.048951048951049e-05,
+      "loss": 0.0059,
+      "step": 30
+    },
+    {
+      "clip_ratio": 0.0011572305811569095,
+      "completion_length": 82.91072082519531,
+      "epoch": 0.043356643356643354,
+      "grad_norm": 0.11830097508590764,
+      "kl": 0.0006256103515625,
+      "learning_rate": 1.0839160839160838e-05,
+      "loss": 0.002,
+      "num_tokens": 310163.0,
+      "reward": 0.9069231748580933,
+      "reward_std": 0.4621748924255371,
+      "rewards/check_gptzero_func": 0.25,
+      "rewards/check_perplexity_diff_func": 0.3571428656578064,
+      "rewards/check_winston_local_func": 0.29978030920028687,
+      "step": 31
+    },
+    {
+      "clip_ratio": 0.00402362085878849,
+      "epoch": 0.044755244755244755,
+      "grad_norm": 0.11861334422671065,
+      "kl": 0.000762939453125,
+      "learning_rate": 1.118881118881119e-05,
+      "loss": 0.0019,
+      "step": 32
+    },
+    {
+      "clip_ratio": 0.0017474278574809432,
+      "completion_length": 90.9464340209961,
+      "epoch": 0.046153846153846156,
+      "grad_norm": 0.1025496804751534,
+      "kl": 0.000701904296875,
+      "learning_rate": 1.153846153846154e-05,
+      "loss": -0.0014,
+      "num_tokens": 328922.0,
+      "reward": 1.1198338270187378,
+      "reward_std": 0.5045111775398254,
+      "rewards/check_gptzero_func": 0.0892857164144516,
+      "rewards/check_perplexity_diff_func": 0.7857142686843872,
+      "rewards/check_winston_local_func": 0.24483375251293182,
+      "step": 33
+    },
+    {
+      "clip_ratio": 0.0012307984288781881,
+      "epoch": 0.04755244755244755,
+      "grad_norm": 0.10274008519956918,
+      "kl": 0.00087738037109375,
+      "learning_rate": 1.188811188811189e-05,
+      "loss": -0.0015,
+      "step": 34
+    },
+    {
+      "clip_ratio": 0.0029886537231504917,
+      "completion_length": 93.00000762939453,
+      "epoch": 0.04895104895104895,
+      "grad_norm": 0.1311802256012859,
+      "kl": 0.001190185546875,
+      "learning_rate": 1.2237762237762239e-05,
+      "loss": -0.0141,
+      "num_tokens": 348852.0,
+      "reward": 0.9456299543380737,
+      "reward_std": 0.6374980807304382,
+      "rewards/check_gptzero_func": 0.2321428507566452,
+      "rewards/check_perplexity_diff_func": 0.4285714328289032,
+      "rewards/check_winston_local_func": 0.28491565585136414,
+      "step": 35
+    },
+    {
+      "clip_ratio": 0.0028984802775084972,
+      "epoch": 0.05034965034965035,
+      "grad_norm": 0.13267305607921692,
+      "kl": 0.00145721435546875,
+      "learning_rate": 1.2587412587412589e-05,
+      "loss": -0.0147,
+      "step": 36
+    },
+    {
+      "clip_ratio": 0.0018746532732620835,
+      "completion_length": 89.85714721679688,
+      "epoch": 0.05174825174825175,
+      "grad_norm": 0.1497182815907709,
+      "kl": 0.002044677734375,
+      "learning_rate": 1.2937062937062939e-05,
+      "loss": -0.0253,
+      "num_tokens": 367482.0,
+      "reward": 1.2605940103530884,
+      "reward_std": 0.66156405210495,
+      "rewards/check_gptzero_func": 0.1785714328289032,
+      "rewards/check_perplexity_diff_func": 0.75,
+      "rewards/check_winston_local_func": 0.3320225179195404,
+      "step": 37
+    },
+    {
+      "clip_ratio": 0.0020914783235639334,
+      "epoch": 0.05314685314685315,
+      "grad_norm": 0.15027543310666272,
+      "kl": 0.0026092529296875,
+      "learning_rate": 1.3286713286713287e-05,
+      "loss": -0.0257,
+      "step": 38
+    },
+    {
+      "clip_ratio": 0.0017028081929311156,
+      "completion_length": 101.58928680419922,
+      "epoch": 0.05454545454545454,
+      "grad_norm": 0.1306748742984494,
+      "kl": 0.0027008056640625,
+      "learning_rate": 1.3636363636363637e-05,
+      "loss": -0.0091,
+      "num_tokens": 388365.0,
+      "reward": 1.1187876462936401,
+      "reward_std": 0.4044211804866791,
+      "rewards/check_gptzero_func": 0.2321428507566452,
+      "rewards/check_perplexity_diff_func": 0.5357142686843872,
+      "rewards/check_winston_local_func": 0.3509305417537689,
+      "step": 39
+    },
+    {
+      "clip_ratio": 0.003386714030057192,
+      "epoch": 0.055944055944055944,
+      "grad_norm": 0.12880289464580882,
+      "kl": 0.003265380859375,
+      "learning_rate": 1.3986013986013988e-05,
+      "loss": -0.0086,
+      "step": 40
+    },
+    {
+      "clip_ratio": 0.0024523374158889055,
+      "completion_length": 100.41072082519531,
+      "epoch": 0.057342657342657345,
+      "grad_norm": 0.12385816200639352,
+      "kl": 0.004547119140625,
+      "learning_rate": 1.4335664335664336e-05,
+      "loss": -0.0161,
+      "num_tokens": 409074.0,
+      "reward": 0.804709792137146,
+      "reward_std": 0.5129754543304443,
+      "rewards/check_gptzero_func": 0.1428571492433548,
+      "rewards/check_perplexity_diff_func": 0.3928571343421936,
+      "rewards/check_winston_local_func": 0.2689954936504364,
+      "step": 41
+    },
+    {
+      "clip_ratio": 0.004267544951289892,
+      "epoch": 0.05874125874125874,
+      "grad_norm": 0.12301160152684054,
+      "kl": 0.005218505859375,
+      "learning_rate": 1.4685314685314686e-05,
+      "loss": -0.0161,
+      "step": 42
+    },
+    {
+      "clip_ratio": 0.0019341636216267943,
+      "completion_length": 90.46428680419922,
+      "epoch": 0.06013986013986014,
+      "grad_norm": 0.1390908713431379,
+      "kl": 0.0048828125,
+      "learning_rate": 1.5034965034965034e-05,
+      "loss": 0.0068,
+      "num_tokens": 428262.0,
+      "reward": 1.087609887123108,
+      "reward_std": 0.6448312997817993,
+      "rewards/check_gptzero_func": 0.2321428507566452,
+      "rewards/check_perplexity_diff_func": 0.3928571343421936,
+      "rewards/check_winston_local_func": 0.46260982751846313,
+      "step": 43
+    },
+    {
+      "clip_ratio": 0.0016832423862069845,
+      "epoch": 0.06153846153846154,
+      "grad_norm": 0.13873080950458352,
+      "kl": 0.00567626953125,
+      "learning_rate": 1.5384615384615387e-05,
+      "loss": 0.0065,
+      "step": 44
+    },
+    {
+      "clip_ratio": 0.0019954824820160866,
+      "completion_length": 96.92857360839844,
+      "epoch": 0.06293706293706294,
+      "grad_norm": 0.11630124487143968,
+      "kl": 0.00897216796875,
+      "learning_rate": 1.5734265734265734e-05,
+      "loss": 0.008,
+      "num_tokens": 448480.0,
+      "reward": 1.0604994297027588,
+      "reward_std": 0.5378546118736267,
+      "rewards/check_gptzero_func": 0.2321428507566452,
+      "rewards/check_perplexity_diff_func": 0.4642857015132904,
+      "rewards/check_winston_local_func": 0.36407074332237244,
+      "step": 45
+    },
+    {
+      "clip_ratio": 0.002663462422788143,
+      "epoch": 0.06433566433566433,
+      "grad_norm": 0.1149796219701467,
+      "kl": 0.01007080078125,
+      "learning_rate": 1.6083916083916083e-05,
+      "loss": 0.0081,
+      "step": 46
+    },
+    {
+      "clip_ratio": 0.0018039485439658165,
+      "completion_length": 79.51786041259766,
+      "epoch": 0.06573426573426573,
+      "grad_norm": 0.12721969778213826,
+      "kl": 0.01123046875,
+      "learning_rate": 1.6433566433566433e-05,
+      "loss": 0.0112,
+      "num_tokens": 465671.0,
+      "reward": 0.9921315312385559,
+      "reward_std": 0.38579052686691284,
+      "rewards/check_gptzero_func": 0.1607142835855484,
+      "rewards/check_perplexity_diff_func": 0.3928571343421936,
+      "rewards/check_winston_local_func": 0.43856000900268555,
+      "step": 47
+    },
+    {
+      "clip_ratio": 0.0027954555116593838,
+      "epoch": 0.06713286713286713,
+      "grad_norm": 0.12930272448754576,
+      "kl": 0.01300048828125,
+      "learning_rate": 1.6783216783216786e-05,
+      "loss": 0.0107,
+      "step": 48
+    },
+    {
+      "clip_ratio": 0.0028269642498344183,
+      "completion_length": 116.5714340209961,
+      "epoch": 0.06853146853146853,
+      "grad_norm": 0.12430052891449103,
+      "kl": 0.01226806640625,
+      "learning_rate": 1.7132867132867133e-05,
+      "loss": 0.025,
+      "num_tokens": 488367.0,
+      "reward": 0.8830849528312683,
+      "reward_std": 0.5276607871055603,
+      "rewards/check_gptzero_func": 0.25,
+      "rewards/check_perplexity_diff_func": 0.3214285671710968,
+      "rewards/check_winston_local_func": 0.31165632605552673,
+      "step": 49
+    },
+    {
+      "clip_ratio": 0.0023628019262105227,
+      "epoch": 0.06993006993006994,
+      "grad_norm": 0.12293590613717971,
+      "kl": 0.0142822265625,
+      "learning_rate": 1.7482517482517483e-05,
+      "loss": 0.0249,
+      "step": 50
+    },
+    {
+      "clip_ratio": 0.0033425339497625828,
+      "completion_length": 83.39286041259766,
+      "epoch": 0.07132867132867132,
+      "grad_norm": 0.11106847343390151,
+      "kl": 0.01177978515625,
+      "learning_rate": 1.7832167832167836e-05,
+      "loss": -0.0041,
+      "num_tokens": 506551.0,
+      "reward": 1.029296875,
+      "reward_std": 0.4901208281517029,
+      "rewards/check_gptzero_func": 0.2857142984867096,
+      "rewards/check_perplexity_diff_func": 0.3214285671710968,
+      "rewards/check_winston_local_func": 0.42215394973754883,
+      "step": 51
+    },
+    {
+      "clip_ratio": 0.0022109400015324354,
+      "epoch": 0.07272727272727272,
+      "grad_norm": 0.11017795132415774,
+      "kl": 0.013671875,
+      "learning_rate": 1.8181818181818182e-05,
+      "loss": -0.0043,
+      "step": 52
+    },
+    {
+      "clip_ratio": 0.0019256824161857367,
+      "completion_length": 86.5714340209961,
+      "epoch": 0.07412587412587412,
+      "grad_norm": 0.11010712395827207,
+      "kl": 0.0208740234375,
+      "learning_rate": 1.8531468531468532e-05,
+      "loss": -0.0032,
+      "num_tokens": 524863.0,
+      "reward": 0.962996244430542,
+      "reward_std": 0.6607940196990967,
+      "rewards/check_gptzero_func": 0.1071428582072258,
+      "rewards/check_perplexity_diff_func": 0.5,
+      "rewards/check_winston_local_func": 0.355853408575058,
+      "step": 53
+    },
+    {
+      "clip_ratio": 0.0025832760147750378,
+      "epoch": 0.07552447552447553,
+      "grad_norm": 0.10695378216267973,
+      "kl": 0.0250244140625,
+      "learning_rate": 1.888111888111888e-05,
+      "loss": -0.0034,
+      "step": 54
+    },
+    {
+      "clip_ratio": 0.0017639577854424715,
+      "completion_length": 96.08928680419922,
+      "epoch": 0.07692307692307693,
+      "grad_norm": 0.17472941592839772,
+      "kl": 0.0322265625,
+      "learning_rate": 1.923076923076923e-05,
+      "loss": -0.0141,
+      "num_tokens": 544736.0,
+      "reward": 1.3478963375091553,
+      "reward_std": 0.8002303242683411,
+      "rewards/check_gptzero_func": 0.2857142984867096,
+      "rewards/check_perplexity_diff_func": 0.7142857313156128,
+      "rewards/check_winston_local_func": 0.34789615869522095,
+      "step": 55
+    },
+    {
+      "clip_ratio": 0.0034089265391230583,
+      "epoch": 0.07832167832167833,
+      "grad_norm": 0.14666884899599142,
+      "kl": 0.041748046875,
+      "learning_rate": 1.958041958041958e-05,
+      "loss": -0.0155,
+      "step": 56
+    },
+    {
+      "clip_ratio": 0.0026615143287926912,
+      "completion_length": 96.91072082519531,
+      "epoch": 0.07972027972027972,
+      "grad_norm": 0.1400337476406121,
+      "kl": 0.035400390625,
+      "learning_rate": 1.993006993006993e-05,
+      "loss": -0.027,
+      "num_tokens": 564803.0,
+      "reward": 1.0594887733459473,
+      "reward_std": 0.6912091970443726,
+      "rewards/check_gptzero_func": 0.2857142984867096,
+      "rewards/check_perplexity_diff_func": 0.4285714328289032,
+      "rewards/check_winston_local_func": 0.34520307183265686,
+      "step": 57
+    },
+    {
+      "clip_ratio": 0.003153084311634302,
+      "epoch": 0.08111888111888112,
+      "grad_norm": 0.138816359725178,
+      "kl": 0.042236328125,
+      "learning_rate": 2.027972027972028e-05,
+      "loss": -0.0274,
+      "step": 58
+    },
+    {
+      "clip_ratio": 0.0012626759707927704,
+      "completion_length": 76.35714721679688,
+      "epoch": 0.08251748251748252,
+      "grad_norm": 0.15940476459351663,
+      "kl": 0.0556640625,
+      "learning_rate": 2.062937062937063e-05,
+      "loss": -0.0057,
+      "num_tokens": 582265.0,
+      "reward": 0.992591381072998,
+      "reward_std": 0.5938137769699097,
+      "rewards/check_gptzero_func": 0.125,
+      "rewards/check_perplexity_diff_func": 0.4642857015132904,
+      "rewards/check_winston_local_func": 0.4033055901527405,
+      "step": 59
+    },
+    {
+      "clip_ratio": 0.004643784370273352,
+      "epoch": 0.08391608391608392,
+      "grad_norm": 0.15305834024157844,
+      "kl": 0.07421875,
+      "learning_rate": 2.097902097902098e-05,
+      "loss": -0.0064,
+      "step": 60
+    },
+    {
+      "clip_ratio": 0.0032869603019207716,
+      "completion_length": 103.33928680419922,
+      "epoch": 0.08531468531468532,
+      "grad_norm": 0.12260760902895611,
+      "kl": 0.0810546875,
+      "learning_rate": 2.132867132867133e-05,
+      "loss": 0.0066,
+      "num_tokens": 603454.0,
+      "reward": 1.191124439239502,
+      "reward_std": 0.5557684898376465,
+      "rewards/check_gptzero_func": 0.2321428507566452,
+      "rewards/check_perplexity_diff_func": 0.5357142686843872,
+      "rewards/check_winston_local_func": 0.4232672154903412,
+      "step": 61
+    },
+    {
+      "clip_ratio": 0.008058370091021061,
+      "epoch": 0.08671328671328671,
+      "grad_norm": 0.10224499878092017,
+      "kl": 0.1162109375,
+      "learning_rate": 2.1678321678321677e-05,
+      "loss": 0.0058,
+      "step": 62
+    },
+    {
+      "clip_ratio": 0.002202474046498537,
+      "completion_length": 91.4464340209961,
+      "epoch": 0.08811188811188811,
+      "grad_norm": 0.09776882919444682,
+      "kl": 0.2138671875,
+      "learning_rate": 2.202797202797203e-05,
+      "loss": 0.0021,
+      "num_tokens": 622705.0,
+      "reward": 1.1320958137512207,
+      "reward_std": 0.4060821831226349,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 0.3214285671710968,
+      "rewards/check_winston_local_func": 0.41781002283096313,
+      "step": 63
+    },
+    {
+      "clip_ratio": 0.0025903189089149237,
+      "epoch": 0.08951048951048951,
+      "grad_norm": 0.0963149975780195,
+      "kl": 0.2109375,
+      "learning_rate": 2.237762237762238e-05,
+      "loss": 0.0014,
+      "step": 64
+    },
+    {
+      "clip_ratio": 0.0016666523879393935,
+      "completion_length": 100.00000762939453,
+      "epoch": 0.09090909090909091,
+      "grad_norm": 0.12490001498293682,
+      "kl": 0.1220703125,
+      "learning_rate": 2.272727272727273e-05,
+      "loss": 0.0081,
+      "num_tokens": 643191.0,
+      "reward": 1.274580478668213,
+      "reward_std": 0.5277792811393738,
+      "rewards/check_gptzero_func": 0.2321428507566452,
+      "rewards/check_perplexity_diff_func": 0.5357142686843872,
+      "rewards/check_winston_local_func": 0.5067232847213745,
+      "step": 65
+    },
+    {
+      "clip_ratio": 0.0032784033101052046,
+      "epoch": 0.09230769230769231,
+      "grad_norm": 0.12114304348028067,
+      "kl": 0.134765625,
+      "learning_rate": 2.307692307692308e-05,
+      "loss": 0.0074,
+      "step": 66
+    },
+    {
+      "clip_ratio": 0.0017520035617053509,
+      "completion_length": 95.16072082519531,
+      "epoch": 0.0937062937062937,
+      "grad_norm": 0.14292853522837679,
+      "kl": 0.216796875,
+      "learning_rate": 2.342657342657343e-05,
+      "loss": -0.0018,
+      "num_tokens": 662736.0,
+      "reward": 1.1007850170135498,
+      "reward_std": 0.45985397696495056,
+      "rewards/check_gptzero_func": 0.25,
+      "rewards/check_perplexity_diff_func": 0.4642857015132904,
+      "rewards/check_winston_local_func": 0.38649922609329224,
+      "step": 67
+    },
+    {
+      "clip_ratio": 0.00579434959217906,
+      "epoch": 0.0951048951048951,
+      "grad_norm": 0.14340586759573087,
+      "kl": 0.2431640625,
+      "learning_rate": 2.377622377622378e-05,
+      "loss": -0.0029,
+      "step": 68
+    },
+    {
+      "clip_ratio": 0.0022546499967575073,
+      "completion_length": 86.17857360839844,
+      "epoch": 0.0965034965034965,
+      "grad_norm": 0.17287042335415018,
+      "kl": 0.26171875,
+      "learning_rate": 2.4125874125874125e-05,
+      "loss": 0.0059,
+      "num_tokens": 681030.0,
+      "reward": 1.4391711950302124,
+      "reward_std": 0.6516181230545044,
+      "rewards/check_gptzero_func": 0.5178571343421936,
+      "rewards/check_perplexity_diff_func": 0.4642857015132904,
+      "rewards/check_winston_local_func": 0.45702821016311646,
+      "step": 69
+    },
+    {
+      "clip_ratio": 0.004240955226123333,
+      "epoch": 0.0979020979020979,
+      "grad_norm": 0.16371494613893736,
+      "kl": 0.28125,
+      "learning_rate": 2.4475524475524478e-05,
+      "loss": 0.0048,
+      "step": 70
+    },
+    {
+      "clip_ratio": 0.003548440057784319,
+      "completion_length": 102.42857360839844,
+      "epoch": 0.0993006993006993,
+      "grad_norm": 0.1615418523026911,
+      "kl": 0.1708984375,
+      "learning_rate": 2.4825174825174828e-05,
+      "loss": 0.0015,
+      "num_tokens": 701874.0,
+      "reward": 1.4929367303848267,
+      "reward_std": 0.6083499193191528,
+      "rewards/check_gptzero_func": 0.3214285671710968,
+      "rewards/check_perplexity_diff_func": 0.7142857313156128,
+      "rewards/check_winston_local_func": 0.4572224020957947,
+      "step": 71
+    },
+    {
+      "clip_ratio": 0.004331245087087154,
+      "epoch": 0.1006993006993007,
+      "grad_norm": 0.12876017762018707,
+      "kl": 0.216796875,
+      "learning_rate": 2.5174825174825178e-05,
+      "loss": 0.0,
+      "step": 72
+    },
+    {
+      "clip_ratio": 0.003951852675527334,
+      "completion_length": 73.46428680419922,
+      "epoch": 0.1020979020979021,
+      "grad_norm": 0.20791748542634808,
+      "kl": 0.365234375,
+      "learning_rate": 2.5524475524475528e-05,
+      "loss": -0.0037,
+      "num_tokens": 718488.0,
+      "reward": 1.7953797578811646,
+      "reward_std": 0.5393027067184448,
+      "rewards/check_gptzero_func": 0.4464285671710968,
+      "rewards/check_perplexity_diff_func": 0.8571428656578064,
+      "rewards/check_winston_local_func": 0.49180838465690613,
+      "step": 73
+    },
+    {
+      "clip_ratio": 0.008312534540891647,
+      "epoch": 0.1034965034965035,
+      "grad_norm": 0.19232465302245624,
+      "kl": 0.5390625,
+      "learning_rate": 2.5874125874125877e-05,
+      "loss": -0.0053,
+      "step": 74
+    },
+    {
+      "clip_ratio": 0.0021736263297498226,
+      "completion_length": 97.73214721679688,
+      "epoch": 0.1048951048951049,
+      "grad_norm": 0.18504738647098173,
+      "kl": 0.2421875,
+      "learning_rate": 2.6223776223776224e-05,
+      "loss": -0.0023,
+      "num_tokens": 738619.0,
+      "reward": 1.284185767173767,
+      "reward_std": 0.6959513425827026,
+      "rewards/check_gptzero_func": 0.4464285671710968,
+      "rewards/check_perplexity_diff_func": 0.3928571343421936,
+      "rewards/check_winston_local_func": 0.4449000358581543,
+      "step": 75
+    },
+    {
+      "clip_ratio": 0.0033756059128791094,
+      "epoch": 0.1062937062937063,
+      "grad_norm": 0.17585639043073228,
+      "kl": 0.255859375,
+      "learning_rate": 2.6573426573426574e-05,
+      "loss": -0.0043,
+      "step": 76
+    },
+    {
+      "clip_ratio": 0.003740633837878704,
+      "completion_length": 95.67857360839844,
+      "epoch": 0.1076923076923077,
+      "grad_norm": 0.21573775826044048,
+      "kl": 0.2451171875,
+      "learning_rate": 2.6923076923076923e-05,
+      "loss": 0.0105,
+      "num_tokens": 758111.0,
+      "reward": 1.4672702550888062,
+      "reward_std": 0.85318922996521,
+      "rewards/check_gptzero_func": 0.25,
+      "rewards/check_perplexity_diff_func": 0.7857142686843872,
+      "rewards/check_winston_local_func": 0.43155592679977417,
+      "step": 77
+    },
+    {
+      "clip_ratio": 0.004517109598964453,
+      "epoch": 0.10909090909090909,
+      "grad_norm": 0.2071235680791843,
+      "kl": 0.302734375,
+      "learning_rate": 2.7272727272727273e-05,
+      "loss": 0.009,
+      "step": 78
+    },
+    {
+      "clip_ratio": 0.004641580395400524,
+      "completion_length": 117.75000762939453,
+      "epoch": 0.11048951048951049,
+      "grad_norm": 0.19278645376555428,
+      "kl": 0.259765625,
+      "learning_rate": 2.762237762237762e-05,
+      "loss": -0.0004,
+      "num_tokens": 780403.0,
+      "reward": 1.4974778890609741,
+      "reward_std": 0.59557044506073,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 0.6785714030265808,
+      "rewards/check_winston_local_func": 0.39033493399620056,
+      "step": 79
+    },
+    {
+      "clip_ratio": 0.006305322516709566,
+      "epoch": 0.11188811188811189,
+      "grad_norm": 0.13786850726930885,
+      "kl": 0.3125,
+      "learning_rate": 2.7972027972027976e-05,
+      "loss": -0.0019,
+      "step": 80
+    },
+    {
+      "clip_ratio": 0.0017778041074052453,
+      "completion_length": 101.03572082519531,
+      "epoch": 0.11328671328671329,
+      "grad_norm": 0.18099911272050398,
+      "kl": 1.7578125,
+      "learning_rate": 2.8321678321678326e-05,
+      "loss": 0.0053,
+      "num_tokens": 800231.0,
+      "reward": 1.420699954032898,
+      "reward_std": 0.5104399919509888,
+      "rewards/check_gptzero_func": 0.2678571343421936,
+      "rewards/check_perplexity_diff_func": 0.6428571343421936,
+      "rewards/check_winston_local_func": 0.5099858045578003,
+      "step": 81
+    },
+    {
+      "clip_ratio": 0.0030425102449953556,
+      "epoch": 0.11468531468531469,
+      "grad_norm": 0.10725025403922342,
+      "kl": 0.9921875,
+      "learning_rate": 2.8671328671328672e-05,
+      "loss": 0.0043,
+      "step": 82
+    },
+    {
+      "clip_ratio": 0.0030319676734507084,
+      "completion_length": 93.3214340209961,
+      "epoch": 0.11608391608391608,
+      "grad_norm": 0.22858957476463732,
+      "kl": 0.35546875,
+      "learning_rate": 2.9020979020979022e-05,
+      "loss": 0.0021,
+      "num_tokens": 819773.0,
+      "reward": 1.4201573133468628,
+      "reward_std": 0.761705756187439,
+      "rewards/check_gptzero_func": 0.3392857015132904,
+      "rewards/check_perplexity_diff_func": 0.5714285969734192,
+      "rewards/check_winston_local_func": 0.5094431042671204,
+      "step": 83
+    },
+    {
+      "clip_ratio": 0.0027519434224814177,
+      "epoch": 0.11748251748251748,
+      "grad_norm": 0.180287994009811,
+      "kl": 0.392578125,
+      "learning_rate": 2.9370629370629372e-05,
+      "loss": -0.0002,
+      "step": 84
+    },
+    {
+      "clip_ratio": 0.0034523813519626856,
+      "completion_length": 102.66072082519531,
+      "epoch": 0.11888111888111888,
+      "grad_norm": 0.25571068617285075,
+      "kl": 0.494140625,
+      "learning_rate": 2.972027972027972e-05,
+      "loss": -0.0103,
+      "num_tokens": 840280.0,
+      "reward": 1.7226576805114746,
+      "reward_std": 0.7545516490936279,
+      "rewards/check_gptzero_func": 0.5714285969734192,
+      "rewards/check_perplexity_diff_func": 0.6785714030265808,
+      "rewards/check_winston_local_func": 0.4726576805114746,
+      "step": 85
+    },
+    {
+      "clip_ratio": 0.00493992306292057,
+      "epoch": 0.12027972027972028,
+      "grad_norm": 0.2172932526081326,
+      "kl": 0.490234375,
+      "learning_rate": 3.0069930069930068e-05,
+      "loss": -0.013,
+      "step": 86
+    },
+    {
+      "clip_ratio": 0.003154533449560404,
+      "completion_length": 98.96428680419922,
+      "epoch": 0.12167832167832168,
+      "grad_norm": 0.3651883135439141,
+      "kl": 0.5859375,
+      "learning_rate": 3.0419580419580425e-05,
+      "loss": -0.0129,
+      "num_tokens": 860422.0,
+      "reward": 1.6449967622756958,
+      "reward_std": 0.7163441777229309,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 0.7142857313156128,
+      "rewards/check_winston_local_func": 0.5378537178039551,
+      "step": 87
+    },
+    {
+      "clip_ratio": 0.007247431669384241,
+      "epoch": 0.12307692307692308,
+      "grad_norm": 0.2167770735292921,
+      "kl": 0.61328125,
+      "learning_rate": 3.0769230769230774e-05,
+      "loss": -0.0168,
+      "step": 88
+    },
+    {
+      "clip_ratio": 0.0027036736719310284,
+      "completion_length": 108.03572082519531,
+      "epoch": 0.12447552447552447,
+      "grad_norm": 0.12604084880729077,
+      "kl": 0.291015625,
+      "learning_rate": 3.111888111888112e-05,
+      "loss": -0.0039,
+      "num_tokens": 881364.0,
+      "reward": 1.3799673318862915,
+      "reward_std": 0.6881024837493896,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 0.5714285969734192,
+      "rewards/check_winston_local_func": 0.4156815707683563,
+      "step": 89
+    },
+    {
+      "clip_ratio": 0.005238416139036417,
+      "epoch": 0.1258741258741259,
+      "grad_norm": 0.12103303780018655,
+      "kl": 0.30859375,
+      "learning_rate": 3.146853146853147e-05,
+      "loss": -0.005,
+      "step": 90
+    },
+    {
+      "clip_ratio": 0.003768512513488531,
+      "completion_length": 122.80357360839844,
+      "epoch": 0.12727272727272726,
+      "grad_norm": 0.18423815092289067,
+      "kl": 0.67578125,
+      "learning_rate": 3.181818181818182e-05,
+      "loss": -0.0171,
+      "num_tokens": 904487.0,
+      "reward": 1.6101170778274536,
+      "reward_std": 0.6241902709007263,
+      "rewards/check_gptzero_func": 0.3392857015132904,
+      "rewards/check_perplexity_diff_func": 0.7857142686843872,
+      "rewards/check_winston_local_func": 0.48511695861816406,
+      "step": 91
+    },
+    {
+      "clip_ratio": 0.005407070741057396,
+      "epoch": 0.12867132867132866,
+      "grad_norm": 0.15457101650134836,
+      "kl": 0.66796875,
+      "learning_rate": 3.216783216783217e-05,
+      "loss": -0.0182,
+      "step": 92
+    },
+    {
+      "clip_ratio": 0.002042532665655017,
+      "completion_length": 108.17857360839844,
+      "epoch": 0.13006993006993006,
+      "grad_norm": 0.19588538280583437,
+      "kl": 1.75,
+      "learning_rate": 3.251748251748252e-05,
+      "loss": -0.0023,
+      "num_tokens": 925633.0,
+      "reward": 1.3932582139968872,
+      "reward_std": 0.5087055563926697,
+      "rewards/check_gptzero_func": 0.4107142984867096,
+      "rewards/check_perplexity_diff_func": 0.4642857015132904,
+      "rewards/check_winston_local_func": 0.5182580351829529,
+      "step": 93
+    },
+    {
+      "clip_ratio": 0.00876330491155386,
+      "epoch": 0.13146853146853146,
+      "grad_norm": 0.1506256243597478,
+      "kl": 0.80078125,
+      "learning_rate": 3.2867132867132866e-05,
+      "loss": -0.0035,
+      "step": 94
+    },
+    {
+      "clip_ratio": 0.003118924330919981,
+      "completion_length": 77.4464340209961,
+      "epoch": 0.13286713286713286,
+      "grad_norm": 0.34784170699852607,
+      "kl": 0.72265625,
+      "learning_rate": 3.321678321678322e-05,
+      "loss": -0.0134,
+      "num_tokens": 942892.0,
+      "reward": 1.832355260848999,
+      "reward_std": 1.064102292060852,
+      "rewards/check_gptzero_func": 0.4821428656578064,
+      "rewards/check_perplexity_diff_func": 0.8214285969734192,
+      "rewards/check_winston_local_func": 0.5287837982177734,
+      "step": 95
+    },
+    {
+      "clip_ratio": 0.00509096821770072,
+      "epoch": 0.13426573426573427,
+      "grad_norm": 0.25968424971577175,
+      "kl": 0.96875,
+      "learning_rate": 3.356643356643357e-05,
+      "loss": -0.0176,
+      "step": 96
+    },
+    {
+      "clip_ratio": 0.0037272910121828318,
+      "completion_length": 95.26786041259766,
+      "epoch": 0.13566433566433567,
+      "grad_norm": 0.16914051199341906,
+      "kl": 0.349609375,
+      "learning_rate": 3.391608391608392e-05,
+      "loss": 0.0037,
+      "num_tokens": 962317.0,
+      "reward": 1.805726170539856,
+      "reward_std": 0.831261396408081,
+      "rewards/check_gptzero_func": 0.4107142984867096,
+      "rewards/check_perplexity_diff_func": 0.9285714030265808,
+      "rewards/check_winston_local_func": 0.46644020080566406,
+      "step": 97
+    },
+    {
+      "clip_ratio": 0.0034601751249283552,
+      "epoch": 0.13706293706293707,
+      "grad_norm": 0.16538870434198547,
+      "kl": 0.388671875,
+      "learning_rate": 3.4265734265734265e-05,
+      "loss": 0.0019,
+      "step": 98
+    },
+    {
+      "clip_ratio": 0.004487877711653709,
+      "completion_length": 111.8214340209961,
+      "epoch": 0.13846153846153847,
+      "grad_norm": 0.12415078751808917,
+      "kl": 0.384765625,
+      "learning_rate": 3.461538461538462e-05,
+      "loss": -0.001,
+      "num_tokens": 983513.0,
+      "reward": 1.2062333822250366,
+      "reward_std": 0.591974139213562,
+      "rewards/check_gptzero_func": 0.3214285671710968,
+      "rewards/check_perplexity_diff_func": 0.3571428656578064,
+      "rewards/check_winston_local_func": 0.5276618599891663,
+      "step": 99
+    },
+    {
+      "clip_ratio": 0.00284260674379766,
+      "epoch": 0.13986013986013987,
+      "grad_norm": 0.1197068572237121,
+      "kl": 0.376953125,
+      "learning_rate": 3.4965034965034965e-05,
+      "loss": -0.0017,
+      "step": 100
+    },
+    {
+      "clip_ratio": 0.0023206709884107113,
+      "completion_length": 101.08928680419922,
+      "epoch": 0.14125874125874127,
+      "grad_norm": 0.16400691003183243,
+      "kl": 0.30078125,
+      "learning_rate": 3.531468531468531e-05,
+      "loss": 0.0033,
+      "num_tokens": 1003878.0,
+      "reward": 1.631854772567749,
+      "reward_std": 0.5223387479782104,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 0.7142857313156128,
+      "rewards/check_winston_local_func": 0.5247119665145874,
+      "step": 101
+    },
+    {
+      "clip_ratio": 0.0016618981026113033,
+      "epoch": 0.14265734265734265,
+      "grad_norm": 0.1573929247787687,
+      "kl": 0.326171875,
+      "learning_rate": 3.566433566433567e-05,
+      "loss": 0.0012,
+      "step": 102
+    },
+    {
+      "clip_ratio": 0.0022201864048838615,
+      "completion_length": 124.46429443359375,
+      "epoch": 0.14405594405594405,
+      "grad_norm": 0.1353861256020412,
+      "kl": 0.3828125,
+      "learning_rate": 3.601398601398602e-05,
+      "loss": 0.005,
+      "num_tokens": 1027066.0,
+      "reward": 1.6920486688613892,
+      "reward_std": 0.48560601472854614,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 0.75,
+      "rewards/check_winston_local_func": 0.5491914749145508,
+      "step": 103
+    },
+    {
+      "clip_ratio": 0.0029075751081109047,
+      "epoch": 0.14545454545454545,
+      "grad_norm": 0.13328591024866146,
+      "kl": 0.39453125,
+      "learning_rate": 3.6363636363636364e-05,
+      "loss": 0.0034,
+      "step": 104
+    },
+    {
+      "clip_ratio": 0.0019614642951637506,
+      "completion_length": 85.26786041259766,
+      "epoch": 0.14685314685314685,
+      "grad_norm": 0.19674478002774554,
+      "kl": 0.333984375,
+      "learning_rate": 3.671328671328672e-05,
+      "loss": -0.0188,
+      "num_tokens": 1045567.0,
+      "reward": 1.6822034120559692,
+      "reward_std": 0.7559517621994019,
+      "rewards/check_gptzero_func": 0.5357142686843872,
+      "rewards/check_perplexity_diff_func": 0.6071428656578064,
+      "rewards/check_winston_local_func": 0.5393460988998413,
+      "step": 105
+    },
+    {
+      "clip_ratio": 0.004534203093498945,
+      "epoch": 0.14825174825174825,
+      "grad_norm": 0.16787611299305724,
+      "kl": 0.3671875,
+      "learning_rate": 3.7062937062937064e-05,
+      "loss": -0.0217,
+      "step": 106
+    },
+    {
+      "clip_ratio": 0.003009920008480549,
+      "completion_length": 106.55357360839844,
+      "epoch": 0.14965034965034965,
+      "grad_norm": 0.19751132540015684,
+      "kl": 0.455078125,
+      "learning_rate": 3.741258741258741e-05,
+      "loss": 0.0111,
+      "num_tokens": 1066466.0,
+      "reward": 2.178619861602783,
+      "reward_std": 0.7016831040382385,
+      "rewards/check_gptzero_func": 0.6071428656578064,
+      "rewards/check_perplexity_diff_func": 0.8928571343421936,
+      "rewards/check_winston_local_func": 0.6786197423934937,
+      "step": 107
+    },
+    {
+      "clip_ratio": 0.004473494831472635,
+      "epoch": 0.15104895104895105,
+      "grad_norm": 0.1431470008572649,
+      "kl": 0.4375,
+      "learning_rate": 3.776223776223776e-05,
+      "loss": 0.0095,
+      "step": 108
+    },
+    {
+      "clip_ratio": 0.0026541000697761774,
+      "completion_length": 100.6964340209961,
+      "epoch": 0.15244755244755245,
+      "grad_norm": 0.33135604940901486,
+      "kl": 0.310546875,
+      "learning_rate": 3.811188811188811e-05,
+      "loss": 0.0138,
+      "num_tokens": 1086751.0,
+      "reward": 1.6315226554870605,
+      "reward_std": 0.7806248068809509,
+      "rewards/check_gptzero_func": 0.4642857015132904,
+      "rewards/check_perplexity_diff_func": 0.6071428656578064,
+      "rewards/check_winston_local_func": 0.5600939393043518,
+      "step": 109
+    },
+    {
+      "clip_ratio": 0.0028892713598906994,
+      "epoch": 0.15384615384615385,
+      "grad_norm": 0.24217207616877234,
+      "kl": 0.361328125,
+      "learning_rate": 3.846153846153846e-05,
+      "loss": 0.0074,
+      "step": 110
+    },
+    {
+      "clip_ratio": 0.0021747422870248556,
+      "completion_length": 84.76786041259766,
+      "epoch": 0.15524475524475526,
+      "grad_norm": 0.15803622964815373,
+      "kl": 0.66796875,
+      "learning_rate": 3.8811188811188816e-05,
+      "loss": -0.0045,
+      "num_tokens": 1104518.0,
+      "reward": 1.6692500114440918,
+      "reward_std": 0.5965060591697693,
+      "rewards/check_gptzero_func": 0.5535714030265808,
+      "rewards/check_perplexity_diff_func": 0.6071428656578064,
+      "rewards/check_winston_local_func": 0.5085356831550598,
+      "step": 111
+    },
+    {
+      "clip_ratio": 0.004626331850886345,
+      "epoch": 0.15664335664335666,
+      "grad_norm": 0.14920406573741435,
+      "kl": 0.7578125,
+      "learning_rate": 3.916083916083916e-05,
+      "loss": -0.0064,
+      "step": 112
+    },
+    {
+      "clip_ratio": 0.002694516209885478,
+      "completion_length": 92.17857360839844,
+      "epoch": 0.15804195804195803,
+      "grad_norm": 0.161427063551978,
+      "kl": 0.361328125,
+      "learning_rate": 3.9510489510489516e-05,
+      "loss": 0.0058,
+      "num_tokens": 1123466.0,
+      "reward": 1.353637456893921,
+      "reward_std": 0.5352396965026855,
+      "rewards/check_gptzero_func": 0.4821428656578064,
+      "rewards/check_perplexity_diff_func": 0.3571428656578064,
+      "rewards/check_winston_local_func": 0.5143517255783081,
+      "step": 113
+    },
+    {
+      "clip_ratio": 0.00391918933019042,
+      "epoch": 0.15944055944055943,
+      "grad_norm": 0.14561571783883442,
+      "kl": 0.39453125,
+      "learning_rate": 3.986013986013986e-05,
+      "loss": 0.0036,
+      "step": 114
+    },
+    {
+      "clip_ratio": 0.001278667594306171,
+      "completion_length": 107.12500762939453,
+      "epoch": 0.16083916083916083,
+      "grad_norm": 0.1805624176716803,
+      "kl": 0.3046875,
+      "learning_rate": 4.020979020979021e-05,
+      "loss": -0.001,
+      "num_tokens": 1144619.0,
+      "reward": 1.6282455921173096,
+      "reward_std": 0.7325619459152222,
+      "rewards/check_gptzero_func": 0.4464285671710968,
+      "rewards/check_perplexity_diff_func": 0.5357142686843872,
+      "rewards/check_winston_local_func": 0.6461027264595032,
+      "step": 115
+    },
+    {
+      "clip_ratio": 0.0017050534952431917,
+      "epoch": 0.16223776223776223,
+      "grad_norm": 0.16942268791902212,
+      "kl": 0.337890625,
+      "learning_rate": 4.055944055944056e-05,
+      "loss": -0.0027,
+      "step": 116
+    },
+    {
+      "clip_ratio": 0.0015462420415133238,
+      "completion_length": 114.8214340209961,
+      "epoch": 0.16363636363636364,
+      "grad_norm": 0.1777133538888581,
+      "kl": 0.322265625,
+      "learning_rate": 4.0909090909090915e-05,
+      "loss": -0.0028,
+      "num_tokens": 1166185.0,
+      "reward": 1.566156268119812,
+      "reward_std": 0.38663557171821594,
+      "rewards/check_gptzero_func": 0.4642857015132904,
+      "rewards/check_perplexity_diff_func": 0.5,
+      "rewards/check_winston_local_func": 0.6018705368041992,
+      "step": 117
+    },
+    {
+      "clip_ratio": 0.0030793240293860435,
+      "epoch": 0.16503496503496504,
+      "grad_norm": 0.15084856022622706,
+      "kl": 0.2890625,
+      "learning_rate": 4.125874125874126e-05,
+      "loss": -0.0048,
+      "step": 118
+    },
+    {
+      "clip_ratio": 0.0018360918620601296,
+      "completion_length": 100.25000762939453,
+      "epoch": 0.16643356643356644,
+      "grad_norm": 0.10943003413749244,
+      "kl": 0.83984375,
+      "learning_rate": 4.1608391608391614e-05,
+      "loss": 0.0043,
+      "num_tokens": 1186317.0,
+      "reward": 1.7977957725524902,
+      "reward_std": 0.5172301530838013,
+      "rewards/check_gptzero_func": 0.5714285969734192,
+      "rewards/check_perplexity_diff_func": 0.7142857313156128,
+      "rewards/check_winston_local_func": 0.5120813250541687,
+      "step": 119
+    },
+    {
+      "clip_ratio": 0.0025589358992874622,
+      "epoch": 0.16783216783216784,
+      "grad_norm": 0.10431032316313237,
+      "kl": 0.68359375,
+      "learning_rate": 4.195804195804196e-05,
+      "loss": 0.0029,
+      "step": 120
+    },
+    {
+      "clip_ratio": 0.002800893737003207,
+      "completion_length": 120.91072082519531,
+      "epoch": 0.16923076923076924,
+      "grad_norm": 0.1567960287886374,
+      "kl": 0.37890625,
+      "learning_rate": 4.230769230769231e-05,
+      "loss": -0.0016,
+      "num_tokens": 1208530.0,
+      "reward": 1.5369055271148682,
+      "reward_std": 0.7073518633842468,
+      "rewards/check_gptzero_func": 0.3214285671710968,
+      "rewards/check_perplexity_diff_func": 0.6785714030265808,
+      "rewards/check_winston_local_func": 0.5369054079055786,
+      "step": 121
+    },
+    {
+      "clip_ratio": 0.0026409339625388384,
+      "epoch": 0.17062937062937064,
+      "grad_norm": 0.15013150065673506,
+      "kl": 0.37890625,
+      "learning_rate": 4.265734265734266e-05,
+      "loss": -0.0035,
+      "step": 122
+    },
+    {
+      "clip_ratio": 0.002211854327470064,
+      "completion_length": 110.66072082519531,
+      "epoch": 0.17202797202797201,
+      "grad_norm": 0.17274409069851862,
+      "kl": 0.380859375,
+      "learning_rate": 4.300699300699301e-05,
+      "loss": -0.0062,
+      "num_tokens": 1229171.0,
+      "reward": 1.6140996217727661,
+      "reward_std": 0.7221139669418335,
+      "rewards/check_gptzero_func": 0.4642857015132904,
+      "rewards/check_perplexity_diff_func": 0.5357142686843872,
+      "rewards/check_winston_local_func": 0.6140995025634766,
+      "step": 123
+    },
+    {
+      "clip_ratio": 0.005159804597496986,
+      "epoch": 0.17342657342657342,
+      "grad_norm": 0.13306053466541726,
+      "kl": 0.443359375,
+      "learning_rate": 4.335664335664335e-05,
+      "loss": -0.009,
+      "step": 124
+    },
+    {
+      "clip_ratio": 0.0031610180158168077,
+      "completion_length": 104.30357360839844,
+      "epoch": 0.17482517482517482,
+      "grad_norm": 0.19261275892706695,
+      "kl": 0.30078125,
+      "learning_rate": 4.370629370629371e-05,
+      "loss": -0.0129,
+      "num_tokens": 1249350.0,
+      "reward": 1.9047484397888184,
+      "reward_std": 0.6767317652702332,
+      "rewards/check_gptzero_func": 0.6071428656578064,
+      "rewards/check_perplexity_diff_func": 0.6071428656578064,
+      "rewards/check_winston_local_func": 0.6904626488685608,
+      "step": 125
+    },
+    {
+      "clip_ratio": 0.004231306724250317,
+      "epoch": 0.17622377622377622,
+      "grad_norm": 0.1752737652506695,
+      "kl": 0.33203125,
+      "learning_rate": 4.405594405594406e-05,
+      "loss": -0.0162,
+      "step": 126
+    },
+    {
+      "clip_ratio": 0.003981932066380978,
+      "completion_length": 109.53572082519531,
+      "epoch": 0.17762237762237762,
+      "grad_norm": 0.1423250496935692,
+      "kl": 0.48828125,
+      "learning_rate": 4.4405594405594406e-05,
+      "loss": -0.0102,
+      "num_tokens": 1269848.0,
+      "reward": 1.7591207027435303,
+      "reward_std": 0.5321380496025085,
+      "rewards/check_gptzero_func": 0.5,
+      "rewards/check_perplexity_diff_func": 0.6785714030265808,
+      "rewards/check_winston_local_func": 0.5805493593215942,
+      "step": 127
+    },
+    {
+      "clip_ratio": 0.005287667270749807,
+      "epoch": 0.17902097902097902,
+      "grad_norm": 0.13255848380623775,
+      "kl": 0.498046875,
+      "learning_rate": 4.475524475524476e-05,
+      "loss": -0.0121,
+      "step": 128
+    },
+    {
+      "clip_ratio": 0.0025025398936122656,
+      "completion_length": 105.30357360839844,
+      "epoch": 0.18041958041958042,
+      "grad_norm": 0.1177341259986552,
+      "kl": 0.283203125,
+      "learning_rate": 4.5104895104895105e-05,
+      "loss": -0.0016,
+      "num_tokens": 1290033.0,
+      "reward": 1.7174798250198364,
+      "reward_std": 0.5096268057823181,
+      "rewards/check_gptzero_func": 0.5357142686843872,
+      "rewards/check_perplexity_diff_func": 0.5357142686843872,
+      "rewards/check_winston_local_func": 0.6460510492324829,
+      "step": 129
+    },
+    {
+      "clip_ratio": 0.0033291254658252,
+      "epoch": 0.18181818181818182,
+      "grad_norm": 0.11141937591016414,
+      "kl": 0.294921875,
+      "learning_rate": 4.545454545454546e-05,
+      "loss": -0.0032,
+      "step": 130
+    },
+    {
+      "clip_ratio": 0.0024425899609923363,
+      "completion_length": 96.30357360839844,
+      "epoch": 0.18321678321678322,
+      "grad_norm": 0.23593816935965856,
+      "kl": 0.341796875,
+      "learning_rate": 4.5804195804195805e-05,
+      "loss": -0.006,
+      "num_tokens": 1309356.0,
+      "reward": 1.7568891048431396,
+      "reward_std": 0.7255779504776001,
+      "rewards/check_gptzero_func": 0.5892857313156128,
+      "rewards/check_perplexity_diff_func": 0.5357142686843872,
+      "rewards/check_winston_local_func": 0.6318890452384949,
+      "step": 131
+    },
+    {
+      "clip_ratio": 0.004331678152084351,
+      "epoch": 0.18461538461538463,
+      "grad_norm": 0.2135627059179365,
+      "kl": 0.349609375,
+      "learning_rate": 4.615384615384616e-05,
+      "loss": -0.0089,
+      "step": 132
+    },
+    {
+      "clip_ratio": 0.0028827630449086428,
+      "completion_length": 111.6964340209961,
+      "epoch": 0.18601398601398603,
+      "grad_norm": 0.1622976058137894,
+      "kl": 0.369140625,
+      "learning_rate": 4.6503496503496505e-05,
+      "loss": 0.014,
+      "num_tokens": 1330017.0,
+      "reward": 1.6692737340927124,
+      "reward_std": 0.628279983997345,
+      "rewards/check_gptzero_func": 0.4464285671710968,
+      "rewards/check_perplexity_diff_func": 0.6428571343421936,
+      "rewards/check_winston_local_func": 0.5799878835678101,
+      "step": 133
+    },
+    {
+      "clip_ratio": 0.004608546383678913,
+      "epoch": 0.1874125874125874,
+      "grad_norm": 0.16503328062657166,
+      "kl": 0.4140625,
+      "learning_rate": 4.685314685314686e-05,
+      "loss": 0.0121,
+      "step": 134
+    },
+    {
+      "clip_ratio": 0.0027426625601947308,
+      "completion_length": 107.58928680419922,
+      "epoch": 0.1888111888111888,
+      "grad_norm": 0.27926954484170907,
+      "kl": 0.5859375,
+      "learning_rate": 4.7202797202797204e-05,
+      "loss": 0.0139,
+      "num_tokens": 1350164.0,
+      "reward": 1.9428951740264893,
+      "reward_std": 0.6565932035446167,
+      "rewards/check_gptzero_func": 0.6785714030265808,
+      "rewards/check_perplexity_diff_func": 0.5714285969734192,
+      "rewards/check_winston_local_func": 0.6928950548171997,
+      "step": 135
+    },
+    {
+      "clip_ratio": 0.00439854059368372,
+      "epoch": 0.1902097902097902,
+      "grad_norm": 0.1925514269864402,
+      "kl": 0.5546875,
+      "learning_rate": 4.755244755244756e-05,
+      "loss": 0.0074,
+      "step": 136
+    },
+    {
+      "clip_ratio": 0.0018738384824246168,
+      "completion_length": 116.4464340209961,
+      "epoch": 0.1916083916083916,
+      "grad_norm": 0.14918417137816656,
+      "kl": 0.58984375,
+      "learning_rate": 4.7902097902097904e-05,
+      "loss": 0.0037,
+      "num_tokens": 1371507.0,
+      "reward": 2.0137577056884766,
+      "reward_std": 0.5378711819648743,
+      "rewards/check_gptzero_func": 0.6071428656578064,
+      "rewards/check_perplexity_diff_func": 0.6785714030265808,
+      "rewards/check_winston_local_func": 0.7280434370040894,
+      "step": 137
+    },
+    {
+      "clip_ratio": 0.003529219189658761,
+      "epoch": 0.193006993006993,
+      "grad_norm": 0.1301060608041501,
+      "kl": 0.6171875,
+      "learning_rate": 4.825174825174825e-05,
+      "loss": 0.0016,
+      "step": 138
+    },
+    {
+      "clip_ratio": 0.0011985624441877007,
+      "completion_length": 112.6964340209961,
+      "epoch": 0.1944055944055944,
+      "grad_norm": 0.13034934146859325,
+      "kl": 0.322265625,
+      "learning_rate": 4.86013986013986e-05,
+      "loss": -0.0024,
+      "num_tokens": 1392414.0,
+      "reward": 1.9355616569519043,
+      "reward_std": 0.4726971983909607,
+      "rewards/check_gptzero_func": 0.5892857313156128,
+      "rewards/check_perplexity_diff_func": 0.75,
+      "rewards/check_winston_local_func": 0.596275806427002,
+      "step": 139
+    },
+    {
+      "clip_ratio": 0.0032699662260711193,
+      "epoch": 0.1958041958041958,
+      "grad_norm": 0.12091032490165694,
+      "kl": 0.318359375,
+      "learning_rate": 4.8951048951048956e-05,
+      "loss": -0.0039,
+      "step": 140
+    },
+    {
+      "clip_ratio": 0.0013384540798142552,
+      "completion_length": 132.05357360839844,
+      "epoch": 0.1972027972027972,
+      "grad_norm": 0.12728573210952063,
+      "kl": 0.59765625,
+      "learning_rate": 4.93006993006993e-05,
+      "loss": 0.0045,
+      "num_tokens": 1416177.0,
+      "reward": 1.7924094200134277,
+      "reward_std": 0.7202263474464417,
+      "rewards/check_gptzero_func": 0.5714285969734192,
+      "rewards/check_perplexity_diff_func": 0.7142857313156128,
+      "rewards/check_winston_local_func": 0.5066950917243958,
+      "step": 141
+    },
+    {
+      "clip_ratio": 0.0017050639726221561,
+      "epoch": 0.1986013986013986,
+      "grad_norm": 0.12231361158637834,
+      "kl": 0.65234375,
+      "learning_rate": 4.9650349650349656e-05,
+      "loss": 0.0025,
+      "step": 142
+    },
+    {
+      "clip_ratio": 0.0015860958956182003,
+      "completion_length": 99.73214721679688,
+      "epoch": 0.2,
+      "grad_norm": 0.1853946177277763,
+      "kl": 0.58984375,
+      "learning_rate": 5e-05,
+      "loss": 0.0021,
+      "num_tokens": 1435732.0,
+      "reward": 2.0557591915130615,
+      "reward_std": 0.503829836845398,
+      "rewards/check_gptzero_func": 0.6964285969734192,
+      "rewards/check_perplexity_diff_func": 0.75,
+      "rewards/check_winston_local_func": 0.6093305945396423,
+      "step": 143
+    },
+    {
+      "clip_ratio": 0.002123563550412655,
+      "epoch": 0.2013986013986014,
+      "grad_norm": 0.13470126319919157,
+      "kl": 0.423828125,
+      "learning_rate": 5.0349650349650356e-05,
+      "loss": -0.0009,
+      "step": 144
+    },
+    {
+      "clip_ratio": 0.0018757216166704893,
+      "completion_length": 95.73214721679688,
+      "epoch": 0.20279720279720279,
+      "grad_norm": 0.14930190515465938,
+      "kl": 0.388671875,
+      "learning_rate": 5.06993006993007e-05,
+      "loss": -0.0026,
+      "num_tokens": 1454705.0,
+      "reward": 2.0560998916625977,
+      "reward_std": 0.5541732311248779,
+      "rewards/check_gptzero_func": 0.6964285969734192,
+      "rewards/check_perplexity_diff_func": 0.5714285969734192,
+      "rewards/check_winston_local_func": 0.788242518901825,
+      "step": 145
+    },
+    {
+      "clip_ratio": 0.003540371311828494,
+      "epoch": 0.2041958041958042,
+      "grad_norm": 0.1304251509197979,
+      "kl": 0.41796875,
+      "learning_rate": 5.1048951048951055e-05,
+      "loss": -0.0046,
+      "step": 146
+    },
+    {
+      "clip_ratio": 0.0031265008728951216,
+      "completion_length": 80.9464340209961,
+      "epoch": 0.2055944055944056,
+      "grad_norm": 0.2087929657840581,
+      "kl": 1.1328125,
+      "learning_rate": 5.1398601398601395e-05,
+      "loss": 0.0026,
+      "num_tokens": 1472182.0,
+      "reward": 1.9076076745986938,
+      "reward_std": 0.46633273363113403,
+      "rewards/check_gptzero_func": 0.6428571343421936,
+      "rewards/check_perplexity_diff_func": 0.5,
+      "rewards/check_winston_local_func": 0.7647504210472107,
+      "step": 147
+    },
+    {
+      "clip_ratio": 0.0029583375435322523,
+      "epoch": 0.206993006993007,
+      "grad_norm": 0.178151823438605,
+      "kl": 0.66015625,
+      "learning_rate": 5.1748251748251755e-05,
+      "loss": -0.0014,
+      "step": 148
+    },
+    {
+      "clip_ratio": 0.0020277961157262325,
+      "completion_length": 126.58929443359375,
+      "epoch": 0.2083916083916084,
+      "grad_norm": 0.13736153026044381,
+      "kl": 0.373046875,
+      "learning_rate": 5.2097902097902094e-05,
+      "loss": 0.0127,
+      "num_tokens": 1494581.0,
+      "reward": 1.870171308517456,
+      "reward_std": 0.4412032663822174,
+      "rewards/check_gptzero_func": 0.4464285671710968,
+      "rewards/check_perplexity_diff_func": 0.8214285969734192,
+      "rewards/check_winston_local_func": 0.6023141741752625,
+      "step": 149
+    },
+    {
+      "clip_ratio": 0.0022643795236945152,
+      "epoch": 0.2097902097902098,
+      "grad_norm": 0.12942677801199462,
+      "kl": 0.40625,
+      "learning_rate": 5.244755244755245e-05,
+      "loss": 0.0106,
+      "step": 150
+    },
+    {
+      "clip_ratio": 0.001886948710307479,
+      "completion_length": 95.14286041259766,
+      "epoch": 0.2111888111888112,
+      "grad_norm": 0.15555312890089615,
+      "kl": 0.62890625,
+      "learning_rate": 5.279720279720281e-05,
+      "loss": 0.0013,
+      "num_tokens": 1513403.0,
+      "reward": 1.9735006093978882,
+      "reward_std": 0.5218394994735718,
+      "rewards/check_gptzero_func": 0.5535714030265808,
+      "rewards/check_perplexity_diff_func": 0.6071428656578064,
+      "rewards/check_winston_local_func": 0.8127861022949219,
+      "step": 151
+    },
+    {
+      "clip_ratio": 0.003476199461147189,
+      "epoch": 0.2125874125874126,
+      "grad_norm": 0.13833602394582134,
+      "kl": 0.5390625,
+      "learning_rate": 5.314685314685315e-05,
+      "loss": -0.0015,
+      "step": 152
+    },
+    {
+      "clip_ratio": 0.0029247915372252464,
+      "completion_length": 104.1964340209961,
+      "epoch": 0.213986013986014,
+      "grad_norm": 0.18332277821774023,
+      "kl": 0.451171875,
+      "learning_rate": 5.34965034965035e-05,
+      "loss": -0.0047,
+      "num_tokens": 1533154.0,
+      "reward": 1.977178931236267,
+      "reward_std": 0.5758928656578064,
+      "rewards/check_gptzero_func": 0.6785714030265808,
+      "rewards/check_perplexity_diff_func": 0.6785714030265808,
+      "rewards/check_winston_local_func": 0.6200361251831055,
+      "step": 153
+    },
+    {
+      "clip_ratio": 0.002894646255299449,
+      "epoch": 0.2153846153846154,
+      "grad_norm": 0.12417162413759981,
+      "kl": 0.48828125,
+      "learning_rate": 5.384615384615385e-05,
+      "loss": -0.0082,
+      "step": 154
+    },
+    {
+      "clip_ratio": 0.002533347113057971,
+      "completion_length": 106.6964340209961,
+      "epoch": 0.21678321678321677,
+      "grad_norm": 0.2165608004438418,
+      "kl": 0.4140625,
+      "learning_rate": 5.41958041958042e-05,
+      "loss": 0.0367,
+      "num_tokens": 1554005.0,
+      "reward": 2.254620313644409,
+      "reward_std": 0.6860859394073486,
+      "rewards/check_gptzero_func": 0.7857142686843872,
+      "rewards/check_perplexity_diff_func": 0.7142857313156128,
+      "rewards/check_winston_local_func": 0.7546200752258301,
+      "step": 155
+    },
+    {
+      "clip_ratio": 0.005156368017196655,
+      "epoch": 0.21818181818181817,
+      "grad_norm": 0.16185524517286934,
+      "kl": 0.4609375,
+      "learning_rate": 5.4545454545454546e-05,
+      "loss": 0.0322,
+      "step": 156
+    },
+    {
+      "clip_ratio": 0.002192203886806965,
+      "completion_length": 129.55357360839844,
+      "epoch": 0.21958041958041957,
+      "grad_norm": 0.13686573639431518,
+      "kl": 0.421875,
+      "learning_rate": 5.48951048951049e-05,
+      "loss": 0.0095,
+      "num_tokens": 1576988.0,
+      "reward": 2.202296495437622,
+      "reward_std": 0.4840867817401886,
+      "rewards/check_gptzero_func": 0.6428571343421936,
+      "rewards/check_perplexity_diff_func": 0.8928571343421936,
+      "rewards/check_winston_local_func": 0.6665821075439453,
+      "step": 157
+    },
+    {
+      "clip_ratio": 0.0037054666317999363,
+      "epoch": 0.22097902097902097,
+      "grad_norm": 0.12040805378810834,
+      "kl": 0.45703125,
+      "learning_rate": 5.524475524475524e-05,
+      "loss": 0.0077,
+      "step": 158
+    },
+    {
+      "clip_ratio": 0.002324200002476573,
+      "completion_length": 141.10714721679688,
+      "epoch": 0.22237762237762237,
+      "grad_norm": 0.1722210877266639,
+      "kl": 0.9765625,
+      "learning_rate": 5.55944055944056e-05,
+      "loss": -0.0049,
+      "num_tokens": 1600826.0,
+      "reward": 2.0245378017425537,
+      "reward_std": 0.45877185463905334,
+      "rewards/check_gptzero_func": 0.5892857313156128,
+      "rewards/check_perplexity_diff_func": 0.75,
+      "rewards/check_winston_local_func": 0.6852518916130066,
+      "step": 159
+    },
+    {
+      "clip_ratio": 0.003470373572781682,
+      "epoch": 0.22377622377622378,
+      "grad_norm": 0.13356363193434526,
+      "kl": 0.65234375,
+      "learning_rate": 5.594405594405595e-05,
+      "loss": -0.0085,
+      "step": 160
+    },
+    {
+      "clip_ratio": 0.0026535126380622387,
+      "completion_length": 109.76786041259766,
+      "epoch": 0.22517482517482518,
+      "grad_norm": 0.409404915148595,
+      "kl": 0.515625,
+      "learning_rate": 5.629370629370629e-05,
+      "loss": -0.0051,
+      "num_tokens": 1621103.0,
+      "reward": 2.0040969848632812,
+      "reward_std": 0.5942196249961853,
+      "rewards/check_gptzero_func": 0.5178571343421936,
+      "rewards/check_perplexity_diff_func": 0.7857142686843872,
+      "rewards/check_winston_local_func": 0.7005256414413452,
+      "step": 161
+    },
+    {
+      "clip_ratio": 0.0058495416305959225,
+      "epoch": 0.22657342657342658,
+      "grad_norm": 0.16376328147020022,
+      "kl": 0.5078125,
+      "learning_rate": 5.664335664335665e-05,
+      "loss": -0.0129,
+      "step": 162
+    },
+    {
+      "clip_ratio": 0.002924378262832761,
+      "completion_length": 114.3214340209961,
+      "epoch": 0.22797202797202798,
+      "grad_norm": 0.17698045766822795,
+      "kl": 0.62109375,
+      "learning_rate": 5.699300699300699e-05,
+      "loss": -0.0053,
+      "num_tokens": 1642629.0,
+      "reward": 2.0913121700286865,
+      "reward_std": 0.533393144607544,
+      "rewards/check_gptzero_func": 0.6785714030265808,
+      "rewards/check_perplexity_diff_func": 0.6785714030265808,
+      "rewards/check_winston_local_func": 0.7341693639755249,
+      "step": 163
+    },
+    {
+      "clip_ratio": 0.006533453240990639,
+      "epoch": 0.22937062937062938,
+      "grad_norm": 0.15566114152311913,
+      "kl": 0.66796875,
+      "learning_rate": 5.7342657342657345e-05,
+      "loss": -0.0085,
+      "step": 164
+    },
+    {
+      "clip_ratio": 0.002033352619037032,
+      "completion_length": 110.48214721679688,
+      "epoch": 0.23076923076923078,
+      "grad_norm": 0.19340316318251602,
+      "kl": 0.4453125,
+      "learning_rate": 5.769230769230769e-05,
+      "loss": 0.0054,
+      "num_tokens": 1663000.0,
+      "reward": 1.809409499168396,
+      "reward_std": 0.5209853053092957,
+      "rewards/check_gptzero_func": 0.6428571343421936,
+      "rewards/check_perplexity_diff_func": 0.3571428656578064,
+      "rewards/check_winston_local_func": 0.809409499168396,
+      "step": 165
+    },
+    {
+      "clip_ratio": 0.005329853855073452,
+      "epoch": 0.23216783216783216,
+      "grad_norm": 0.15299329983895626,
+      "kl": 0.55078125,
+      "learning_rate": 5.8041958041958044e-05,
+      "loss": 0.0018,
+      "step": 166
+    },
+    {
+      "clip_ratio": 0.0025733087677508593,
+      "completion_length": 95.73214721679688,
+      "epoch": 0.23356643356643356,
+      "grad_norm": 0.16102285972903455,
+      "kl": 0.458984375,
+      "learning_rate": 5.83916083916084e-05,
+      "loss": -0.0008,
+      "num_tokens": 1682123.0,
+      "reward": 1.9705681800842285,
+      "reward_std": 0.5465096235275269,
+      "rewards/check_gptzero_func": 0.6428571343421936,
+      "rewards/check_perplexity_diff_func": 0.6428571343421936,
+      "rewards/check_winston_local_func": 0.6848538517951965,
+      "step": 167
+    },
+    {
+      "clip_ratio": 0.0034487086813896894,
+      "epoch": 0.23496503496503496,
+      "grad_norm": 0.12896214559568192,
+      "kl": 0.474609375,
+      "learning_rate": 5.8741258741258744e-05,
+      "loss": -0.0039,
+      "step": 168
+    },
+    {
+      "clip_ratio": 0.0012721805833280087,
+      "completion_length": 72.125,
+      "epoch": 0.23636363636363636,
+      "grad_norm": 0.19976827264055144,
+      "kl": 0.5703125,
+      "learning_rate": 5.90909090909091e-05,
+      "loss": -0.0046,
+      "num_tokens": 1698078.0,
+      "reward": 1.8698246479034424,
+      "reward_std": 0.4430284798145294,
+      "rewards/check_gptzero_func": 0.7142857313156128,
+      "rewards/check_perplexity_diff_func": 0.3571428656578064,
+      "rewards/check_winston_local_func": 0.7983959913253784,
+      "step": 169
+    },
+    {
+      "clip_ratio": 0.005269515328109264,
+      "epoch": 0.23776223776223776,
+      "grad_norm": 0.14855282442633314,
+      "kl": 0.61328125,
+      "learning_rate": 5.944055944055944e-05,
+      "loss": -0.0079,
+      "step": 170
+    },
+    {
+      "clip_ratio": 0.0027731256559491158,
+      "completion_length": 117.83929443359375,
+      "epoch": 0.23916083916083916,
+      "grad_norm": 0.23058249563979205,
+      "kl": 0.412109375,
+      "learning_rate": 5.9790209790209796e-05,
+      "loss": 0.0055,
+      "num_tokens": 1719673.0,
+      "reward": 2.2663450241088867,
+      "reward_std": 0.46630793809890747,
+      "rewards/check_gptzero_func": 0.6785714030265808,
+      "rewards/check_perplexity_diff_func": 0.7142857313156128,
+      "rewards/check_winston_local_func": 0.8734878301620483,
+      "step": 171
+    },
+    {
+      "clip_ratio": 0.0034511894918978214,
+      "epoch": 0.24055944055944056,
+      "grad_norm": 0.17368207545699044,
+      "kl": 0.373046875,
+      "learning_rate": 6.0139860139860136e-05,
+      "loss": -0.0007,
+      "step": 172
+    },
+    {
+      "clip_ratio": 0.0032431341242045164,
+      "completion_length": 126.85714721679688,
+      "epoch": 0.24195804195804196,
+      "grad_norm": 0.17995690811350934,
+      "kl": 0.353515625,
+      "learning_rate": 6.048951048951049e-05,
+      "loss": -0.0011,
+      "num_tokens": 1742569.0,
+      "reward": 2.258990526199341,
+      "reward_std": 0.6008436679840088,
+      "rewards/check_gptzero_func": 0.75,
+      "rewards/check_perplexity_diff_func": 0.75,
+      "rewards/check_winston_local_func": 0.7589904069900513,
+      "step": 173
+    },
+    {
+      "clip_ratio": 0.006950656417757273,
+      "epoch": 0.24335664335664337,
+      "grad_norm": 0.13892848528506246,
+      "kl": 0.359375,
+      "learning_rate": 6.083916083916085e-05,
+      "loss": -0.0033,
+      "step": 174
+    },
+    {
+      "clip_ratio": 0.0027662119828164577,
+      "completion_length": 130.8928680419922,
+      "epoch": 0.24475524475524477,
+      "grad_norm": 0.10040261340554028,
+      "kl": 0.345703125,
+      "learning_rate": 6.118881118881119e-05,
+      "loss": -0.0032,
+      "num_tokens": 1765643.0,
+      "reward": 2.074061632156372,
+      "reward_std": 0.3745954632759094,
+      "rewards/check_gptzero_func": 0.6607142686843872,
+      "rewards/check_perplexity_diff_func": 0.6785714030265808,
+      "rewards/check_winston_local_func": 0.7347758412361145,
+      "step": 175
+    },
+    {
+      "clip_ratio": 0.0030742601957172155,
+      "epoch": 0.24615384615384617,
+      "grad_norm": 0.0921134439221548,
+      "kl": 0.3359375,
+      "learning_rate": 6.153846153846155e-05,
+      "loss": -0.0044,
+      "step": 176
+    },
+    {
+      "clip_ratio": 0.002027927665039897,
+      "completion_length": 89.73214721679688,
+      "epoch": 0.24755244755244754,
+      "grad_norm": 0.1709976566266231,
+      "kl": 0.50390625,
+      "learning_rate": 6.188811188811188e-05,
+      "loss": 0.0051,
+      "num_tokens": 1783674.0,
+      "reward": 1.7810758352279663,
+      "reward_std": 0.4943030774593353,
+      "rewards/check_gptzero_func": 0.6428571343421936,
+      "rewards/check_perplexity_diff_func": 0.4285714328289032,
+      "rewards/check_winston_local_func": 0.7096471786499023,
+      "step": 177
+    },
+    {
+      "clip_ratio": 0.0020612890366464853,
+      "epoch": 0.24895104895104894,
+      "grad_norm": 0.1477275348028994,
+      "kl": 0.50390625,
+      "learning_rate": 6.223776223776224e-05,
+      "loss": 0.0011,
+      "step": 178
+    },
+    {
+      "clip_ratio": 0.0021104796323925257,
+      "completion_length": 132.10714721679688,
+      "epoch": 0.25034965034965034,
+      "grad_norm": 0.08842848405697774,
+      "kl": 0.39453125,
+      "learning_rate": 6.258741258741259e-05,
+      "loss": 0.0005,
+      "num_tokens": 1806754.0,
+      "reward": 2.107767105102539,
+      "reward_std": 0.269815057516098,
+      "rewards/check_gptzero_func": 0.75,
+      "rewards/check_perplexity_diff_func": 0.5357142686843872,
+      "rewards/check_winston_local_func": 0.8220529556274414,
+      "step": 179
+    },
+    {
+      "clip_ratio": 0.004407648928463459,
+      "epoch": 0.2517482517482518,
+      "grad_norm": 0.07542595290297384,
+      "kl": 0.416015625,
+      "learning_rate": 6.293706293706293e-05,
+      "loss": -0.0006,
+      "step": 180
+    },
+    {
+      "clip_ratio": 0.0018615310546010733,
+      "completion_length": 89.30357360839844,
+      "epoch": 0.25314685314685315,
+      "grad_norm": 0.2895549842864223,
+      "kl": 0.498046875,
+      "learning_rate": 6.32867132867133e-05,
+      "loss": -0.0082,
+      "num_tokens": 1824795.0,
+      "reward": 1.8801069259643555,
+      "reward_std": 0.5523244738578796,
+      "rewards/check_gptzero_func": 0.6071428656578064,
+      "rewards/check_perplexity_diff_func": 0.5,
+      "rewards/check_winston_local_func": 0.7729640603065491,
+      "step": 181
+    },
+    {
+      "clip_ratio": 0.007555535528808832,
+      "epoch": 0.2545454545454545,
+      "grad_norm": 0.20874610098022578,
+      "kl": 0.59375,
+      "learning_rate": 6.363636363636364e-05,
+      "loss": -0.016,
+      "step": 182
+    },
+    {
+      "clip_ratio": 0.0016618422232568264,
+      "completion_length": 84.46428680419922,
+      "epoch": 0.25594405594405595,
+      "grad_norm": 0.20653753743630154,
+      "kl": 0.5625,
+      "learning_rate": 6.398601398601399e-05,
+      "loss": 0.0058,
+      "num_tokens": 1842573.0,
+      "reward": 2.1319751739501953,
+      "reward_std": 0.5030335187911987,
+      "rewards/check_gptzero_func": 0.7321428656578064,
+      "rewards/check_perplexity_diff_func": 0.5714285969734192,
+      "rewards/check_winston_local_func": 0.8284037709236145,
+      "step": 183
+    },
+    {
+      "clip_ratio": 0.00579726742580533,
+      "epoch": 0.2573426573426573,
+      "grad_norm": 0.15782001907156346,
+      "kl": 0.578125,
+      "learning_rate": 6.433566433566433e-05,
+      "loss": 0.0001,
+      "step": 184
+    },
+    {
+      "clip_ratio": 0.0016176491044461727,
+      "completion_length": 125.39286041259766,
+      "epoch": 0.25874125874125875,
+      "grad_norm": 0.12385736188741701,
+      "kl": 0.4921875,
+      "learning_rate": 6.46853146853147e-05,
+      "loss": -0.0027,
+      "num_tokens": 1864297.0,
+      "reward": 1.7732529640197754,
+      "reward_std": 0.3774341642856598,
+      "rewards/check_gptzero_func": 0.625,
+      "rewards/check_perplexity_diff_func": 0.4285714328289032,
+      "rewards/check_winston_local_func": 0.7196813821792603,
+      "step": 185
+    },
+    {
+      "clip_ratio": 0.00623862212523818,
+      "epoch": 0.2601398601398601,
+      "grad_norm": 0.10877531810926387,
+      "kl": 0.53125,
+      "learning_rate": 6.503496503496504e-05,
+      "loss": -0.0048,
+      "step": 186
+    },
+    {
+      "clip_ratio": 0.0031279984395951033,
+      "completion_length": 98.00000762939453,
+      "epoch": 0.26153846153846155,
+      "grad_norm": 0.3661429387824046,
+      "kl": 0.6328125,
+      "learning_rate": 6.538461538461539e-05,
+      "loss": -0.0063,
+      "num_tokens": 1882931.0,
+      "reward": 2.0694050788879395,
+      "reward_std": 0.6090536117553711,
+      "rewards/check_gptzero_func": 0.6785714030265808,
+      "rewards/check_perplexity_diff_func": 0.5714285969734192,
+      "rewards/check_winston_local_func": 0.819405198097229,
+      "step": 187
+    },
+    {
+      "clip_ratio": 0.010017934255301952,
+      "epoch": 0.2629370629370629,
+      "grad_norm": 0.34977841113727764,
+      "kl": 0.65234375,
+      "learning_rate": 6.573426573426573e-05,
+      "loss": -0.0185,
+      "step": 188
+    },
+    {
+      "clip_ratio": 0.00222155568189919,
+      "completion_length": 133.44644165039062,
+      "epoch": 0.26433566433566436,
+      "grad_norm": 0.18715243199492715,
+      "kl": 0.46484375,
+      "learning_rate": 6.608391608391609e-05,
+      "loss": 0.0062,
+      "num_tokens": 1906084.0,
+      "reward": 2.364468812942505,
+      "reward_std": 0.5593475699424744,
+      "rewards/check_gptzero_func": 0.6964285969734192,
+      "rewards/check_perplexity_diff_func": 0.8571428656578064,
+      "rewards/check_winston_local_func": 0.8108974099159241,
+      "step": 189
+    },
+    {
+      "clip_ratio": 0.008103223517537117,
+      "epoch": 0.26573426573426573,
+      "grad_norm": 0.14058389303467,
+      "kl": 0.45703125,
+      "learning_rate": 6.643356643356644e-05,
+      "loss": 0.0025,
+      "step": 190
+    },
+    {
+      "clip_ratio": 0.0019668787717819214,
+      "completion_length": 142.07144165039062,
+      "epoch": 0.26713286713286716,
+      "grad_norm": 0.14399812178023874,
+      "kl": 0.43359375,
+      "learning_rate": 6.678321678321679e-05,
+      "loss": -0.0031,
+      "num_tokens": 1930224.0,
+      "reward": 2.1098926067352295,
+      "reward_std": 0.5390018820762634,
+      "rewards/check_gptzero_func": 0.8214285969734192,
+      "rewards/check_perplexity_diff_func": 0.4642857015132904,
+      "rewards/check_winston_local_func": 0.8241782784461975,
+      "step": 191
+    },
+    {
+      "clip_ratio": 0.005082913674414158,
+      "epoch": 0.26853146853146853,
+      "grad_norm": 0.12024703070308183,
+      "kl": 0.447265625,
+      "learning_rate": 6.713286713286715e-05,
+      "loss": -0.0055,
+      "step": 192
+    },
+    {
+      "clip_ratio": 0.0032251765951514244,
+      "completion_length": 84.98214721679688,
+      "epoch": 0.2699300699300699,
+      "grad_norm": 0.32527529059153654,
+      "kl": 0.6640625,
+      "learning_rate": 6.748251748251748e-05,
+      "loss": 0.0186,
+      "num_tokens": 1947399.0,
+      "reward": 1.9443087577819824,
+      "reward_std": 0.6417390704154968,
+      "rewards/check_gptzero_func": 0.7678571343421936,
+      "rewards/check_perplexity_diff_func": 0.2857142984867096,
+      "rewards/check_winston_local_func": 0.8907372355461121,
+      "step": 193
+    },
+    {
+      "clip_ratio": 0.008085422217845917,
+      "epoch": 0.27132867132867133,
+      "grad_norm": 0.21806984416977268,
+      "kl": 0.69140625,
+      "learning_rate": 6.783216783216784e-05,
+      "loss": 0.0068,
+      "step": 194
+    },
+    {
+      "clip_ratio": 0.0025381618179380894,
+      "completion_length": 91.66072082519531,
+      "epoch": 0.2727272727272727,
+      "grad_norm": 0.1831840025015104,
+      "kl": 0.53125,
+      "learning_rate": 6.818181818181818e-05,
+      "loss": 0.0014,
+      "num_tokens": 1965402.0,
+      "reward": 2.3979477882385254,
+      "reward_std": 0.45264866948127747,
+      "rewards/check_gptzero_func": 0.7678571343421936,
+      "rewards/check_perplexity_diff_func": 0.7857142686843872,
+      "rewards/check_winston_local_func": 0.8443759083747864,
+      "step": 195
+    },
+    {
+      "clip_ratio": 0.005082620773464441,
+      "epoch": 0.27412587412587414,
+      "grad_norm": 0.14399845631168656,
+      "kl": 0.515625,
+      "learning_rate": 6.853146853146853e-05,
+      "loss": -0.0023,
+      "step": 196
+    },
+    {
+      "clip_ratio": 0.0025359569117426872,
+      "completion_length": 118.80357360839844,
+      "epoch": 0.2755244755244755,
+      "grad_norm": 0.1617404225919446,
+      "kl": 0.5625,
+      "learning_rate": 6.888111888111889e-05,
+      "loss": 0.0146,
+      "num_tokens": 1986731.0,
+      "reward": 2.078895330429077,
+      "reward_std": 0.31289052963256836,
+      "rewards/check_gptzero_func": 0.8035714030265808,
+      "rewards/check_perplexity_diff_func": 0.3928571343421936,
+      "rewards/check_winston_local_func": 0.882466733455658,
+      "step": 197
+    },
+    {
+      "clip_ratio": 0.0062335156835615635,
+      "epoch": 0.27692307692307694,
+      "grad_norm": 0.12006362566688389,
+      "kl": 0.69921875,
+      "learning_rate": 6.923076923076924e-05,
+      "loss": 0.0125,
+      "step": 198
+    },
+    {
+      "clip_ratio": 0.0023628328926861286,
+      "completion_length": 113.76786041259766,
+      "epoch": 0.2783216783216783,
+      "grad_norm": 0.19235514876641358,
+      "kl": 0.58984375,
+      "learning_rate": 6.958041958041958e-05,
+      "loss": -0.0069,
+      "num_tokens": 2007822.0,
+      "reward": 2.0453529357910156,
+      "reward_std": 0.576248288154602,
+      "rewards/check_gptzero_func": 0.7321428656578064,
+      "rewards/check_perplexity_diff_func": 0.5714285969734192,
+      "rewards/check_winston_local_func": 0.7417814135551453,
+      "step": 199
+    },
+    {
+      "clip_ratio": 0.0030245708767324686,
+      "epoch": 0.27972027972027974,
+      "grad_norm": 0.15320753759426498,
+      "kl": 0.578125,
+      "learning_rate": 6.993006993006993e-05,
+      "loss": -0.0115,
+      "step": 200
+    },
+    {
+      "clip_ratio": 0.0028001146856695414,
+      "completion_length": 121.51786041259766,
+      "epoch": 0.2811188811188811,
+      "grad_norm": 0.2770839119590159,
+      "kl": 0.5859375,
+      "learning_rate": 7.027972027972029e-05,
+      "loss": -0.0321,
+      "num_tokens": 2030153.0,
+      "reward": 2.223505735397339,
+      "reward_std": 0.580276370048523,
+      "rewards/check_gptzero_func": 0.8392857313156128,
+      "rewards/check_perplexity_diff_func": 0.5,
+      "rewards/check_winston_local_func": 0.8842198252677917,
+      "step": 201
+    },
+    {
+      "clip_ratio": 0.004435424692928791,
+      "epoch": 0.28251748251748254,
+      "grad_norm": 0.1752123363810394,
+      "kl": 0.61328125,
+      "learning_rate": 7.062937062937062e-05,
+      "loss": -0.0398,
+      "step": 202
+    },
+    {
+      "clip_ratio": 0.003775001736357808,
+      "completion_length": 132.96429443359375,
+      "epoch": 0.2839160839160839,
+      "grad_norm": 0.20528144857768493,
+      "kl": 0.5390625,
+      "learning_rate": 7.097902097902098e-05,
+      "loss": -0.0054,
+      "num_tokens": 2053611.0,
+      "reward": 2.4398131370544434,
+      "reward_std": 0.6120996475219727,
+      "rewards/check_gptzero_func": 0.8214285969734192,
+      "rewards/check_perplexity_diff_func": 0.8571428656578064,
+      "rewards/check_winston_local_func": 0.761241614818573,
+      "step": 203
+    },
+    {
+      "clip_ratio": 0.0037133130244910717,
+      "epoch": 0.2853146853146853,
+      "grad_norm": 0.14916355666745199,
+      "kl": 0.55859375,
+      "learning_rate": 7.132867132867134e-05,
+      "loss": -0.0112,
+      "step": 204
+    },
+    {
+      "clip_ratio": 0.002573356730863452,
+      "completion_length": 109.51786041259766,
+      "epoch": 0.2867132867132867,
+      "grad_norm": 0.15672915743695726,
+      "kl": 1.4921875,
+      "learning_rate": 7.167832167832168e-05,
+      "loss": 0.0032,
+      "num_tokens": 2073846.0,
+      "reward": 2.221467971801758,
+      "reward_std": 0.492183119058609,
+      "rewards/check_gptzero_func": 0.6964285969734192,
+      "rewards/check_perplexity_diff_func": 0.6428571343421936,
+      "rewards/check_winston_local_func": 0.8821821212768555,
+      "step": 205
+    },
+    {
+      "clip_ratio": 0.0036338225472718477,
+      "epoch": 0.2881118881118881,
+      "grad_norm": 0.27785877634425765,
+      "kl": 0.74609375,
+      "learning_rate": 7.202797202797204e-05,
+      "loss": 0.0013,
+      "step": 206
+    },
+    {
+      "clip_ratio": 0.004580673761665821,
+      "completion_length": 95.05357360839844,
+      "epoch": 0.2895104895104895,
+      "grad_norm": 0.1746155777114636,
+      "kl": 0.671875,
+      "learning_rate": 7.237762237762238e-05,
+      "loss": -0.0068,
+      "num_tokens": 2092355.0,
+      "reward": 2.0429205894470215,
+      "reward_std": 0.36572587490081787,
+      "rewards/check_gptzero_func": 0.6964285969734192,
+      "rewards/check_perplexity_diff_func": 0.4642857015132904,
+      "rewards/check_winston_local_func": 0.8822061419487,
+      "step": 207
+    },
+    {
+      "clip_ratio": 0.007289381232112646,
+      "epoch": 0.2909090909090909,
+      "grad_norm": 0.12704093759252294,
+      "kl": 0.75390625,
+      "learning_rate": 7.272727272727273e-05,
+      "loss": -0.0111,
+      "step": 208
+    },
+    {
+      "clip_ratio": 0.0038105440326035023,
+      "completion_length": 113.4464340209961,
+      "epoch": 0.2923076923076923,
+      "grad_norm": 0.20958397036994925,
+      "kl": 0.6015625,
+      "learning_rate": 7.307692307692307e-05,
+      "loss": 0.0082,
+      "num_tokens": 2113542.0,
+      "reward": 2.162785530090332,
+      "reward_std": 0.49807849526405334,
+      "rewards/check_gptzero_func": 0.75,
+      "rewards/check_perplexity_diff_func": 0.5714285969734192,
+      "rewards/check_winston_local_func": 0.8413568139076233,
+      "step": 209
+    },
+    {
+      "clip_ratio": 0.007606918923556805,
+      "epoch": 0.2937062937062937,
+      "grad_norm": 0.1766466418371376,
+      "kl": 0.63671875,
+      "learning_rate": 7.342657342657343e-05,
+      "loss": 0.0025,
+      "step": 210
+    },
+    {
+      "clip_ratio": 0.0019986790139228106,
+      "completion_length": 92.73214721679688,
+      "epoch": 0.2951048951048951,
+      "grad_norm": 0.2716166901527731,
+      "kl": 0.6171875,
+      "learning_rate": 7.377622377622378e-05,
+      "loss": -0.0005,
+      "num_tokens": 2131873.0,
+      "reward": 2.099912405014038,
+      "reward_std": 0.5661742687225342,
+      "rewards/check_gptzero_func": 0.8214285969734192,
+      "rewards/check_perplexity_diff_func": 0.4285714328289032,
+      "rewards/check_winston_local_func": 0.8499122858047485,
+      "step": 211
+    },
+    {
+      "clip_ratio": 0.005917501635849476,
+      "epoch": 0.2965034965034965,
+      "grad_norm": 0.19270405158731238,
+      "kl": 0.62890625,
+      "learning_rate": 7.412587412587413e-05,
+      "loss": -0.0079,
+      "step": 212
+    },
+    {
+      "clip_ratio": 0.002530448604375124,
+      "completion_length": 96.51786041259766,
+      "epoch": 0.29790209790209793,
+      "grad_norm": 0.22723256705117592,
+      "kl": 0.69921875,
+      "learning_rate": 7.447552447552449e-05,
+      "loss": -0.009,
+      "num_tokens": 2150462.0,
+      "reward": 2.1355700492858887,
+      "reward_std": 0.3738899827003479,
+      "rewards/check_gptzero_func": 0.8571428656578064,
+      "rewards/check_perplexity_diff_func": 0.3571428656578064,
+      "rewards/check_winston_local_func": 0.9212842583656311,
+      "step": 213
+    },
+    {
+      "clip_ratio": 0.011397747322916985,
+      "epoch": 0.2993006993006993,
+      "grad_norm": 0.18189402035580596,
+      "kl": 0.7265625,
+      "learning_rate": 7.482517482517482e-05,
+      "loss": -0.0139,
+      "step": 214
+    },
+    {
+      "clip_ratio": 0.004513947293162346,
+      "completion_length": 94.66072082519531,
+      "epoch": 0.3006993006993007,
+      "grad_norm": 0.2675295151019955,
+      "kl": 0.67578125,
+      "learning_rate": 7.517482517482518e-05,
+      "loss": 0.0014,
+      "num_tokens": 2169243.0,
+      "reward": 2.2817916870117188,
+      "reward_std": 0.4118366539478302,
+      "rewards/check_gptzero_func": 0.8571428656578064,
+      "rewards/check_perplexity_diff_func": 0.4642857015132904,
+      "rewards/check_winston_local_func": 0.9603630900382996,
+      "step": 215
+    },
+    {
+      "clip_ratio": 0.010964194312691689,
+      "epoch": 0.3020979020979021,
+      "grad_norm": 0.17031623307626814,
+      "kl": 0.875,
+      "learning_rate": 7.552447552447553e-05,
+      "loss": -0.0057,
+      "step": 216
+    },
+    {
+      "clip_ratio": 0.003371638245880604,
+      "completion_length": 118.17857360839844,
+      "epoch": 0.3034965034965035,
+      "grad_norm": 0.1847364030092435,
+      "kl": 0.60546875,
+      "learning_rate": 7.587412587412587e-05,
+      "loss": -0.0022,
+      "num_tokens": 2190571.0,
+      "reward": 2.1225900650024414,
+      "reward_std": 0.5403507947921753,
+      "rewards/check_gptzero_func": 0.75,
+      "rewards/check_perplexity_diff_func": 0.4642857015132904,
+      "rewards/check_winston_local_func": 0.9083043336868286,
+      "step": 217
+    },
+    {
+      "clip_ratio": 0.01002263929694891,
+      "epoch": 0.3048951048951049,
+      "grad_norm": 0.1449002041728082,
+      "kl": 0.61328125,
+      "learning_rate": 7.622377622377622e-05,
+      "loss": -0.0067,
+      "step": 218
+    },
+    {
+      "clip_ratio": 0.0013253266224637628,
+      "completion_length": 129.33929443359375,
+      "epoch": 0.3062937062937063,
+      "grad_norm": 0.1504069021149342,
+      "kl": 0.578125,
+      "learning_rate": 7.657342657342658e-05,
+      "loss": -0.0075,
+      "num_tokens": 2213408.0,
+      "reward": 2.1974618434906006,
+      "reward_std": 0.4602964520454407,
+      "rewards/check_gptzero_func": 0.7857142686843872,
+      "rewards/check_perplexity_diff_func": 0.5,
+      "rewards/check_winston_local_func": 0.911747395992279,
+      "step": 219
+    },
+    {
+      "clip_ratio": 0.006876418832689524,
+      "epoch": 0.3076923076923077,
+      "grad_norm": 0.11992971275284438,
+      "kl": 0.57421875,
+      "learning_rate": 7.692307692307693e-05,
+      "loss": -0.0101,
+      "step": 220
+    },
+    {
+      "clip_ratio": 0.0024474726524204016,
+      "completion_length": 89.67857360839844,
+      "epoch": 0.3090909090909091,
+      "grad_norm": 0.46915411380362765,
+      "kl": 0.73046875,
+      "learning_rate": 7.727272727272727e-05,
+      "loss": 0.0333,
+      "num_tokens": 2231248.0,
+      "reward": 2.161350727081299,
+      "reward_std": 0.47136664390563965,
+      "rewards/check_gptzero_func": 0.8214285969734192,
+      "rewards/check_perplexity_diff_func": 0.4642857015132904,
+      "rewards/check_winston_local_func": 0.8756363987922668,
+      "step": 221
+    },
+    {
+      "clip_ratio": 0.02464432455599308,
+      "epoch": 0.3104895104895105,
+      "grad_norm": 0.30204889343007607,
+      "kl": 0.8046875,
+      "learning_rate": 7.762237762237763e-05,
+      "loss": 0.0211,
+      "step": 222
+    },
+    {
+      "clip_ratio": 0.0017527465242892504,
+      "completion_length": 134.17857360839844,
+      "epoch": 0.3118881118881119,
+      "grad_norm": 0.18948229022832336,
+      "kl": 0.6328125,
+      "learning_rate": 7.797202797202798e-05,
+      "loss": -0.0049,
+      "num_tokens": 2255162.0,
+      "reward": 2.4446465969085693,
+      "reward_std": 0.5171672105789185,
+      "rewards/check_gptzero_func": 0.7678571343421936,
+      "rewards/check_perplexity_diff_func": 0.7857142686843872,
+      "rewards/check_winston_local_func": 0.8910752534866333,
+      "step": 223
+    },
+    {
+      "clip_ratio": 0.006593752186745405,
+      "epoch": 0.3132867132867133,
+      "grad_norm": 0.15013602627394823,
+      "kl": 0.640625,
+      "learning_rate": 7.832167832167832e-05,
+      "loss": -0.0097,
+      "step": 224
+    },
+    {
+      "clip_ratio": 0.0019926591776311398,
+      "completion_length": 127.00000762939453,
+      "epoch": 0.3146853146853147,
+      "grad_norm": 0.22875481051435598,
+      "kl": 0.6875,
+      "learning_rate": 7.867132867132867e-05,
+      "loss": 0.0021,
+      "num_tokens": 2277320.0,
+      "reward": 2.2040371894836426,
+      "reward_std": 0.5528277158737183,
+      "rewards/check_gptzero_func": 0.7321428656578064,
+      "rewards/check_perplexity_diff_func": 0.5357142686843872,
+      "rewards/check_winston_local_func": 0.936180055141449,
+      "step": 225
+    },
+    {
+      "clip_ratio": 0.005839359946548939,
+      "epoch": 0.31608391608391606,
+      "grad_norm": 0.17829200080123941,
+      "kl": 0.68359375,
+      "learning_rate": 7.902097902097903e-05,
+      "loss": -0.0053,
+      "step": 226
+    },
+    {
+      "clip_ratio": 0.0026252593379467726,
+      "completion_length": 130.0357208251953,
+      "epoch": 0.3174825174825175,
+      "grad_norm": 0.16076101957107508,
+      "kl": 0.66015625,
+      "learning_rate": 7.937062937062938e-05,
+      "loss": -0.0033,
+      "num_tokens": 2299818.0,
+      "reward": 2.3278117179870605,
+      "reward_std": 0.48680880665779114,
+      "rewards/check_gptzero_func": 0.8035714030265808,
+      "rewards/check_perplexity_diff_func": 0.6428571343421936,
+      "rewards/check_winston_local_func": 0.8813830614089966,
+      "step": 227
+    },
+    {
+      "clip_ratio": 0.004954828415066004,
+      "epoch": 0.31888111888111886,
+      "grad_norm": 0.12734264227867195,
+      "kl": 0.6875,
+      "learning_rate": 7.972027972027972e-05,
+      "loss": -0.0071,
+      "step": 228
+    },
+    {
+      "clip_ratio": 0.0034341278951615095,
+      "completion_length": 133.7857208251953,
+      "epoch": 0.3202797202797203,
+      "grad_norm": 0.24065024754984402,
+      "kl": 0.6875,
+      "learning_rate": 8.006993006993007e-05,
+      "loss": -0.0052,
+      "num_tokens": 2322948.0,
+      "reward": 2.5632615089416504,
+      "reward_std": 0.6206622123718262,
+      "rewards/check_gptzero_func": 0.9642857313156128,
+      "rewards/check_perplexity_diff_func": 0.6785714030265808,
+      "rewards/check_winston_local_func": 0.9204041361808777,
+      "step": 229
+    },
+    {
+      "clip_ratio": 0.006521092262119055,
+      "epoch": 0.32167832167832167,
+      "grad_norm": 0.18602361781154253,
+      "kl": 0.6796875,
+      "learning_rate": 8.041958041958042e-05,
+      "loss": -0.013,
+      "step": 230
+    },
+    {
+      "clip_ratio": 0.004207184072583914,
+      "completion_length": 116.53572082519531,
+      "epoch": 0.3230769230769231,
+      "grad_norm": 0.36704569865090053,
+      "kl": 0.69921875,
+      "learning_rate": 8.076923076923078e-05,
+      "loss": 0.0055,
+      "num_tokens": 2344106.0,
+      "reward": 2.4917171001434326,
+      "reward_std": 0.6073458790779114,
+      "rewards/check_gptzero_func": 0.8392857313156128,
+      "rewards/check_perplexity_diff_func": 0.75,
+      "rewards/check_winston_local_func": 0.902431309223175,
+      "step": 231
+    },
+    {
+      "clip_ratio": 0.013317321427166462,
+      "epoch": 0.32447552447552447,
+      "grad_norm": 0.13844634287775134,
+      "kl": 0.73046875,
+      "learning_rate": 8.111888111888112e-05,
+      "loss": -0.0039,
+      "step": 232
+    },
+    {
+      "clip_ratio": 0.004024635534733534,
+      "completion_length": 134.92857360839844,
+      "epoch": 0.3258741258741259,
+      "grad_norm": 0.34168858248313216,
+      "kl": 0.64453125,
+      "learning_rate": 8.146853146853147e-05,
+      "loss": -0.0057,
+      "num_tokens": 2367724.0,
+      "reward": 2.20210599899292,
+      "reward_std": 0.5888614654541016,
+      "rewards/check_gptzero_func": 0.7321428656578064,
+      "rewards/check_perplexity_diff_func": 0.6071428656578064,
+      "rewards/check_winston_local_func": 0.8628200888633728,
+      "step": 233
+    },
+    {
+      "clip_ratio": 0.009708845987915993,
+      "epoch": 0.32727272727272727,
+      "grad_norm": 0.19135959991877635,
+      "kl": 0.67578125,
+      "learning_rate": 8.181818181818183e-05,
+      "loss": -0.0146,
+      "step": 234
+    },
+    {
+      "clip_ratio": 0.005571329966187477,
+      "completion_length": 98.75000762939453,
+      "epoch": 0.32867132867132864,
+      "grad_norm": 0.32328761096657005,
+      "kl": 0.77734375,
+      "learning_rate": 8.216783216783218e-05,
+      "loss": 0.0128,
+      "num_tokens": 2386636.0,
+      "reward": 2.2358410358428955,
+      "reward_std": 0.599101185798645,
+      "rewards/check_gptzero_func": 0.8214285969734192,
+      "rewards/check_perplexity_diff_func": 0.5357142686843872,
+      "rewards/check_winston_local_func": 0.8786982297897339,
+      "step": 235
+    },
+    {
+      "clip_ratio": 0.009436835534870625,
+      "epoch": 0.3300699300699301,
+      "grad_norm": 0.19557085227388898,
+      "kl": 0.8515625,
+      "learning_rate": 8.251748251748252e-05,
+      "loss": 0.0022,
+      "step": 236
+    },
+    {
+      "clip_ratio": 0.006164635997265577,
+      "completion_length": 79.75,
+      "epoch": 0.33146853146853145,
+      "grad_norm": 0.3250433111655094,
+      "kl": 0.98828125,
+      "learning_rate": 8.286713286713287e-05,
+      "loss": -0.0126,
+      "num_tokens": 2403400.0,
+      "reward": 2.1074206829071045,
+      "reward_std": 0.41642776131629944,
+      "rewards/check_gptzero_func": 0.9107142686843872,
+      "rewards/check_perplexity_diff_func": 0.25,
+      "rewards/check_winston_local_func": 0.946706235408783,
+      "step": 237
+    },
+    {
+      "clip_ratio": 0.0066314926370978355,
+      "epoch": 0.3328671328671329,
+      "grad_norm": 0.1612667131666411,
+      "kl": 0.98828125,
+      "learning_rate": 8.321678321678323e-05,
+      "loss": -0.0225,
+      "step": 238
+    },
+    {
+      "clip_ratio": 0.003079179208725691,
+      "completion_length": 117.10714721679688,
+      "epoch": 0.33426573426573425,
+      "grad_norm": 0.2505534626607256,
+      "kl": 0.6796875,
+      "learning_rate": 8.356643356643356e-05,
+      "loss": -0.0103,
+      "num_tokens": 2425354.0,
+      "reward": 2.3166799545288086,
+      "reward_std": 0.5270359516143799,
+      "rewards/check_gptzero_func": 0.8214285969734192,
+      "rewards/check_perplexity_diff_func": 0.5714285969734192,
+      "rewards/check_winston_local_func": 0.9238227605819702,
+      "step": 239
+    },
+    {
+      "clip_ratio": 0.006801524665206671,
+      "epoch": 0.3356643356643357,
+      "grad_norm": 0.16826419365505382,
+      "kl": 0.671875,
+      "learning_rate": 8.391608391608392e-05,
+      "loss": -0.0182,
+      "step": 240
+    },
+    {
+      "clip_ratio": 0.004480496048927307,
+      "completion_length": 125.05357360839844,
+      "epoch": 0.33706293706293705,
+      "grad_norm": 0.26807661638826913,
+      "kl": 0.60546875,
+      "learning_rate": 8.426573426573428e-05,
+      "loss": -0.0086,
+      "num_tokens": 2447333.0,
+      "reward": 2.2939677238464355,
+      "reward_std": 0.39558398723602295,
+      "rewards/check_gptzero_func": 0.9107142686843872,
+      "rewards/check_perplexity_diff_func": 0.5357142686843872,
+      "rewards/check_winston_local_func": 0.8475390672683716,
+      "step": 241
+    },
+    {
+      "clip_ratio": 0.007587091531604528,
+      "epoch": 0.3384615384615385,
+      "grad_norm": 0.13602673124742348,
+      "kl": 0.57421875,
+      "learning_rate": 8.461538461538461e-05,
+      "loss": -0.0152,
+      "step": 242
+    },
+    {
+      "clip_ratio": 0.0034923183266073465,
+      "completion_length": 98.71428680419922,
+      "epoch": 0.33986013986013985,
+      "grad_norm": 0.7773422467135167,
+      "kl": 6.5,
+      "learning_rate": 8.496503496503497e-05,
+      "loss": 0.0071,
+      "num_tokens": 2466435.0,
+      "reward": 2.4026196002960205,
+      "reward_std": 0.4264836013317108,
+      "rewards/check_gptzero_func": 0.8571428656578064,
+      "rewards/check_perplexity_diff_func": 0.6785714030265808,
+      "rewards/check_winston_local_func": 0.8669052124023438,
+      "step": 243
+    },
+    {
+      "clip_ratio": 0.026189187541604042,
+      "epoch": 0.3412587412587413,
+      "grad_norm": 130.9971047023429,
+      "kl": 0.70703125,
+      "learning_rate": 8.531468531468532e-05,
+      "loss": 1.8998,
+      "step": 244
+    },
+    {
+      "clip_ratio": 0.004780410788953304,
+      "completion_length": 135.0,
+      "epoch": 0.34265734265734266,
+      "grad_norm": 0.2329479243774761,
+      "kl": 0.87109375,
+      "learning_rate": 8.566433566433567e-05,
+      "loss": 0.0011,
+      "num_tokens": 2489651.0,
+      "reward": 2.2166569232940674,
+      "reward_std": 0.5797281265258789,
+      "rewards/check_gptzero_func": 0.9107142686843872,
+      "rewards/check_perplexity_diff_func": 0.4285714328289032,
+      "rewards/check_winston_local_func": 0.8773713111877441,
+      "step": 245
+    },
+    {
+      "clip_ratio": 0.021273447200655937,
+      "epoch": 0.34405594405594403,
+      "grad_norm": 0.22672502397660207,
+      "kl": 0.9609375,
+      "learning_rate": 8.601398601398601e-05,
+      "loss": -0.0045,
+      "step": 246
+    },
+    {
+      "clip_ratio": 0.002856873208656907,
+      "completion_length": 122.66072082519531,
+      "epoch": 0.34545454545454546,
+      "grad_norm": 0.19772681964626543,
+      "kl": 0.984375,
+      "learning_rate": 8.636363636363637e-05,
+      "loss": 0.0028,
+      "num_tokens": 2510506.0,
+      "reward": 2.323245048522949,
+      "reward_std": 0.46726977825164795,
+      "rewards/check_gptzero_func": 0.9107142686843872,
+      "rewards/check_perplexity_diff_func": 0.5,
+      "rewards/check_winston_local_func": 0.9125306010246277,
+      "step": 247
+    },
+    {
+      "clip_ratio": 0.02299562282860279,
+      "epoch": 0.34685314685314683,
+      "grad_norm": 0.1812005085005865,
+      "kl": 0.87890625,
+      "learning_rate": 8.67132867132867e-05,
+      "loss": -0.001,
+      "step": 248
+    },
+    {
+      "clip_ratio": 0.0026335662696510553,
+      "completion_length": 116.85714721679688,
+      "epoch": 0.34825174825174826,
+      "grad_norm": 0.28731547808119623,
+      "kl": 0.76953125,
+      "learning_rate": 8.706293706293707e-05,
+      "loss": 0.024,
+      "num_tokens": 2531254.0,
+      "reward": 2.5312695503234863,
+      "reward_std": 0.5851892828941345,
+      "rewards/check_gptzero_func": 0.8571428656578064,
+      "rewards/check_perplexity_diff_func": 0.8214285969734192,
+      "rewards/check_winston_local_func": 0.8526979088783264,
+      "step": 249
+    },
+    {
+      "clip_ratio": 0.012148472480475903,
+      "epoch": 0.34965034965034963,
+      "grad_norm": 0.21250398409595742,
+      "kl": 0.734375,
+      "learning_rate": 8.741258741258743e-05,
+      "loss": 0.0156,
+      "step": 250
+    },
+    {
+      "clip_ratio": 0.0036881309933960438,
+      "completion_length": 109.12500762939453,
+      "epoch": 0.35104895104895106,
+      "grad_norm": 0.278787459589393,
+      "kl": 1.3671875,
+      "learning_rate": 8.776223776223776e-05,
+      "loss": -0.0127,
+      "num_tokens": 2551265.0,
+      "reward": 2.24423885345459,
+      "reward_std": 0.6221095323562622,
+      "rewards/check_gptzero_func": 0.875,
+      "rewards/check_perplexity_diff_func": 0.5,
+      "rewards/check_winston_local_func": 0.8692389130592346,
+      "step": 251
+    },
+    {
+      "clip_ratio": 0.011438491754233837,
+      "epoch": 0.35244755244755244,
+      "grad_norm": 0.21324998809887244,
+      "kl": 1.0859375,
+      "learning_rate": 8.811188811188812e-05,
+      "loss": -0.023,
+      "step": 252
+    },
+    {
+      "clip_ratio": 0.0039367591962218285,
+      "completion_length": 112.14286041259766,
+      "epoch": 0.35384615384615387,
+      "grad_norm": 0.31069817037713615,
+      "kl": 0.91015625,
+      "learning_rate": 8.846153846153847e-05,
+      "loss": 0.0167,
+      "num_tokens": 2572119.0,
+      "reward": 2.492161989212036,
+      "reward_std": 0.6407585144042969,
+      "rewards/check_gptzero_func": 0.875,
+      "rewards/check_perplexity_diff_func": 0.75,
+      "rewards/check_winston_local_func": 0.8671619296073914,
+      "step": 253
+    },
+    {
+      "clip_ratio": 0.005880096461623907,
+      "epoch": 0.35524475524475524,
+      "grad_norm": 0.19008007229653096,
+      "kl": 0.90234375,
+      "learning_rate": 8.881118881118881e-05,
+      "loss": 0.0057,
+      "step": 254
+    },
+    {
+      "clip_ratio": 0.003151817014440894,
+      "completion_length": 120.78572082519531,
+      "epoch": 0.35664335664335667,
+      "grad_norm": 0.4113945542101958,
+      "kl": 0.890625,
+      "learning_rate": 8.916083916083916e-05,
+      "loss": 0.0079,
+      "num_tokens": 2593081.0,
+      "reward": 2.3228745460510254,
+      "reward_std": 0.5329480171203613,
+      "rewards/check_gptzero_func": 0.875,
+      "rewards/check_perplexity_diff_func": 0.6428571343421936,
+      "rewards/check_winston_local_func": 0.8050170540809631,
+      "step": 255
+    },
+    {
+      "clip_ratio": 0.01224872563034296,
+      "epoch": 0.35804195804195804,
+      "grad_norm": 0.24677625747594412,
+      "kl": 1.0078125,
+      "learning_rate": 8.951048951048952e-05,
+      "loss": 0.0006,
+      "step": 256
+    },
+    {
+      "clip_ratio": 0.003666934324428439,
+      "completion_length": 146.5178680419922,
+      "epoch": 0.3594405594405594,
+      "grad_norm": 0.21537926236506497,
+      "kl": 0.484375,
+      "learning_rate": 8.986013986013986e-05,
+      "loss": 0.0068,
+      "num_tokens": 2618758.0,
+      "reward": 2.198406934738159,
+      "reward_std": 0.3875991106033325,
+      "rewards/check_gptzero_func": 0.9285714030265808,
+      "rewards/check_perplexity_diff_func": 0.4285714328289032,
+      "rewards/check_winston_local_func": 0.8412641882896423,
+      "step": 257
+    },
+    {
+      "clip_ratio": 0.012731654569506645,
+      "epoch": 0.36083916083916084,
+      "grad_norm": 0.1543455831571161,
+      "kl": 0.478515625,
+      "learning_rate": 9.020979020979021e-05,
+      "loss": 0.0025,
+      "step": 258
+    },
+    {
+      "clip_ratio": 0.0012050624936819077,
+      "completion_length": 129.375,
+      "epoch": 0.3622377622377622,
+      "grad_norm": 0.16107873823673136,
+      "kl": 1.6640625,
+      "learning_rate": 9.055944055944057e-05,
+      "loss": 0.0054,
+      "num_tokens": 2641815.0,
+      "reward": 2.4805288314819336,
+      "reward_std": 0.3510296940803528,
+      "rewards/check_gptzero_func": 0.8392857313156128,
+      "rewards/check_perplexity_diff_func": 0.8214285969734192,
+      "rewards/check_winston_local_func": 0.8198142647743225,
+      "step": 259
+    },
+    {
+      "clip_ratio": 0.006208478473126888,
+      "epoch": 0.36363636363636365,
+      "grad_norm": 0.10546346547579268,
+      "kl": 0.8046875,
+      "learning_rate": 9.090909090909092e-05,
+      "loss": 0.0019,
+      "step": 260
+    },
+    {
+      "clip_ratio": 0.0025593352038413286,
+      "completion_length": 128.57144165039062,
+      "epoch": 0.365034965034965,
+      "grad_norm": 0.1986264685447053,
+      "kl": 0.6171875,
+      "learning_rate": 9.125874125874126e-05,
+      "loss": -0.0025,
+      "num_tokens": 2664343.0,
+      "reward": 2.261420249938965,
+      "reward_std": 0.5704914927482605,
+      "rewards/check_gptzero_func": 0.7857142686843872,
+      "rewards/check_perplexity_diff_func": 0.6428571343421936,
+      "rewards/check_winston_local_func": 0.8328486680984497,
+      "step": 261
+    },
+    {
+      "clip_ratio": 0.002767725382000208,
+      "epoch": 0.36643356643356645,
+      "grad_norm": 0.14267481856078115,
+      "kl": 0.66015625,
+      "learning_rate": 9.160839160839161e-05,
+      "loss": -0.0071,
+      "step": 262
+    },
+    {
+      "clip_ratio": 0.0036019759718328714,
+      "completion_length": 108.30357360839844,
+      "epoch": 0.3678321678321678,
+      "grad_norm": 0.25429899311591847,
+      "kl": 0.61328125,
+      "learning_rate": 9.195804195804196e-05,
+      "loss": 0.0114,
+      "num_tokens": 2684460.0,
+      "reward": 2.1853349208831787,
+      "reward_std": 0.5545978546142578,
+      "rewards/check_gptzero_func": 0.7142857313156128,
+      "rewards/check_perplexity_diff_func": 0.6428571343421936,
+      "rewards/check_winston_local_func": 0.828191876411438,
+      "step": 263
+    },
+    {
+      "clip_ratio": 0.002995865885168314,
+      "epoch": 0.36923076923076925,
+      "grad_norm": 0.17454926732818832,
+      "kl": 0.62109375,
+      "learning_rate": 9.230769230769232e-05,
+      "loss": 0.0028,
+      "step": 264
+    },
+    {
+      "clip_ratio": 0.002558046253398061,
+      "completion_length": 102.80357360839844,
+      "epoch": 0.3706293706293706,
+      "grad_norm": 0.18701691067416304,
+      "kl": 0.734375,
+      "learning_rate": 9.265734265734266e-05,
+      "loss": 0.0401,
+      "num_tokens": 2704225.0,
+      "reward": 2.1770143508911133,
+      "reward_std": 0.573693037033081,
+      "rewards/check_gptzero_func": 0.8035714030265808,
+      "rewards/check_perplexity_diff_func": 0.5357142686843872,
+      "rewards/check_winston_local_func": 0.8377286195755005,
+      "step": 265
+    },
+    {
+      "clip_ratio": 0.006838452070951462,
+      "epoch": 0.37202797202797205,
+      "grad_norm": 0.13942096443359253,
+      "kl": 0.71875,
+      "learning_rate": 9.300699300699301e-05,
+      "loss": 0.0354,
+      "step": 266
+    },
+    {
+      "clip_ratio": 0.00375689216889441,
+      "completion_length": 106.37500762939453,
+      "epoch": 0.3734265734265734,
+      "grad_norm": 0.25266992310982356,
+      "kl": 0.7421875,
+      "learning_rate": 9.335664335664336e-05,
+      "loss": -0.0156,
+      "num_tokens": 2724530.0,
+      "reward": 2.11887526512146,
+      "reward_std": 0.5366143584251404,
+      "rewards/check_gptzero_func": 0.875,
+      "rewards/check_perplexity_diff_func": 0.4642857015132904,
+      "rewards/check_winston_local_func": 0.7795897126197815,
+      "step": 267
+    },
+    {
+      "clip_ratio": 0.006593361962586641,
+      "epoch": 0.3748251748251748,
+      "grad_norm": 0.18497903526945206,
+      "kl": 0.7265625,
+      "learning_rate": 9.370629370629372e-05,
+      "loss": -0.0243,
+      "step": 268
+    },
+    {
+      "clip_ratio": 0.004569682292640209,
+      "completion_length": 124.39286041259766,
+      "epoch": 0.37622377622377623,
+      "grad_norm": 0.21331332062489702,
+      "kl": 0.953125,
+      "learning_rate": 9.405594405594406e-05,
+      "loss": 0.0346,
+      "num_tokens": 2746246.0,
+      "reward": 2.403602123260498,
+      "reward_std": 0.5621734857559204,
+      "rewards/check_gptzero_func": 0.7857142686843872,
+      "rewards/check_perplexity_diff_func": 0.75,
+      "rewards/check_winston_local_func": 0.8678879141807556,
+      "step": 269
+    },
+    {
+      "clip_ratio": 0.006006812676787376,
+      "epoch": 0.3776223776223776,
+      "grad_norm": 0.15420322253654764,
+      "kl": 0.8671875,
+      "learning_rate": 9.440559440559441e-05,
+      "loss": 0.0283,
+      "step": 270
+    },
+    {
+      "clip_ratio": 0.002397142816334963,
+      "completion_length": 102.41072082519531,
+      "epoch": 0.37902097902097903,
+      "grad_norm": 0.22567114655173642,
+      "kl": 0.609375,
+      "learning_rate": 9.475524475524477e-05,
+      "loss": 0.0231,
+      "num_tokens": 2765203.0,
+      "reward": 2.649076223373413,
+      "reward_std": 0.47521141171455383,
+      "rewards/check_gptzero_func": 0.875,
+      "rewards/check_perplexity_diff_func": 1.0,
+      "rewards/check_winston_local_func": 0.7740762829780579,
+      "step": 271
+    },
+    {
+      "clip_ratio": 0.007639365270733833,
+      "epoch": 0.3804195804195804,
+      "grad_norm": 0.18662930190408594,
+      "kl": 0.578125,
+      "learning_rate": 9.510489510489511e-05,
+      "loss": 0.0165,
+      "step": 272
+    },
+    {
+      "clip_ratio": 0.0010056205792352557,
+      "completion_length": 115.28572082519531,
+      "epoch": 0.38181818181818183,
+      "grad_norm": 0.14954423121719349,
+      "kl": 0.625,
+      "learning_rate": 9.545454545454546e-05,
+      "loss": 0.0237,
+      "num_tokens": 2785985.0,
+      "reward": 2.3541975021362305,
+      "reward_std": 0.5445494651794434,
+      "rewards/check_gptzero_func": 0.8392857313156128,
+      "rewards/check_perplexity_diff_func": 0.75,
+      "rewards/check_winston_local_func": 0.7649118304252625,
+      "step": 273
+    },
+    {
+      "clip_ratio": 0.003297280054539442,
+      "epoch": 0.3832167832167832,
+      "grad_norm": 0.11713806873969078,
+      "kl": 0.640625,
+      "learning_rate": 9.580419580419581e-05,
+      "loss": 0.0206,
+      "step": 274
+    },
+    {
+      "clip_ratio": 0.0016172031173482537,
+      "completion_length": 128.73214721679688,
+      "epoch": 0.38461538461538464,
+      "grad_norm": 0.155009547701029,
+      "kl": 0.51171875,
+      "learning_rate": 9.615384615384617e-05,
+      "loss": 0.0191,
+      "num_tokens": 2808414.0,
+      "reward": 2.376694679260254,
+      "reward_std": 0.33215656876564026,
+      "rewards/check_gptzero_func": 0.8392857313156128,
+      "rewards/check_perplexity_diff_func": 0.8571428656578064,
+      "rewards/check_winston_local_func": 0.6802659630775452,
+      "step": 275
+    },
+    {
+      "clip_ratio": 0.004294655751436949,
+      "epoch": 0.386013986013986,
+      "grad_norm": 0.11409067239204185,
+      "kl": 0.515625,
+      "learning_rate": 9.65034965034965e-05,
+      "loss": 0.0155,
+      "step": 276
+    },
+    {
+      "clip_ratio": 0.002925862092524767,
+      "completion_length": 116.41072082519531,
+      "epoch": 0.38741258741258744,
+      "grad_norm": 0.2764503095198425,
+      "kl": 0.578125,
+      "learning_rate": 9.685314685314686e-05,
+      "loss": 0.01,
+      "num_tokens": 2829285.0,
+      "reward": 2.3352210521698,
+      "reward_std": 0.7207943201065063,
+      "rewards/check_gptzero_func": 0.6607142686843872,
+      "rewards/check_perplexity_diff_func": 0.9642857313156128,
+      "rewards/check_winston_local_func": 0.7102211713790894,
+      "step": 277
+    },
+    {
+      "clip_ratio": 0.005557883996516466,
+      "epoch": 0.3888111888111888,
+      "grad_norm": 0.2140511595370052,
+      "kl": 0.5859375,
+      "learning_rate": 9.72027972027972e-05,
+      "loss": -0.0004,
+      "step": 278
+    },
+    {
+      "clip_ratio": 0.0029403052758425474,
+      "completion_length": 114.25000762939453,
+      "epoch": 0.3902097902097902,
+      "grad_norm": 0.21363596109825803,
+      "kl": 0.63671875,
+      "learning_rate": 9.755244755244755e-05,
+      "loss": 0.0259,
+      "num_tokens": 2849989.0,
+      "reward": 2.6570937633514404,
+      "reward_std": 0.5007120370864868,
+      "rewards/check_gptzero_func": 0.75,
+      "rewards/check_perplexity_diff_func": 1.2142857313156128,
+      "rewards/check_winston_local_func": 0.6928080320358276,
+      "step": 279
+    },
+    {
+      "clip_ratio": 0.00384224159643054,
+      "epoch": 0.3916083916083916,
+      "grad_norm": 0.15102706915256958,
+      "kl": 0.6875,
+      "learning_rate": 9.790209790209791e-05,
+      "loss": 0.0185,
+      "step": 280
+    },
+    {
+      "clip_ratio": 0.0027283646631985903,
+      "completion_length": 90.25000762939453,
+      "epoch": 0.393006993006993,
+      "grad_norm": 0.3585134615134873,
+      "kl": 0.73046875,
+      "learning_rate": 9.825174825174826e-05,
+      "loss": -0.006,
+      "num_tokens": 2867567.0,
+      "reward": 2.4528698921203613,
+      "reward_std": 0.6383817791938782,
+      "rewards/check_gptzero_func": 0.6071428656578064,
+      "rewards/check_perplexity_diff_func": 1.1428571939468384,
+      "rewards/check_winston_local_func": 0.7028695940971375,
+      "step": 281
+    },
+    {
+      "clip_ratio": 0.016295742243528366,
+      "epoch": 0.3944055944055944,
+      "grad_norm": 0.2358766583499307,
+      "kl": 0.73046875,
+      "learning_rate": 9.86013986013986e-05,
+      "loss": -0.0151,
+      "step": 282
+    },
+    {
+      "clip_ratio": 0.0025735762901604176,
+      "completion_length": 132.125,
+      "epoch": 0.3958041958041958,
+      "grad_norm": 0.12427534341566092,
+      "kl": 0.515625,
+      "learning_rate": 9.895104895104895e-05,
+      "loss": 0.006,
+      "num_tokens": 2890954.0,
+      "reward": 2.172079086303711,
+      "reward_std": 0.31105130910873413,
+      "rewards/check_gptzero_func": 0.6607142686843872,
+      "rewards/check_perplexity_diff_func": 0.7857142686843872,
+      "rewards/check_winston_local_func": 0.7256504893302917,
+      "step": 283
+    },
+    {
+      "clip_ratio": 0.008568070828914642,
+      "epoch": 0.3972027972027972,
+      "grad_norm": 0.10751149742872082,
+      "kl": 0.5234375,
+      "learning_rate": 9.930069930069931e-05,
+      "loss": 0.0038,
+      "step": 284
+    },
+    {
+      "clip_ratio": 0.0013303400482982397,
+      "completion_length": 115.50000762939453,
+      "epoch": 0.3986013986013986,
+      "grad_norm": 0.11155145645226237,
+      "kl": 0.427734375,
+      "learning_rate": 9.965034965034964e-05,
+      "loss": 0.0136,
+      "num_tokens": 2912320.0,
+      "reward": 2.499699354171753,
+      "reward_std": 0.22870054841041565,
+      "rewards/check_gptzero_func": 0.6071428656578064,
+      "rewards/check_perplexity_diff_func": 1.3214285373687744,
+      "rewards/check_winston_local_func": 0.5711276531219482,
+      "step": 285
+    },
+    {
+      "clip_ratio": 0.004690885543823242,
+      "epoch": 0.4,
+      "grad_norm": 0.08246302090980155,
+      "kl": 0.443359375,
+      "learning_rate": 0.0001,
+      "loss": 0.0117,
+      "step": 286
+    },
+    {
+      "clip_ratio": 0.0013457606546580791,
+      "completion_length": 153.60714721679688,
+      "epoch": 0.4013986013986014,
+      "grad_norm": 0.0959414180998777,
+      "kl": 0.458984375,
+      "learning_rate": 9.999996275889018e-05,
+      "loss": -0.0037,
+      "num_tokens": 2937286.0,
+      "reward": 2.60290789604187,
+      "reward_std": 0.2643657624721527,
+      "rewards/check_gptzero_func": 0.625,
+      "rewards/check_perplexity_diff_func": 1.3214285373687744,
+      "rewards/check_winston_local_func": 0.6564791798591614,
+      "step": 287
+    },
+    {
+      "clip_ratio": 0.0025680752005428076,
+      "epoch": 0.4027972027972028,
+      "grad_norm": 0.08295696905412306,
+      "kl": 0.453125,
+      "learning_rate": 9.999985103561615e-05,
+      "loss": -0.0053,
+      "step": 288
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 106.39286041259766,
+      "epoch": 0.4041958041958042,
+      "grad_norm": 0.12524695486629228,
+      "kl": 0.3671875,
+      "learning_rate": 9.999966483034437e-05,
+      "loss": -0.0031,
+      "num_tokens": 2957572.0,
+      "reward": 2.607825994491577,
+      "reward_std": 0.2113000899553299,
+      "rewards/check_gptzero_func": 0.5535714030265808,
+      "rewards/check_perplexity_diff_func": 1.4642857313156128,
+      "rewards/check_winston_local_func": 0.5899689793586731,
+      "step": 289
+    },
+    {
+      "clip_ratio": 0.0015743181575089693,
+      "epoch": 0.40559440559440557,
+      "grad_norm": 0.08958540244405325,
+      "kl": 0.36328125,
+      "learning_rate": 9.999940414335222e-05,
+      "loss": -0.0046,
+      "step": 290
+    },
+    {
+      "clip_ratio": 0.0017556955572217703,
+      "completion_length": 96.0714340209961,
+      "epoch": 0.406993006993007,
+      "grad_norm": 0.19185238369967877,
+      "kl": 2.109375,
+      "learning_rate": 9.999906897502803e-05,
+      "loss": 0.008,
+      "num_tokens": 2976076.0,
+      "reward": 2.6148557662963867,
+      "reward_std": 0.24822747707366943,
+      "rewards/check_gptzero_func": 0.5178571343421936,
+      "rewards/check_perplexity_diff_func": 1.4285714626312256,
+      "rewards/check_winston_local_func": 0.6684269309043884,
+      "step": 291
+    },
+    {
+      "clip_ratio": 0.00337741756811738,
+      "epoch": 0.4083916083916084,
+      "grad_norm": 0.08769186584743054,
+      "kl": 0.671875,
+      "learning_rate": 9.999865932587107e-05,
+      "loss": 0.0046,
+      "step": 292
+    },
+    {
+      "clip_ratio": 0.0012849332997575402,
+      "completion_length": 91.87500762939453,
+      "epoch": 0.4097902097902098,
+      "grad_norm": 0.13036265265723118,
+      "kl": 0.4453125,
+      "learning_rate": 9.999817519649158e-05,
+      "loss": 0.0127,
+      "num_tokens": 2994773.0,
+      "reward": 2.552300214767456,
+      "reward_std": 0.14500018954277039,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 1.5,
+      "rewards/check_winston_local_func": 0.6237286329269409,
+      "step": 293
+    },
+    {
+      "clip_ratio": 0.0027297178748995066,
+      "epoch": 0.4111888111888112,
+      "grad_norm": 0.0976002240930887,
+      "kl": 0.439453125,
+      "learning_rate": 9.99976165876107e-05,
+      "loss": 0.0101,
+      "step": 294
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 113.35714721679688,
+      "epoch": 0.4125874125874126,
+      "grad_norm": 0.02609480806956132,
+      "kl": 0.1591796875,
+      "learning_rate": 9.999698350006063e-05,
+      "loss": 0.0012,
+      "num_tokens": 3015993.0,
+      "reward": 2.874926805496216,
+      "reward_std": 0.0874238833785057,
+      "rewards/check_gptzero_func": 0.4821428656578064,
+      "rewards/check_perplexity_diff_func": 1.8928571939468384,
+      "rewards/check_winston_local_func": 0.4999266266822815,
+      "step": 295
+    },
+    {
+      "clip_ratio": 9.294544724980369e-05,
+      "epoch": 0.413986013986014,
+      "grad_norm": 0.02050573548257188,
+      "kl": 0.16015625,
+      "learning_rate": 9.999627593478442e-05,
+      "loss": 0.0011,
+      "step": 296
+    },
+    {
+      "clip_ratio": 0.00012664640962611884,
+      "completion_length": 98.5714340209961,
+      "epoch": 0.4153846153846154,
+      "grad_norm": 0.01680753751461254,
+      "kl": 0.201171875,
+      "learning_rate": 9.999549389283606e-05,
+      "loss": -0.0003,
+      "num_tokens": 3035155.0,
+      "reward": 3.0224545001983643,
+      "reward_std": 0.016296973451972008,
+      "rewards/check_gptzero_func": 0.4107142984867096,
+      "rewards/check_perplexity_diff_func": 1.9642857313156128,
+      "rewards/check_winston_local_func": 0.647454559803009,
+      "step": 297
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.4167832167832168,
+      "grad_norm": 0.012884367061070623,
+      "kl": 0.203125,
+      "learning_rate": 9.999463737538053e-05,
+      "loss": -0.0003,
+      "step": 298
+    },
+    {
+      "clip_ratio": 0.0003541912592481822,
+      "completion_length": 95.76786041259766,
+      "epoch": 0.41818181818181815,
+      "grad_norm": 0.04526732622569909,
+      "kl": 0.291015625,
+      "learning_rate": 9.999370638369377e-05,
+      "loss": 0.0002,
+      "num_tokens": 3054174.0,
+      "reward": 2.3303959369659424,
+      "reward_std": 0.08971218019723892,
+      "rewards/check_gptzero_func": 0.3035714328289032,
+      "rewards/check_perplexity_diff_func": 1.6071428060531616,
+      "rewards/check_winston_local_func": 0.4196813404560089,
+      "step": 299
+    },
+    {
+      "clip_ratio": 0.00011806376278400421,
+      "epoch": 0.4195804195804196,
+      "grad_norm": 0.03862986955837794,
+      "kl": 0.296875,
+      "learning_rate": 9.999270091916257e-05,
+      "loss": -0.0002,
+      "step": 300
+    },
+    {
+      "clip_ratio": 0.0012040403671562672,
+      "completion_length": 101.73214721679688,
+      "epoch": 0.42097902097902096,
+      "grad_norm": 0.12616717395678975,
+      "kl": 0.35546875,
+      "learning_rate": 9.999162098328474e-05,
+      "loss": 0.0033,
+      "num_tokens": 3073401.0,
+      "reward": 2.8153316974639893,
+      "reward_std": 0.18918544054031372,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 1.7142857313156128,
+      "rewards/check_winston_local_func": 0.6724745631217957,
+      "step": 301
+    },
+    {
+      "clip_ratio": 0.0013847867958247662,
+      "epoch": 0.4223776223776224,
+      "grad_norm": 0.06929991057149441,
+      "kl": 0.412109375,
+      "learning_rate": 9.999046657766903e-05,
+      "loss": 0.0012,
+      "step": 302
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 81.3214340209961,
+      "epoch": 0.42377622377622376,
+      "grad_norm": 0.06730614630696258,
+      "kl": 0.2099609375,
+      "learning_rate": 9.998923770403505e-05,
+      "loss": -0.0019,
+      "num_tokens": 3090625.0,
+      "reward": 2.802220106124878,
+      "reward_std": 0.03781326860189438,
+      "rewards/check_gptzero_func": 0.375,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.42722010612487793,
+      "step": 303
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.4251748251748252,
+      "grad_norm": 0.022175315902565728,
+      "kl": 0.2109375,
+      "learning_rate": 9.99879343642134e-05,
+      "loss": -0.0024,
+      "step": 304
+    },
+    {
+      "clip_ratio": 0.00013683634460903704,
+      "completion_length": 105.53572082519531,
+      "epoch": 0.42657342657342656,
+      "grad_norm": 0.04867704636152166,
+      "kl": 0.173828125,
+      "learning_rate": 9.998655656014561e-05,
+      "loss": 0.0117,
+      "num_tokens": 3110877.0,
+      "reward": 2.6083197593688965,
+      "reward_std": 0.13502858579158783,
+      "rewards/check_gptzero_func": 0.3571428656578064,
+      "rewards/check_perplexity_diff_func": 1.8928571939468384,
+      "rewards/check_winston_local_func": 0.35831958055496216,
+      "step": 305
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.427972027972028,
+      "grad_norm": 0.043987709695075865,
+      "kl": 0.1728515625,
+      "learning_rate": 9.99851042938841e-05,
+      "loss": 0.0113,
+      "step": 306
+    },
+    {
+      "clip_ratio": 0.0004155792994424701,
+      "completion_length": 124.03572082519531,
+      "epoch": 0.42937062937062936,
+      "grad_norm": 0.03684760048760383,
+      "kl": 0.2080078125,
+      "learning_rate": 9.998357756759222e-05,
+      "loss": -0.0258,
+      "num_tokens": 3134097.0,
+      "reward": 2.7439539432525635,
+      "reward_std": 0.11075940728187561,
+      "rewards/check_gptzero_func": 0.4642857015132904,
+      "rewards/check_perplexity_diff_func": 1.7857142686843872,
+      "rewards/check_winston_local_func": 0.4939536452293396,
+      "step": 307
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.4307692307692308,
+      "grad_norm": 0.013648247572273129,
+      "kl": 0.20703125,
+      "learning_rate": 9.998197638354428e-05,
+      "loss": -0.026,
+      "step": 308
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 89.12500762939453,
+      "epoch": 0.43216783216783217,
+      "grad_norm": 0.02465210474674629,
+      "kl": 0.443359375,
+      "learning_rate": 9.998030074412545e-05,
+      "loss": 0.0014,
+      "num_tokens": 3152358.0,
+      "reward": 2.420307159423828,
+      "reward_std": 0.11704181134700775,
+      "rewards/check_gptzero_func": 0.2321428507566452,
+      "rewards/check_perplexity_diff_func": 1.75,
+      "rewards/check_winston_local_func": 0.4381641745567322,
+      "step": 309
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.43356643356643354,
+      "grad_norm": 0.021184461362474653,
+      "kl": 0.48828125,
+      "learning_rate": 9.997855065183184e-05,
+      "loss": 0.0014,
+      "step": 310
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 115.73214721679688,
+      "epoch": 0.43496503496503497,
+      "grad_norm": 0.0049134080725686995,
+      "kl": 0.162109375,
+      "learning_rate": 9.997672610927047e-05,
+      "loss": -0.0047,
+      "num_tokens": 3173997.0,
+      "reward": 2.61786150932312,
+      "reward_std": 0.013112460263073444,
+      "rewards/check_gptzero_func": 0.375,
+      "rewards/check_perplexity_diff_func": 1.8214285373687744,
+      "rewards/check_winston_local_func": 0.4214327931404114,
+      "step": 311
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.43636363636363634,
+      "grad_norm": 0.00595537475288129,
+      "kl": 0.173828125,
+      "learning_rate": 9.997482711915927e-05,
+      "loss": -0.0047,
+      "step": 312
+    },
+    {
+      "clip_ratio": 0.0001360544265480712,
+      "completion_length": 90.42857360839844,
+      "epoch": 0.43776223776223777,
+      "grad_norm": 0.007409687643468917,
+      "kl": 0.2275390625,
+      "learning_rate": 9.997285368432703e-05,
+      "loss": 0.0006,
+      "num_tokens": 3192417.0,
+      "reward": 2.682650566101074,
+      "reward_std": 0.02417343109846115,
+      "rewards/check_gptzero_func": 0.375,
+      "rewards/check_perplexity_diff_func": 1.8214285373687744,
+      "rewards/check_winston_local_func": 0.4862218499183655,
+      "step": 313
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.43916083916083914,
+      "grad_norm": 0.007129921917131849,
+      "kl": 0.2275390625,
+      "learning_rate": 9.997080580771349e-05,
+      "loss": 0.0006,
+      "step": 314
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 83.14286041259766,
+      "epoch": 0.4405594405594406,
+      "grad_norm": 1.3606081670938062e-05,
+      "kl": 0.2255859375,
+      "learning_rate": 9.996868349236927e-05,
+      "loss": 0.0002,
+      "num_tokens": 3209875.0,
+      "reward": 2.671295404434204,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3571428656578064,
+      "rewards/check_perplexity_diff_func": 1.8571428060531616,
+      "rewards/check_winston_local_func": 0.45700937509536743,
+      "step": 315
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.44195804195804195,
+      "grad_norm": 1.3537787639105168e-05,
+      "kl": 0.2255859375,
+      "learning_rate": 9.996648674145583e-05,
+      "loss": 0.0002,
+      "step": 316
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 126.21429443359375,
+      "epoch": 0.4433566433566434,
+      "grad_norm": 0.00991208286904361,
+      "kl": 0.15625,
+      "learning_rate": 9.996421555824558e-05,
+      "loss": 0.0103,
+      "num_tokens": 3232963.0,
+      "reward": 2.6654393672943115,
+      "reward_std": 0.0725974440574646,
+      "rewards/check_gptzero_func": 0.5357142686843872,
+      "rewards/check_perplexity_diff_func": 1.6071428060531616,
+      "rewards/check_winston_local_func": 0.5225821733474731,
+      "step": 317
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.44475524475524475,
+      "grad_norm": 0.010677817125212348,
+      "kl": 0.154296875,
+      "learning_rate": 9.996186994612176e-05,
+      "loss": 0.0102,
+      "step": 318
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 103.4464340209961,
+      "epoch": 0.4461538461538462,
+      "grad_norm": 5.275581365835224e-05,
+      "kl": 0.16796875,
+      "learning_rate": 9.995944990857849e-05,
+      "loss": 0.0002,
+      "num_tokens": 3252898.0,
+      "reward": 3.0420854091644287,
+      "reward_std": 0.00039901022682897747,
+      "rewards/check_gptzero_func": 0.5714285969734192,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.5420854687690735,
+      "step": 319
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.44755244755244755,
+      "grad_norm": 6.10762506510971e-05,
+      "kl": 0.16796875,
+      "learning_rate": 9.995695544922077e-05,
+      "loss": 0.0002,
+      "step": 320
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 102.28572082519531,
+      "epoch": 0.4489510489510489,
+      "grad_norm": 0.010696195320132886,
+      "kl": 0.1669921875,
+      "learning_rate": 9.995438657176448e-05,
+      "loss": -0.0177,
+      "num_tokens": 3272574.0,
+      "reward": 2.7592146396636963,
+      "reward_std": 0.03917063772678375,
+      "rewards/check_gptzero_func": 0.3571428656578064,
+      "rewards/check_perplexity_diff_func": 1.9642857313156128,
+      "rewards/check_winston_local_func": 0.437785804271698,
+      "step": 321
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.45034965034965035,
+      "grad_norm": 0.01083841197384593,
+      "kl": 0.1669921875,
+      "learning_rate": 9.995174328003631e-05,
+      "loss": -0.0177,
+      "step": 322
+    },
+    {
+      "clip_ratio": 8.473140769638121e-05,
+      "completion_length": 117.41072082519531,
+      "epoch": 0.45174825174825173,
+      "grad_norm": 0.004522641361347185,
+      "kl": 0.1640625,
+      "learning_rate": 9.994902557797382e-05,
+      "loss": -0.0002,
+      "num_tokens": 3294025.0,
+      "reward": 2.92221736907959,
+      "reward_std": 0.025547320023179054,
+      "rewards/check_gptzero_func": 0.5535714030265808,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.36864587664604187,
+      "step": 323
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.45314685314685316,
+      "grad_norm": 0.004737684573095642,
+      "kl": 0.1650390625,
+      "learning_rate": 9.994623346962544e-05,
+      "loss": -0.0002,
+      "step": 324
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 95.26786041259766,
+      "epoch": 0.45454545454545453,
+      "grad_norm": 3.372088632491828e-06,
+      "kl": 0.1796875,
+      "learning_rate": 9.99433669591504e-05,
+      "loss": 0.0002,
+      "num_tokens": 3312902.0,
+      "reward": 2.595499038696289,
+      "reward_std": 4.388691013446078e-05,
+      "rewards/check_gptzero_func": 0.25,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.34549909830093384,
+      "step": 325
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.45594405594405596,
+      "grad_norm": 3.349405120445447e-06,
+      "kl": 0.1796875,
+      "learning_rate": 9.994042605081879e-05,
+      "loss": 0.0002,
+      "step": 326
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 97.80357360839844,
+      "epoch": 0.45734265734265733,
+      "grad_norm": 0.010175912326793437,
+      "kl": 0.177734375,
+      "learning_rate": 9.993741074901153e-05,
+      "loss": 0.0003,
+      "num_tokens": 3331815.0,
+      "reward": 2.496483325958252,
+      "reward_std": 0.04960284009575844,
+      "rewards/check_gptzero_func": 0.2142857164144516,
+      "rewards/check_perplexity_diff_func": 1.9642857313156128,
+      "rewards/check_winston_local_func": 0.31791168451309204,
+      "step": 327
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.45874125874125876,
+      "grad_norm": 0.01017493586702021,
+      "kl": 0.177734375,
+      "learning_rate": 9.993432105822034e-05,
+      "loss": 0.0003,
+      "step": 328
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 96.30357360839844,
+      "epoch": 0.46013986013986014,
+      "grad_norm": 0.0165654292436505,
+      "kl": 0.1572265625,
+      "learning_rate": 9.993115698304774e-05,
+      "loss": 0.0003,
+      "num_tokens": 3351210.0,
+      "reward": 2.8074374198913574,
+      "reward_std": 0.07505974173545837,
+      "rewards/check_gptzero_func": 0.5178571343421936,
+      "rewards/check_perplexity_diff_func": 1.8571428060531616,
+      "rewards/check_winston_local_func": 0.4324372708797455,
+      "step": 329
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.46153846153846156,
+      "grad_norm": 0.014211382809754432,
+      "kl": 0.16015625,
+      "learning_rate": 9.992791852820709e-05,
+      "loss": 0.0003,
+      "step": 330
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 86.1964340209961,
+      "epoch": 0.46293706293706294,
+      "grad_norm": 0.0580909978469744,
+      "kl": 0.2119140625,
+      "learning_rate": 9.992460569852256e-05,
+      "loss": 0.0013,
+      "num_tokens": 3369025.0,
+      "reward": 2.986760377883911,
+      "reward_std": 0.10425443947315216,
+      "rewards/check_gptzero_func": 0.4642857015132904,
+      "rewards/check_perplexity_diff_func": 1.8571428060531616,
+      "rewards/check_winston_local_func": 0.6653318405151367,
+      "step": 331
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.4643356643356643,
+      "grad_norm": 0.0012843504186470114,
+      "kl": 0.2109375,
+      "learning_rate": 9.992121849892904e-05,
+      "loss": 0.0012,
+      "step": 332
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 113.42857360839844,
+      "epoch": 0.46573426573426574,
+      "grad_norm": 0.03054414776989089,
+      "kl": 0.1494140625,
+      "learning_rate": 9.99177569344723e-05,
+      "loss": -0.0,
+      "num_tokens": 3390093.0,
+      "reward": 2.5890934467315674,
+      "reward_std": 0.1023683100938797,
+      "rewards/check_gptzero_func": 0.25,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.41052165627479553,
+      "step": 333
+    },
+    {
+      "clip_ratio": 0.00022784233442507684,
+      "epoch": 0.4671328671328671,
+      "grad_norm": 0.006141110067038759,
+      "kl": 0.150390625,
+      "learning_rate": 9.99142210103088e-05,
+      "loss": -0.0001,
+      "step": 334
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 109.00000762939453,
+      "epoch": 0.46853146853146854,
+      "grad_norm": 0.011702000814979486,
+      "kl": 0.158203125,
+      "learning_rate": 9.991061073170585e-05,
+      "loss": 0.0001,
+      "num_tokens": 3410511.0,
+      "reward": 2.601181745529175,
+      "reward_std": 0.04859553650021553,
+      "rewards/check_gptzero_func": 0.2857142984867096,
+      "rewards/check_perplexity_diff_func": 1.8928571939468384,
+      "rewards/check_winston_local_func": 0.42261001467704773,
+      "step": 335
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.4699300699300699,
+      "grad_norm": 0.004380253375033637,
+      "kl": 0.158203125,
+      "learning_rate": 9.990692610404145e-05,
+      "loss": 0.0001,
+      "step": 336
+    },
+    {
+      "clip_ratio": 0.00015908369095996022,
+      "completion_length": 124.37500762939453,
+      "epoch": 0.47132867132867134,
+      "grad_norm": 0.011635533606345248,
+      "kl": 0.1318359375,
+      "learning_rate": 9.99031671328044e-05,
+      "loss": 0.0002,
+      "num_tokens": 3433250.0,
+      "reward": 2.719287395477295,
+      "reward_std": 0.05448899790644646,
+      "rewards/check_gptzero_func": 0.5,
+      "rewards/check_perplexity_diff_func": 1.75,
+      "rewards/check_winston_local_func": 0.4692873954772949,
+      "step": 337
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.4727272727272727,
+      "grad_norm": 0.009393665715625162,
+      "kl": 0.1318359375,
+      "learning_rate": 9.989933382359422e-05,
+      "loss": 0.0001,
+      "step": 338
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 110.8214340209961,
+      "epoch": 0.47412587412587415,
+      "grad_norm": 6.954019260271828e-07,
+      "kl": 0.15625,
+      "learning_rate": 9.98954261821212e-05,
+      "loss": 0.0002,
+      "num_tokens": 3454202.0,
+      "reward": 2.5842461585998535,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 1.7857142686843872,
+      "rewards/check_winston_local_func": 0.40567442774772644,
+      "step": 339
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.4755244755244755,
+      "grad_norm": 6.417345751201852e-07,
+      "kl": 0.15625,
+      "learning_rate": 9.98914442142063e-05,
+      "loss": 0.0002,
+      "step": 340
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 76.55357360839844,
+      "epoch": 0.47692307692307695,
+      "grad_norm": 0.008322229771779063,
+      "kl": 0.263671875,
+      "learning_rate": 9.988738792578126e-05,
+      "loss": 0.0005,
+      "num_tokens": 3471041.0,
+      "reward": 2.916919231414795,
+      "reward_std": 0.05185602605342865,
+      "rewards/check_gptzero_func": 0.3214285671710968,
+      "rewards/check_perplexity_diff_func": 1.9642857313156128,
+      "rewards/check_winston_local_func": 0.631205141544342,
+      "step": 341
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.4783216783216783,
+      "grad_norm": 0.010116392524351476,
+      "kl": 0.263671875,
+      "learning_rate": 9.988325732288851e-05,
+      "loss": 0.0004,
+      "step": 342
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 122.21429443359375,
+      "epoch": 0.4797202797202797,
+      "grad_norm": 1.4137707959903356e-05,
+      "kl": 0.134765625,
+      "learning_rate": 9.987905241168117e-05,
+      "loss": 0.0001,
+      "num_tokens": 3493483.0,
+      "reward": 2.7157416343688965,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 1.8571428060531616,
+      "rewards/check_winston_local_func": 0.46574172377586365,
+      "step": 343
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.4811188811188811,
+      "grad_norm": 1.3729789968238485e-05,
+      "kl": 0.134765625,
+      "learning_rate": 9.987477319842307e-05,
+      "loss": 0.0001,
+      "step": 344
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 93.53572082519531,
+      "epoch": 0.4825174825174825,
+      "grad_norm": 7.468121696306164e-07,
+      "kl": 0.1845703125,
+      "learning_rate": 9.987041968948869e-05,
+      "loss": 0.0002,
+      "num_tokens": 3512009.0,
+      "reward": 2.5505218505859375,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.1785714328289032,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.44337886571884155,
+      "step": 345
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.48391608391608393,
+      "grad_norm": 7.137709602226676e-07,
+      "kl": 0.1845703125,
+      "learning_rate": 9.986599189136325e-05,
+      "loss": 0.0002,
+      "step": 346
+    },
+    {
+      "clip_ratio": 0.00031959093757905066,
+      "completion_length": 102.58928680419922,
+      "epoch": 0.4853146853146853,
+      "grad_norm": 0.04707770778424934,
+      "kl": 0.1982421875,
+      "learning_rate": 9.986148981064258e-05,
+      "loss": 0.0004,
+      "num_tokens": 3531574.0,
+      "reward": 2.7341363430023193,
+      "reward_std": 0.0231277234852314,
+      "rewards/check_gptzero_func": 0.4107142984867096,
+      "rewards/check_perplexity_diff_func": 1.8928571939468384,
+      "rewards/check_winston_local_func": 0.430564820766449,
+      "step": 347
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.48671328671328673,
+      "grad_norm": 0.020919799034125892,
+      "kl": 0.2060546875,
+      "learning_rate": 9.985691345403316e-05,
+      "loss": 0.0002,
+      "step": 348
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 127.67857360839844,
+      "epoch": 0.4881118881118881,
+      "grad_norm": 6.517786151476589e-05,
+      "kl": 0.1328125,
+      "learning_rate": 9.985226282835216e-05,
+      "loss": 0.0001,
+      "num_tokens": 3554440.0,
+      "reward": 2.926541566848755,
+      "reward_std": 0.0005768488626927137,
+      "rewards/check_gptzero_func": 0.5357142686843872,
+      "rewards/check_perplexity_diff_func": 1.8571428060531616,
+      "rewards/check_winston_local_func": 0.5336844325065613,
+      "step": 349
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.48951048951048953,
+      "grad_norm": 6.370254129427888e-05,
+      "kl": 0.1328125,
+      "learning_rate": 9.984753794052735e-05,
+      "loss": 0.0001,
+      "step": 350
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 98.75000762939453,
+      "epoch": 0.4909090909090909,
+      "grad_norm": 2.085540464425244e-05,
+      "kl": 0.205078125,
+      "learning_rate": 9.984273879759713e-05,
+      "loss": 0.0002,
+      "num_tokens": 3573976.0,
+      "reward": 2.8204286098480225,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.4989997446537018,
+      "step": 351
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.49230769230769234,
+      "grad_norm": 2.009578048518353e-05,
+      "kl": 0.205078125,
+      "learning_rate": 9.983786540671051e-05,
+      "loss": 0.0002,
+      "step": 352
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 85.21428680419922,
+      "epoch": 0.4937062937062937,
+      "grad_norm": 2.1499404490472988e-06,
+      "kl": 0.1826171875,
+      "learning_rate": 9.983291777512711e-05,
+      "loss": 0.0002,
+      "num_tokens": 3591644.0,
+      "reward": 2.8764142990112305,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.44784265756607056,
+      "step": 353
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.4951048951048951,
+      "grad_norm": 2.160124410750654e-06,
+      "kl": 0.1826171875,
+      "learning_rate": 9.982789591021715e-05,
+      "loss": 0.0002,
+      "step": 354
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 112.00000762939453,
+      "epoch": 0.4965034965034965,
+      "grad_norm": 0.00014814082463704463,
+      "kl": 0.1669921875,
+      "learning_rate": 9.982279981946143e-05,
+      "loss": 0.0002,
+      "num_tokens": 3612596.0,
+      "reward": 2.8398191928863525,
+      "reward_std": 0.0006443771999329329,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.5183902382850647,
+      "step": 355
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.4979020979020979,
+      "grad_norm": 0.00014606763289714295,
+      "kl": 0.1669921875,
+      "learning_rate": 9.98176295104513e-05,
+      "loss": 0.0002,
+      "step": 356
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 88.50000762939453,
+      "epoch": 0.4993006993006993,
+      "grad_norm": 3.33550016470454e-07,
+      "kl": 0.1669921875,
+      "learning_rate": 9.98123849908887e-05,
+      "loss": 0.0002,
+      "num_tokens": 3630862.0,
+      "reward": 2.655492067337036,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3214285671710968,
+      "rewards/check_perplexity_diff_func": 1.8571428060531616,
+      "rewards/check_winston_local_func": 0.47692039608955383,
+      "step": 357
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5006993006993007,
+      "grad_norm": 3.3462192763241657e-07,
+      "kl": 0.1669921875,
+      "learning_rate": 9.980706626858607e-05,
+      "loss": 0.0002,
+      "step": 358
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 126.39286041259766,
+      "epoch": 0.5020979020979021,
+      "grad_norm": 3.4860786526471938e-06,
+      "kl": 0.1416015625,
+      "learning_rate": 9.980167335146642e-05,
+      "loss": 0.0001,
+      "num_tokens": 3653738.0,
+      "reward": 2.765826463699341,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.5,
+      "rewards/check_perplexity_diff_func": 1.8571428060531616,
+      "rewards/check_winston_local_func": 0.4086834490299225,
+      "step": 359
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5034965034965035,
+      "grad_norm": 3.71619859471226e-06,
+      "kl": 0.1416015625,
+      "learning_rate": 9.979620624756329e-05,
+      "loss": 0.0001,
+      "step": 360
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 129.92857360839844,
+      "epoch": 0.5048951048951049,
+      "grad_norm": 8.195642286342138e-07,
+      "kl": 0.1376953125,
+      "learning_rate": 9.979066496502074e-05,
+      "loss": 0.0001,
+      "num_tokens": 3677274.0,
+      "reward": 2.8696835041046143,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.5357142686843872,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.4053979218006134,
+      "step": 361
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5062937062937063,
+      "grad_norm": 8.760444459307852e-07,
+      "kl": 0.1376953125,
+      "learning_rate": 9.978504951209327e-05,
+      "loss": 0.0001,
+      "step": 362
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 118.0714340209961,
+      "epoch": 0.5076923076923077,
+      "grad_norm": 1.595860947799347e-06,
+      "kl": 0.162109375,
+      "learning_rate": 9.977935989714595e-05,
+      "loss": 0.0002,
+      "num_tokens": 3698896.0,
+      "reward": 2.6664552688598633,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.2857142984867096,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.45216941833496094,
+      "step": 363
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.509090909090909,
+      "grad_norm": 1.6260194368178069e-06,
+      "kl": 0.162109375,
+      "learning_rate": 9.977359612865423e-05,
+      "loss": 0.0002,
+      "step": 364
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 85.96428680419922,
+      "epoch": 0.5104895104895105,
+      "grad_norm": 4.357301744161908e-06,
+      "kl": 0.2138671875,
+      "learning_rate": 9.976775821520412e-05,
+      "loss": 0.0002,
+      "num_tokens": 3716536.0,
+      "reward": 2.8877956867218018,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.566366970539093,
+      "step": 365
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5118881118881119,
+      "grad_norm": 4.353902612581284e-06,
+      "kl": 0.2138671875,
+      "learning_rate": 9.976184616549203e-05,
+      "loss": 0.0002,
+      "step": 366
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 116.96429443359375,
+      "epoch": 0.5132867132867133,
+      "grad_norm": 1.5148419096170459e-06,
+      "kl": 0.146484375,
+      "learning_rate": 9.97558599883248e-05,
+      "loss": 0.0001,
+      "num_tokens": 3737950.0,
+      "reward": 2.687312602996826,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.36588403582572937,
+      "step": 367
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5146853146853146,
+      "grad_norm": 1.4470246412763422e-06,
+      "kl": 0.146484375,
+      "learning_rate": 9.97497996926197e-05,
+      "loss": 0.0001,
+      "step": 368
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 94.83928680419922,
+      "epoch": 0.5160839160839161,
+      "grad_norm": 0.0004647591690002515,
+      "kl": 0.2138671875,
+      "learning_rate": 9.974366528740441e-05,
+      "loss": 0.0002,
+      "num_tokens": 3756935.0,
+      "reward": 2.634523391723633,
+      "reward_std": 0.0005499019753187895,
+      "rewards/check_gptzero_func": 0.2857142984867096,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.4202377200126648,
+      "step": 369
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5174825174825175,
+      "grad_norm": 0.0004885464229234562,
+      "kl": 0.2138671875,
+      "learning_rate": 9.973745678181705e-05,
+      "loss": 0.0002,
+      "step": 370
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 100.17857360839844,
+      "epoch": 0.5188811188811189,
+      "grad_norm": 3.0418176986251483e-06,
+      "kl": 0.1640625,
+      "learning_rate": 9.973117418510605e-05,
+      "loss": 0.0002,
+      "num_tokens": 3776087.0,
+      "reward": 2.5216457843780518,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.2857142984867096,
+      "rewards/check_perplexity_diff_func": 1.8571428060531616,
+      "rewards/check_winston_local_func": 0.3787885308265686,
+      "step": 371
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5202797202797202,
+      "grad_norm": 3.0723635480561103e-06,
+      "kl": 0.1640625,
+      "learning_rate": 9.972481750663026e-05,
+      "loss": 0.0002,
+      "step": 372
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 89.66072082519531,
+      "epoch": 0.5216783216783217,
+      "grad_norm": 0.008590656277289149,
+      "kl": 0.40234375,
+      "learning_rate": 9.971838675585888e-05,
+      "loss": 0.0007,
+      "num_tokens": 3794550.0,
+      "reward": 2.5267839431762695,
+      "reward_std": 0.05061079189181328,
+      "rewards/check_gptzero_func": 0.3214285671710968,
+      "rewards/check_perplexity_diff_func": 1.8214285373687744,
+      "rewards/check_winston_local_func": 0.383926659822464,
+      "step": 373
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5230769230769231,
+      "grad_norm": 0.003770576536622864,
+      "kl": 0.4296875,
+      "learning_rate": 9.97118819423714e-05,
+      "loss": 0.0007,
+      "step": 374
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 80.17857360839844,
+      "epoch": 0.5244755244755245,
+      "grad_norm": 3.560813902520289e-07,
+      "kl": 0.2216796875,
+      "learning_rate": 9.970530307585773e-05,
+      "loss": 0.0002,
+      "num_tokens": 3811756.0,
+      "reward": 2.883481740951538,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.5263389348983765,
+      "step": 375
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5258741258741259,
+      "grad_norm": 3.5463275242856317e-07,
+      "kl": 0.2216796875,
+      "learning_rate": 9.9698650166118e-05,
+      "loss": 0.0002,
+      "step": 376
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 86.75000762939453,
+      "epoch": 0.5272727272727272,
+      "grad_norm": 4.776682547251814e-07,
+      "kl": 0.205078125,
+      "learning_rate": 9.969192322306271e-05,
+      "loss": 0.0002,
+      "num_tokens": 3829616.0,
+      "reward": 2.9529268741607666,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4642857015132904,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.5600695610046387,
+      "step": 377
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5286713286713287,
+      "grad_norm": 4.804658823710114e-07,
+      "kl": 0.205078125,
+      "learning_rate": 9.96851222567126e-05,
+      "loss": 0.0002,
+      "step": 378
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 117.89286041259766,
+      "epoch": 0.5300699300699301,
+      "grad_norm": 9.143102926453055e-06,
+      "kl": 0.158203125,
+      "learning_rate": 9.96782472771987e-05,
+      "loss": 0.0002,
+      "num_tokens": 3851170.0,
+      "reward": 2.7617037296295166,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.40456071496009827,
+      "step": 379
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5314685314685315,
+      "grad_norm": 9.41817239216292e-06,
+      "kl": 0.158203125,
+      "learning_rate": 9.967129829476228e-05,
+      "loss": 0.0002,
+      "step": 380
+    },
+    {
+      "clip_ratio": 0.000286861730273813,
+      "completion_length": 80.8214340209961,
+      "epoch": 0.5328671328671328,
+      "grad_norm": 0.03337399461775748,
+      "kl": 0.224609375,
+      "learning_rate": 9.966427531975483e-05,
+      "loss": -0.0402,
+      "num_tokens": 3868508.0,
+      "reward": 2.820319652557373,
+      "reward_std": 0.0946396142244339,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 1.8571428060531616,
+      "rewards/check_winston_local_func": 0.5346053838729858,
+      "step": 381
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5342657342657343,
+      "grad_norm": 0.00034351439093039966,
+      "kl": 0.224609375,
+      "learning_rate": 9.965717836263812e-05,
+      "loss": -0.0403,
+      "step": 382
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 95.03572082519531,
+      "epoch": 0.5356643356643357,
+      "grad_norm": 3.3945579906577137e-07,
+      "kl": 0.216796875,
+      "learning_rate": 9.965000743398408e-05,
+      "loss": 0.0002,
+      "num_tokens": 3887044.0,
+      "reward": 2.8529021739959717,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.4243304431438446,
+      "step": 383
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5370629370629371,
+      "grad_norm": 2.8390305863820515e-07,
+      "kl": 0.216796875,
+      "learning_rate": 9.964276254447484e-05,
+      "loss": 0.0002,
+      "step": 384
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 140.1428680419922,
+      "epoch": 0.5384615384615384,
+      "grad_norm": 8.521451122880038e-07,
+      "kl": 0.134765625,
+      "learning_rate": 9.96354437049027e-05,
+      "loss": 0.0001,
+      "num_tokens": 3911264.0,
+      "reward": 2.4770619869232178,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.2857142984867096,
+      "rewards/check_perplexity_diff_func": 1.8571428060531616,
+      "rewards/check_winston_local_func": 0.33420488238334656,
+      "step": 385
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5398601398601398,
+      "grad_norm": 6.753306754213034e-07,
+      "kl": 0.134765625,
+      "learning_rate": 9.962805092617016e-05,
+      "loss": 0.0001,
+      "step": 386
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 97.64286041259766,
+      "epoch": 0.5412587412587413,
+      "grad_norm": 0.08784199050747857,
+      "kl": 2.28125,
+      "learning_rate": 9.962058421928979e-05,
+      "loss": 0.0023,
+      "num_tokens": 3930478.0,
+      "reward": 2.897810459136963,
+      "reward_std": 0.002231778111308813,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.4692386984825134,
+      "step": 387
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5426573426573427,
+      "grad_norm": 0.03620083942117924,
+      "kl": 0.98828125,
+      "learning_rate": 9.961304359538437e-05,
+      "loss": 0.001,
+      "step": 388
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 102.21428680419922,
+      "epoch": 0.544055944055944,
+      "grad_norm": 0.025447754609912356,
+      "kl": 0.1806640625,
+      "learning_rate": 9.96054290656867e-05,
+      "loss": 0.0002,
+      "num_tokens": 3950362.0,
+      "reward": 2.8110532760620117,
+      "reward_std": 0.04895726963877678,
+      "rewards/check_gptzero_func": 0.5357142686843872,
+      "rewards/check_perplexity_diff_func": 1.8214285373687744,
+      "rewards/check_winston_local_func": 0.45391014218330383,
+      "step": 389
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5454545454545454,
+      "grad_norm": 5.1964811610508515e-05,
+      "kl": 0.1806640625,
+      "learning_rate": 9.959774064153977e-05,
+      "loss": 0.0001,
+      "step": 390
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 103.5714340209961,
+      "epoch": 0.5468531468531469,
+      "grad_norm": 3.816406536402076e-07,
+      "kl": 0.173828125,
+      "learning_rate": 9.958997833439657e-05,
+      "loss": 0.0002,
+      "num_tokens": 3970164.0,
+      "reward": 2.5441431999206543,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3214285671710968,
+      "rewards/check_perplexity_diff_func": 1.8571428060531616,
+      "rewards/check_winston_local_func": 0.3655718266963959,
+      "step": 391
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5482517482517483,
+      "grad_norm": 4.199010025491445e-07,
+      "kl": 0.173828125,
+      "learning_rate": 9.958214215582018e-05,
+      "loss": 0.0002,
+      "step": 392
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 126.8214340209961,
+      "epoch": 0.5496503496503496,
+      "grad_norm": 6.309179959293235e-06,
+      "kl": 0.1533203125,
+      "learning_rate": 9.957423211748374e-05,
+      "loss": 0.0002,
+      "num_tokens": 3992994.0,
+      "reward": 2.7432734966278076,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.42184481024742126,
+      "step": 393
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.551048951048951,
+      "grad_norm": 5.547540566387994e-06,
+      "kl": 0.1533203125,
+      "learning_rate": 9.956624823117036e-05,
+      "loss": 0.0002,
+      "step": 394
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 125.62500762939453,
+      "epoch": 0.5524475524475524,
+      "grad_norm": 0.007987942179452279,
+      "kl": 0.142578125,
+      "learning_rate": 9.955819050877321e-05,
+      "loss": 0.0001,
+      "num_tokens": 4015669.0,
+      "reward": 3.091567039489746,
+      "reward_std": 0.018998777493834496,
+      "rewards/check_gptzero_func": 0.5535714030265808,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.6094242334365845,
+      "step": 395
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5538461538461539,
+      "grad_norm": 0.0034758785486018436,
+      "kl": 0.142578125,
+      "learning_rate": 9.955005896229543e-05,
+      "loss": 0.0,
+      "step": 396
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 96.75000762939453,
+      "epoch": 0.5552447552447553,
+      "grad_norm": 0.011495641299277456,
+      "kl": 0.283203125,
+      "learning_rate": 9.954185360385013e-05,
+      "loss": 0.0006,
+      "num_tokens": 4035041.0,
+      "reward": 2.950004816055298,
+      "reward_std": 0.01496803853660822,
+      "rewards/check_gptzero_func": 0.4821428656578064,
+      "rewards/check_perplexity_diff_func": 1.8928571939468384,
+      "rewards/check_winston_local_func": 0.5750047564506531,
+      "step": 397
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5566433566433566,
+      "grad_norm": 0.010082500068390739,
+      "kl": 0.283203125,
+      "learning_rate": 9.953357444566039e-05,
+      "loss": 0.0006,
+      "step": 398
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 110.8214340209961,
+      "epoch": 0.558041958041958,
+      "grad_norm": 8.489376613416e-06,
+      "kl": 0.1787109375,
+      "learning_rate": 9.952522150005919e-05,
+      "loss": 0.0002,
+      "num_tokens": 4055855.0,
+      "reward": 2.913240909576416,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4642857015132904,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.5203836560249329,
+      "step": 399
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5594405594405595,
+      "grad_norm": 8.766485288799225e-06,
+      "kl": 0.1787109375,
+      "learning_rate": 9.951679477948947e-05,
+      "loss": 0.0002,
+      "step": 400
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 105.46428680419922,
+      "epoch": 0.5608391608391609,
+      "grad_norm": 8.652849000461684e-06,
+      "kl": 0.150390625,
+      "learning_rate": 9.9508294296504e-05,
+      "loss": 0.0002,
+      "num_tokens": 4076051.0,
+      "reward": 2.871354818344116,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.44278329610824585,
+      "step": 401
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5622377622377622,
+      "grad_norm": 8.18246821524729e-06,
+      "kl": 0.150390625,
+      "learning_rate": 9.949972006376556e-05,
+      "loss": 0.0002,
+      "step": 402
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 75.25,
+      "epoch": 0.5636363636363636,
+      "grad_norm": 7.351342876079361e-07,
+      "kl": 0.1953125,
+      "learning_rate": 9.949107209404665e-05,
+      "loss": 0.0002,
+      "num_tokens": 4092527.0,
+      "reward": 2.9322519302368164,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3571428656578064,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.5751089453697205,
+      "step": 403
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5650349650349651,
+      "grad_norm": 7.136344464618312e-07,
+      "kl": 0.1953125,
+      "learning_rate": 9.948235040022966e-05,
+      "loss": 0.0002,
+      "step": 404
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 108.08928680419922,
+      "epoch": 0.5664335664335665,
+      "grad_norm": 0.007956878335048736,
+      "kl": 0.169921875,
+      "learning_rate": 9.947355499530683e-05,
+      "loss": 0.0049,
+      "num_tokens": 4113282.0,
+      "reward": 3.1798436641693115,
+      "reward_std": 0.06495096534490585,
+      "rewards/check_gptzero_func": 0.6785714030265808,
+      "rewards/check_perplexity_diff_func": 1.9642857313156128,
+      "rewards/check_winston_local_func": 0.5369864702224731,
+      "step": 405
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5678321678321678,
+      "grad_norm": 0.006640329404323574,
+      "kl": 0.1728515625,
+      "learning_rate": 9.946468589238021e-05,
+      "loss": 0.0049,
+      "step": 406
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 121.39286041259766,
+      "epoch": 0.5692307692307692,
+      "grad_norm": 1.5620797981026132e-06,
+      "kl": 0.1396484375,
+      "learning_rate": 9.945574310466159e-05,
+      "loss": 0.0001,
+      "num_tokens": 4135506.0,
+      "reward": 2.573127269744873,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.2857142984867096,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.3588416278362274,
+      "step": 407
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5706293706293706,
+      "grad_norm": 1.51593967204227e-06,
+      "kl": 0.1396484375,
+      "learning_rate": 9.944672664547252e-05,
+      "loss": 0.0001,
+      "step": 408
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 107.8214340209961,
+      "epoch": 0.5720279720279721,
+      "grad_norm": 0.0007775272826913423,
+      "kl": 0.17578125,
+      "learning_rate": 9.943763652824436e-05,
+      "loss": 0.0002,
+      "num_tokens": 4156060.0,
+      "reward": 2.55350661277771,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3571428656578064,
+      "rewards/check_perplexity_diff_func": 1.7142857313156128,
+      "rewards/check_winston_local_func": 0.48207801580429077,
+      "step": 409
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5734265734265734,
+      "grad_norm": 0.0008914625567854122,
+      "kl": 0.1767578125,
+      "learning_rate": 9.942847276651811e-05,
+      "loss": 0.0002,
+      "step": 410
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 101.62500762939453,
+      "epoch": 0.5748251748251748,
+      "grad_norm": 4.9539161428254635e-05,
+      "kl": 0.1474609375,
+      "learning_rate": 9.941923537394456e-05,
+      "loss": 0.0001,
+      "num_tokens": 4176019.0,
+      "reward": 2.8906424045562744,
+      "reward_std": 0.00012528452498372644,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.5334992408752441,
+      "step": 411
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5762237762237762,
+      "grad_norm": 5.844422779033556e-05,
+      "kl": 0.1474609375,
+      "learning_rate": 9.940992436428409e-05,
+      "loss": 0.0001,
+      "step": 412
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 105.28572082519531,
+      "epoch": 0.5776223776223777,
+      "grad_norm": 0.0006008878616346907,
+      "kl": 0.2216796875,
+      "learning_rate": 9.940053975140684e-05,
+      "loss": 0.0002,
+      "num_tokens": 4196667.0,
+      "reward": 2.943253755569458,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.5357142686843872,
+      "rewards/check_perplexity_diff_func": 1.8571428060531616,
+      "rewards/check_winston_local_func": 0.5503963828086853,
+      "step": 413
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.579020979020979,
+      "grad_norm": 0.0003365492175069851,
+      "kl": 0.220703125,
+      "learning_rate": 9.939108154929252e-05,
+      "loss": 0.0002,
+      "step": 414
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 114.60714721679688,
+      "epoch": 0.5804195804195804,
+      "grad_norm": 0.008680207230787525,
+      "kl": 0.171875,
+      "learning_rate": 9.938154977203049e-05,
+      "loss": -0.0005,
+      "num_tokens": 4217727.0,
+      "reward": 2.7773444652557373,
+      "reward_std": 0.051937274634838104,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 1.8214285373687744,
+      "rewards/check_winston_local_func": 0.5273441076278687,
+      "step": 415
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5818181818181818,
+      "grad_norm": 0.00033864647249595475,
+      "kl": 0.171875,
+      "learning_rate": 9.937194443381972e-05,
+      "loss": -0.0005,
+      "step": 416
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 100.60714721679688,
+      "epoch": 0.5832167832167832,
+      "grad_norm": 4.017405087628847e-06,
+      "kl": 0.193359375,
+      "learning_rate": 9.936226554896875e-05,
+      "loss": 0.0002,
+      "num_tokens": 4237189.0,
+      "reward": 2.7684807777404785,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.44705215096473694,
+      "step": 417
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5846153846153846,
+      "grad_norm": 3.880139616189675e-06,
+      "kl": 0.193359375,
+      "learning_rate": 9.935251313189564e-05,
+      "loss": 0.0002,
+      "step": 418
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 107.75000762939453,
+      "epoch": 0.586013986013986,
+      "grad_norm": 0.00032680437267053027,
+      "kl": 0.1826171875,
+      "learning_rate": 9.934268719712807e-05,
+      "loss": 0.0002,
+      "num_tokens": 4258213.0,
+      "reward": 2.577641487121582,
+      "reward_std": 0.0015505586052313447,
+      "rewards/check_gptzero_func": 0.2857142984867096,
+      "rewards/check_perplexity_diff_func": 1.8571428060531616,
+      "rewards/check_winston_local_func": 0.4347843825817108,
+      "step": 419
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5874125874125874,
+      "grad_norm": 0.0003238465648274268,
+      "kl": 0.1826171875,
+      "learning_rate": 9.933278775930317e-05,
+      "loss": 0.0002,
+      "step": 420
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 108.67857360839844,
+      "epoch": 0.5888111888111888,
+      "grad_norm": 0.015591092439371165,
+      "kl": 0.16015625,
+      "learning_rate": 9.932281483316758e-05,
+      "loss": 0.0004,
+      "num_tokens": 4278847.0,
+      "reward": 2.615447998046875,
+      "reward_std": 0.05428782477974892,
+      "rewards/check_gptzero_func": 0.3214285671710968,
+      "rewards/check_perplexity_diff_func": 1.8214285373687744,
+      "rewards/check_winston_local_func": 0.47259071469306946,
+      "step": 421
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5902097902097903,
+      "grad_norm": 0.00016911180587449477,
+      "kl": 0.16015625,
+      "learning_rate": 9.931276843357742e-05,
+      "loss": 0.0003,
+      "step": 422
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 111.16072082519531,
+      "epoch": 0.5916083916083916,
+      "grad_norm": 1.6847767080478394e-05,
+      "kl": 0.1630859375,
+      "learning_rate": 9.930264857549825e-05,
+      "loss": 0.0002,
+      "num_tokens": 4300096.0,
+      "reward": 2.857067823410034,
+      "reward_std": 0.00101040443405509,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 1.8571428060531616,
+      "rewards/check_winston_local_func": 0.6070676445960999,
+      "step": 423
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.593006993006993,
+      "grad_norm": 2.453025078669727e-05,
+      "kl": 0.1630859375,
+      "learning_rate": 9.929245527400503e-05,
+      "loss": 0.0002,
+      "step": 424
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 102.50000762939453,
+      "epoch": 0.5944055944055944,
+      "grad_norm": 0.0015183266942611472,
+      "kl": 0.173828125,
+      "learning_rate": 9.928218854428221e-05,
+      "loss": 0.0002,
+      "num_tokens": 4320064.0,
+      "reward": 2.6430578231811523,
+      "reward_std": 5.020291791879572e-05,
+      "rewards/check_gptzero_func": 0.3571428656578064,
+      "rewards/check_perplexity_diff_func": 1.7857142686843872,
+      "rewards/check_winston_local_func": 0.5002006888389587,
+      "step": 425
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5958041958041959,
+      "grad_norm": 0.0013596061771866538,
+      "kl": 0.1708984375,
+      "learning_rate": 9.927184840162354e-05,
+      "loss": 0.0002,
+      "step": 426
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 110.66072082519531,
+      "epoch": 0.5972027972027972,
+      "grad_norm": 0.12076121626694686,
+      "kl": 0.189453125,
+      "learning_rate": 9.926143486143214e-05,
+      "loss": -0.001,
+      "num_tokens": 4341145.0,
+      "reward": 2.6579177379608154,
+      "reward_std": 0.05154726281762123,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 1.8214285373687744,
+      "rewards/check_winston_local_func": 0.4079175889492035,
+      "step": 427
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5986013986013986,
+      "grad_norm": 0.0008289218843548021,
+      "kl": 0.19140625,
+      "learning_rate": 9.92509479392205e-05,
+      "loss": -0.0013,
+      "step": 428
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 89.67857360839844,
+      "epoch": 0.6,
+      "grad_norm": 0.005379412774279031,
+      "kl": 0.1787109375,
+      "learning_rate": 9.924038765061042e-05,
+      "loss": 0.0001,
+      "num_tokens": 4359277.0,
+      "reward": 2.633777141571045,
+      "reward_std": 0.05039219558238983,
+      "rewards/check_gptzero_func": 0.2857142984867096,
+      "rewards/check_perplexity_diff_func": 1.9642857313156128,
+      "rewards/check_winston_local_func": 0.38377735018730164,
+      "step": 429
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6013986013986014,
+      "grad_norm": 0.010439668945298157,
+      "kl": 0.1787109375,
+      "learning_rate": 9.922975401133293e-05,
+      "loss": 0.0002,
+      "step": 430
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 102.67857360839844,
+      "epoch": 0.6027972027972028,
+      "grad_norm": 0.06317955454404899,
+      "kl": 0.201171875,
+      "learning_rate": 9.92190470372284e-05,
+      "loss": 0.0068,
+      "num_tokens": 4379109.0,
+      "reward": 2.6650938987731934,
+      "reward_std": 0.05930115655064583,
+      "rewards/check_gptzero_func": 0.25,
+      "rewards/check_perplexity_diff_func": 1.9642857313156128,
+      "rewards/check_winston_local_func": 0.4508081078529358,
+      "step": 431
+    },
+    {
+      "clip_ratio": 0.0002587991766631603,
+      "epoch": 0.6041958041958042,
+      "grad_norm": 0.05122842630547611,
+      "kl": 0.1865234375,
+      "learning_rate": 9.920826674424642e-05,
+      "loss": 0.0063,
+      "step": 432
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 80.35714721679688,
+      "epoch": 0.6055944055944056,
+      "grad_norm": 7.162658137250354e-07,
+      "kl": 0.1875,
+      "learning_rate": 9.919741314844577e-05,
+      "loss": 0.0002,
+      "num_tokens": 4396693.0,
+      "reward": 3.0637807846069336,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.5357142686843872,
+      "rewards/check_perplexity_diff_func": 1.8571428060531616,
+      "rewards/check_winston_local_func": 0.6709235310554504,
+      "step": 433
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.606993006993007,
+      "grad_norm": 8.451796450955314e-07,
+      "kl": 0.1875,
+      "learning_rate": 9.918648626599447e-05,
+      "loss": 0.0002,
+      "step": 434
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 124.98214721679688,
+      "epoch": 0.6083916083916084,
+      "grad_norm": 0.016099687268680437,
+      "kl": 0.1455078125,
+      "learning_rate": 9.91754861131697e-05,
+      "loss": -0.0001,
+      "num_tokens": 4418950.0,
+      "reward": 2.94807505607605,
+      "reward_std": 0.050821539014577866,
+      "rewards/check_gptzero_func": 0.5357142686843872,
+      "rewards/check_perplexity_diff_func": 1.8928571939468384,
+      "rewards/check_winston_local_func": 0.5195035338401794,
+      "step": 435
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6097902097902098,
+      "grad_norm": 0.014390989758817311,
+      "kl": 0.1455078125,
+      "learning_rate": 9.916441270635772e-05,
+      "loss": -0.0001,
+      "step": 436
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 106.64286041259766,
+      "epoch": 0.6111888111888112,
+      "grad_norm": 1.968332582426962e-06,
+      "kl": 0.14453125,
+      "learning_rate": 9.915326606205404e-05,
+      "loss": 0.0001,
+      "num_tokens": 4439018.0,
+      "reward": 2.77024507522583,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.4131018817424774,
+      "step": 437
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6125874125874126,
+      "grad_norm": 1.8961430119120163e-06,
+      "kl": 0.14453125,
+      "learning_rate": 9.914204619686314e-05,
+      "loss": 0.0001,
+      "step": 438
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 102.21428680419922,
+      "epoch": 0.6139860139860139,
+      "grad_norm": 0.000668351376631818,
+      "kl": 0.1728515625,
+      "learning_rate": 9.913075312749866e-05,
+      "loss": 0.0002,
+      "num_tokens": 4458806.0,
+      "reward": 2.8182172775268555,
+      "reward_std": 0.0025242711417376995,
+      "rewards/check_gptzero_func": 0.5357142686843872,
+      "rewards/check_perplexity_diff_func": 1.8571428060531616,
+      "rewards/check_winston_local_func": 0.42535990476608276,
+      "step": 439
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6153846153846154,
+      "grad_norm": 0.0006751756209535994,
+      "kl": 0.1728515625,
+      "learning_rate": 9.911938687078324e-05,
+      "loss": 0.0002,
+      "step": 440
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 81.25,
+      "epoch": 0.6167832167832168,
+      "grad_norm": 0.007904133813646444,
+      "kl": 0.240234375,
+      "learning_rate": 9.910794744364857e-05,
+      "loss": -0.0001,
+      "num_tokens": 4475982.0,
+      "reward": 2.8269970417022705,
+      "reward_std": 0.05014092102646828,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 1.8928571939468384,
+      "rewards/check_winston_local_func": 0.5412828326225281,
+      "step": 441
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6181818181818182,
+      "grad_norm": 0.00827597712943806,
+      "kl": 0.240234375,
+      "learning_rate": 9.909643486313533e-05,
+      "loss": -0.0001,
+      "step": 442
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 113.00000762939453,
+      "epoch": 0.6195804195804195,
+      "grad_norm": 6.016032313066646e-06,
+      "kl": 0.16796875,
+      "learning_rate": 9.908484914639318e-05,
+      "loss": 0.0002,
+      "num_tokens": 4497200.0,
+      "reward": 2.9243876934051514,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.5315303206443787,
+      "step": 443
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.620979020979021,
+      "grad_norm": 7.198804696892823e-06,
+      "kl": 0.16796875,
+      "learning_rate": 9.90731903106807e-05,
+      "loss": 0.0002,
+      "step": 444
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 129.375,
+      "epoch": 0.6223776223776224,
+      "grad_norm": 0.02075011817403209,
+      "kl": 0.158203125,
+      "learning_rate": 9.90614583733654e-05,
+      "loss": -0.0001,
+      "num_tokens": 4519807.0,
+      "reward": 2.866011619567871,
+      "reward_std": 0.10101933032274246,
+      "rewards/check_gptzero_func": 0.4642857015132904,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.47315436601638794,
+      "step": 445
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6237762237762238,
+      "grad_norm": 0.0034619333956659956,
+      "kl": 0.1591796875,
+      "learning_rate": 9.904965335192373e-05,
+      "loss": -0.0002,
+      "step": 446
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 98.92857360839844,
+      "epoch": 0.6251748251748251,
+      "grad_norm": 0.010057352537761423,
+      "kl": 0.212890625,
+      "learning_rate": 9.903777526394094e-05,
+      "loss": -0.0086,
+      "num_tokens": 4539039.0,
+      "reward": 2.980358362197876,
+      "reward_std": 0.05100167542695999,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 1.9642857313156128,
+      "rewards/check_winston_local_func": 0.587501049041748,
+      "step": 447
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6265734265734266,
+      "grad_norm": 0.006958390301115274,
+      "kl": 0.21875,
+      "learning_rate": 9.90258241271112e-05,
+      "loss": -0.0087,
+      "step": 448
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 111.71429443359375,
+      "epoch": 0.627972027972028,
+      "grad_norm": 1.6536770511152556e-06,
+      "kl": 0.142578125,
+      "learning_rate": 9.901379995923738e-05,
+      "loss": 0.0001,
+      "num_tokens": 4559675.0,
+      "reward": 2.681791067123413,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.2857142984867096,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.4675052762031555,
+      "step": 449
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6293706293706294,
+      "grad_norm": 1.6773170531853214e-06,
+      "kl": 0.142578125,
+      "learning_rate": 9.900170277823129e-05,
+      "loss": 0.0001,
+      "step": 450
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 107.35714721679688,
+      "epoch": 0.6307692307692307,
+      "grad_norm": 2.0958023033337787e-06,
+      "kl": 0.2109375,
+      "learning_rate": 9.898953260211338e-05,
+      "loss": 0.0002,
+      "num_tokens": 4579995.0,
+      "reward": 2.7400293350219727,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.4186006188392639,
+      "step": 451
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6321678321678321,
+      "grad_norm": 1.982971590769392e-06,
+      "kl": 0.2109375,
+      "learning_rate": 9.897728944901292e-05,
+      "loss": 0.0002,
+      "step": 452
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 118.26786041259766,
+      "epoch": 0.6335664335664336,
+      "grad_norm": 0.009450375112303724,
+      "kl": 0.1416015625,
+      "learning_rate": 9.896497333716783e-05,
+      "loss": -0.0051,
+      "num_tokens": 4602048.0,
+      "reward": 2.640944004058838,
+      "reward_std": 0.043768420815467834,
+      "rewards/check_gptzero_func": 0.5357142686843872,
+      "rewards/check_perplexity_diff_func": 1.5357142686843872,
+      "rewards/check_winston_local_func": 0.5695151090621948,
+      "step": 453
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.634965034965035,
+      "grad_norm": 0.0017713963386990425,
+      "kl": 0.1416015625,
+      "learning_rate": 9.895258428492475e-05,
+      "loss": -0.0051,
+      "step": 454
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 98.42857360839844,
+      "epoch": 0.6363636363636364,
+      "grad_norm": 0.014431904681184264,
+      "kl": 0.1943359375,
+      "learning_rate": 9.894012231073894e-05,
+      "loss": 0.0002,
+      "num_tokens": 4621654.0,
+      "reward": 2.7700040340423584,
+      "reward_std": 0.03728308901190758,
+      "rewards/check_gptzero_func": 0.4107142984867096,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.43071839213371277,
+      "step": 455
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6377622377622377,
+      "grad_norm": 0.008239966094546339,
+      "kl": 0.1943359375,
+      "learning_rate": 9.892758743317434e-05,
+      "loss": 0.0002,
+      "step": 456
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 90.50000762939453,
+      "epoch": 0.6391608391608392,
+      "grad_norm": 2.579298856684113e-06,
+      "kl": 0.1923828125,
+      "learning_rate": 9.891497967090344e-05,
+      "loss": 0.0002,
+      "num_tokens": 4640140.0,
+      "reward": 2.6383018493652344,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.25,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.4597306251525879,
+      "step": 457
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6405594405594406,
+      "grad_norm": 2.6625863420088633e-06,
+      "kl": 0.1923828125,
+      "learning_rate": 9.890229904270731e-05,
+      "loss": 0.0002,
+      "step": 458
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 113.03572082519531,
+      "epoch": 0.641958041958042,
+      "grad_norm": 1.7932858894376123e-05,
+      "kl": 0.1748046875,
+      "learning_rate": 9.888954556747563e-05,
+      "loss": 0.0002,
+      "num_tokens": 4661276.0,
+      "reward": 2.74210524559021,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3571428656578064,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.3849623203277588,
+      "step": 459
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6433566433566433,
+      "grad_norm": 1.7905922488331998e-05,
+      "kl": 0.1748046875,
+      "learning_rate": 9.887671926420648e-05,
+      "loss": 0.0002,
+      "step": 460
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 112.42857360839844,
+      "epoch": 0.6447552447552447,
+      "grad_norm": 0.014077582219597583,
+      "kl": 0.173828125,
+      "learning_rate": 9.886382015200652e-05,
+      "loss": 0.0002,
+      "num_tokens": 4682450.0,
+      "reward": 2.8680295944213867,
+      "reward_std": 0.05049533396959305,
+      "rewards/check_gptzero_func": 0.4642857015132904,
+      "rewards/check_perplexity_diff_func": 1.9642857313156128,
+      "rewards/check_winston_local_func": 0.43945807218551636,
+      "step": 461
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6461538461538462,
+      "grad_norm": 0.010796297977350395,
+      "kl": 0.173828125,
+      "learning_rate": 9.885084825009086e-05,
+      "loss": 0.0001,
+      "step": 462
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 105.14286041259766,
+      "epoch": 0.6475524475524476,
+      "grad_norm": 0.02303930816984521,
+      "kl": 0.17578125,
+      "learning_rate": 9.883780357778299e-05,
+      "loss": 0.0003,
+      "num_tokens": 4702710.0,
+      "reward": 2.885434150695801,
+      "reward_std": 0.06019994616508484,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 1.75,
+      "rewards/check_winston_local_func": 0.7068625092506409,
+      "step": 463
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6489510489510489,
+      "grad_norm": 0.0015050223064125838,
+      "kl": 0.1767578125,
+      "learning_rate": 9.882468615451484e-05,
+      "loss": 0.0002,
+      "step": 464
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 127.5714340209961,
+      "epoch": 0.6503496503496503,
+      "grad_norm": 3.0460982652821775e-06,
+      "kl": 0.1650390625,
+      "learning_rate": 9.881149599982671e-05,
+      "loss": 0.0002,
+      "num_tokens": 4725654.0,
+      "reward": 2.407341957092285,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.2857142984867096,
+      "rewards/check_perplexity_diff_func": 1.8571428060531616,
+      "rewards/check_winston_local_func": 0.2644847333431244,
+      "step": 465
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6517482517482518,
+      "grad_norm": 2.8166666383288673e-06,
+      "kl": 0.1650390625,
+      "learning_rate": 9.879823313336722e-05,
+      "loss": 0.0002,
+      "step": 466
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 122.16072082519531,
+      "epoch": 0.6531468531468532,
+      "grad_norm": 0.0001021700279542065,
+      "kl": 0.140625,
+      "learning_rate": 9.878489757489337e-05,
+      "loss": 0.0001,
+      "num_tokens": 4748375.0,
+      "reward": 2.6119117736816406,
+      "reward_std": 2.7804879209725186e-05,
+      "rewards/check_gptzero_func": 0.5357142686843872,
+      "rewards/check_perplexity_diff_func": 1.5714285373687744,
+      "rewards/check_winston_local_func": 0.5047687888145447,
+      "step": 467
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6545454545454545,
+      "grad_norm": 9.33433928584176e-05,
+      "kl": 0.140625,
+      "learning_rate": 9.877148934427037e-05,
+      "loss": 0.0001,
+      "step": 468
+    },
+    {
+      "clip_ratio": 0.0006711409660056233,
+      "completion_length": 95.64286041259766,
+      "epoch": 0.6559440559440559,
+      "grad_norm": 0.0004684591425632982,
+      "kl": 0.171875,
+      "learning_rate": 9.87580084614717e-05,
+      "loss": 0.0002,
+      "num_tokens": 4767439.0,
+      "reward": 2.3873257637023926,
+      "reward_std": 0.0017833748133853078,
+      "rewards/check_gptzero_func": 0.25,
+      "rewards/check_perplexity_diff_func": 1.7142857313156128,
+      "rewards/check_winston_local_func": 0.42303988337516785,
+      "step": 469
+    },
+    {
+      "clip_ratio": 0.0004793864209204912,
+      "epoch": 0.6573426573426573,
+      "grad_norm": 0.00047799981287926325,
+      "kl": 0.171875,
+      "learning_rate": 9.874445494657911e-05,
+      "loss": 0.0002,
+      "step": 470
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 79.39286041259766,
+      "epoch": 0.6587412587412588,
+      "grad_norm": 0.011441113949782547,
+      "kl": 0.216796875,
+      "learning_rate": 9.873082881978251e-05,
+      "loss": -0.0064,
+      "num_tokens": 4784715.0,
+      "reward": 3.037661075592041,
+      "reward_std": 0.023328183218836784,
+      "rewards/check_gptzero_func": 0.4821428656578064,
+      "rewards/check_perplexity_diff_func": 1.9642857313156128,
+      "rewards/check_winston_local_func": 0.5912323594093323,
+      "step": 471
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6601398601398601,
+      "grad_norm": 0.008433986682608557,
+      "kl": 0.216796875,
+      "learning_rate": 9.871713010137997e-05,
+      "loss": -0.0064,
+      "step": 472
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 119.42857360839844,
+      "epoch": 0.6615384615384615,
+      "grad_norm": 7.638914800437301e-06,
+      "kl": 0.1435546875,
+      "learning_rate": 9.870335881177774e-05,
+      "loss": 0.0001,
+      "num_tokens": 4806207.0,
+      "reward": 2.9706830978393555,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4642857015132904,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.5063972473144531,
+      "step": 473
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6629370629370629,
+      "grad_norm": 7.505737589947172e-06,
+      "kl": 0.1435546875,
+      "learning_rate": 9.868951497149011e-05,
+      "loss": 0.0001,
+      "step": 474
+    },
+    {
+      "clip_ratio": 0.00019073051225859672,
+      "completion_length": 89.87500762939453,
+      "epoch": 0.6643356643356644,
+      "grad_norm": 0.03662860311910791,
+      "kl": 0.2060546875,
+      "learning_rate": 9.86755986011395e-05,
+      "loss": 0.0065,
+      "num_tokens": 4824332.0,
+      "reward": 2.643022060394287,
+      "reward_std": 0.04776443541049957,
+      "rewards/check_gptzero_func": 0.3214285671710968,
+      "rewards/check_perplexity_diff_func": 1.8928571939468384,
+      "rewards/check_winston_local_func": 0.42873620986938477,
+      "step": 475
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6657342657342658,
+      "grad_norm": 0.03126231173733144,
+      "kl": 0.2060546875,
+      "learning_rate": 9.866160972145634e-05,
+      "loss": 0.0062,
+      "step": 476
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 123.10714721679688,
+      "epoch": 0.6671328671328671,
+      "grad_norm": 0.0002559996362709361,
+      "kl": 0.1357421875,
+      "learning_rate": 9.864754835327909e-05,
+      "loss": 0.0001,
+      "num_tokens": 4847094.0,
+      "reward": 2.767104387283325,
+      "reward_std": 0.0007265785825438797,
+      "rewards/check_gptzero_func": 0.5,
+      "rewards/check_perplexity_diff_func": 1.7142857313156128,
+      "rewards/check_winston_local_func": 0.5528185963630676,
+      "step": 477
+    },
+    {
+      "clip_ratio": 8.97343925316818e-05,
+      "epoch": 0.6685314685314685,
+      "grad_norm": 0.00042633622080268514,
+      "kl": 0.1357421875,
+      "learning_rate": 9.86334145175542e-05,
+      "loss": 0.0001,
+      "step": 478
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 96.67857360839844,
+      "epoch": 0.66993006993007,
+      "grad_norm": 0.0008693689003033054,
+      "kl": 0.16015625,
+      "learning_rate": 9.861920823533606e-05,
+      "loss": 0.0002,
+      "num_tokens": 4866394.0,
+      "reward": 2.5761330127716064,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3571428656578064,
+      "rewards/check_perplexity_diff_func": 1.7142857313156128,
+      "rewards/check_winston_local_func": 0.504704475402832,
+      "step": 479
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6713286713286714,
+      "grad_norm": 0.0008435837049153919,
+      "kl": 0.16015625,
+      "learning_rate": 9.860492952778696e-05,
+      "loss": 0.0002,
+      "step": 480
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 94.39286041259766,
+      "epoch": 0.6727272727272727,
+      "grad_norm": 3.349433978828943e-06,
+      "kl": 0.177734375,
+      "learning_rate": 9.859057841617709e-05,
+      "loss": 0.0002,
+      "num_tokens": 4885318.0,
+      "reward": 2.8612334728240967,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3571428656578064,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.5755191445350647,
+      "step": 481
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6741258741258741,
+      "grad_norm": 3.5095106255593373e-06,
+      "kl": 0.177734375,
+      "learning_rate": 9.857615492188452e-05,
+      "loss": 0.0002,
+      "step": 482
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 87.98214721679688,
+      "epoch": 0.6755244755244755,
+      "grad_norm": 0.00016582191324414645,
+      "kl": 0.208984375,
+      "learning_rate": 9.856165906639513e-05,
+      "loss": 0.0002,
+      "num_tokens": 4903401.0,
+      "reward": 2.797102928161621,
+      "reward_std": 0.0008326892857439816,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.4756740629673004,
+      "step": 483
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.676923076923077,
+      "grad_norm": 0.00017592290286556547,
+      "kl": 0.208984375,
+      "learning_rate": 9.85470908713026e-05,
+      "loss": 0.0002,
+      "step": 484
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 122.14286041259766,
+      "epoch": 0.6783216783216783,
+      "grad_norm": 3.840262302649335e-05,
+      "kl": 0.140625,
+      "learning_rate": 9.853245035830834e-05,
+      "loss": 0.0001,
+      "num_tokens": 4925777.0,
+      "reward": 2.80380916595459,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4642857015132904,
+      "rewards/check_perplexity_diff_func": 1.8571428060531616,
+      "rewards/check_winston_local_func": 0.4823804199695587,
+      "step": 485
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6797202797202797,
+      "grad_norm": 3.855750270963149e-05,
+      "kl": 0.140625,
+      "learning_rate": 9.851773754922152e-05,
+      "loss": 0.0001,
+      "step": 486
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 110.5714340209961,
+      "epoch": 0.6811188811188811,
+      "grad_norm": 1.0000521495087345e-06,
+      "kl": 0.1572265625,
+      "learning_rate": 9.850295246595898e-05,
+      "loss": 0.0002,
+      "num_tokens": 4946515.0,
+      "reward": 2.9591448307037354,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.5,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.5305731296539307,
+      "step": 487
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6825174825174826,
+      "grad_norm": 9.579988148024946e-07,
+      "kl": 0.1572265625,
+      "learning_rate": 9.848809513054523e-05,
+      "loss": 0.0002,
+      "step": 488
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 104.67857360839844,
+      "epoch": 0.6839160839160839,
+      "grad_norm": 4.4290499511336513e-07,
+      "kl": 0.1728515625,
+      "learning_rate": 9.847316556511245e-05,
+      "loss": 0.0002,
+      "num_tokens": 4966537.0,
+      "reward": 3.0352413654327393,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4642857015132904,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.5709553956985474,
+      "step": 489
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6853146853146853,
+      "grad_norm": 4.228282290104698e-07,
+      "kl": 0.1728515625,
+      "learning_rate": 9.845816379190036e-05,
+      "loss": 0.0002,
+      "step": 490
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 106.92857360839844,
+      "epoch": 0.6867132867132867,
+      "grad_norm": 7.161264654013381e-07,
+      "kl": 0.1865234375,
+      "learning_rate": 9.844308983325625e-05,
+      "loss": 0.0002,
+      "num_tokens": 4986837.0,
+      "reward": 2.8011527061462402,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4642857015132904,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.40829533338546753,
+      "step": 491
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6881118881118881,
+      "grad_norm": 6.606409864369399e-07,
+      "kl": 0.1865234375,
+      "learning_rate": 9.842794371163501e-05,
+      "loss": 0.0002,
+      "step": 492
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 127.3214340209961,
+      "epoch": 0.6895104895104895,
+      "grad_norm": 3.5056088482557826e-05,
+      "kl": 0.11767578125,
+      "learning_rate": 9.841272544959892e-05,
+      "loss": 0.0001,
+      "num_tokens": 5009783.0,
+      "reward": 2.6788315773010254,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.5,
+      "rewards/check_perplexity_diff_func": 1.7857142686843872,
+      "rewards/check_winston_local_func": 0.393117219209671,
+      "step": 493
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6909090909090909,
+      "grad_norm": 3.5363245519599165e-05,
+      "kl": 0.11767578125,
+      "learning_rate": 9.839743506981782e-05,
+      "loss": 0.0001,
+      "step": 494
+    },
+    {
+      "clip_ratio": 0.0004093328316230327,
+      "completion_length": 99.08928680419922,
+      "epoch": 0.6923076923076923,
+      "grad_norm": 0.0028759704706919223,
+      "kl": 0.1650390625,
+      "learning_rate": 9.838207259506891e-05,
+      "loss": 0.0002,
+      "num_tokens": 5029238.0,
+      "reward": 2.5216493606567383,
+      "reward_std": 0.0011310166446492076,
+      "rewards/check_gptzero_func": 0.3571428656578064,
+      "rewards/check_perplexity_diff_func": 1.7142857313156128,
+      "rewards/check_winston_local_func": 0.45022064447402954,
+      "step": 495
+    },
+    {
+      "clip_ratio": 0.0004093328316230327,
+      "epoch": 0.6937062937062937,
+      "grad_norm": 0.0035050811312865673,
+      "kl": 0.1650390625,
+      "learning_rate": 9.836663804823683e-05,
+      "loss": 0.0002,
+      "step": 496
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 124.85714721679688,
+      "epoch": 0.6951048951048951,
+      "grad_norm": 3.8191875557546916e-05,
+      "kl": 0.158203125,
+      "learning_rate": 9.835113145231356e-05,
+      "loss": 0.0002,
+      "num_tokens": 5051408.0,
+      "reward": 2.8089945316314697,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4642857015132904,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.4161372184753418,
+      "step": 497
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6965034965034965,
+      "grad_norm": 3.1854181850026e-05,
+      "kl": 0.158203125,
+      "learning_rate": 9.833555283039842e-05,
+      "loss": 0.0002,
+      "step": 498
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 115.17857360839844,
+      "epoch": 0.6979020979020979,
+      "grad_norm": 3.735234931957505e-07,
+      "kl": 0.15234375,
+      "learning_rate": 9.831990220569801e-05,
+      "loss": 0.0002,
+      "num_tokens": 5072486.0,
+      "reward": 2.7535111904144287,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3571428656578064,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.46779707074165344,
+      "step": 499
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6993006993006993,
+      "grad_norm": 3.749561048042854e-07,
+      "kl": 0.15234375,
+      "learning_rate": 9.83041796015262e-05,
+      "loss": 0.0002,
+      "step": 500
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 137.85714721679688,
+      "epoch": 0.7006993006993008,
+      "grad_norm": 0.004581651252064326,
+      "kl": 0.1357421875,
+      "learning_rate": 9.828838504130406e-05,
+      "loss": 0.0001,
+      "num_tokens": 24128.0,
+      "reward": 2.5751242637634277,
+      "reward_std": 0.050465863198041916,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 1.8214285373687744,
+      "rewards/check_winston_local_func": 0.36083847284317017,
+      "step": 501
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7020979020979021,
+      "grad_norm": 0.004623783518082634,
+      "kl": 0.1357421875,
+      "learning_rate": 9.827251854855991e-05,
+      "loss": 0.0001,
+      "step": 502
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 84.0714340209961,
+      "epoch": 0.7034965034965035,
+      "grad_norm": 5.4224474596128455e-06,
+      "kl": 0.1962890625,
+      "learning_rate": 9.825658014692914e-05,
+      "loss": 0.0002,
+      "num_tokens": 41660.0,
+      "reward": 2.881941556930542,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.48908427357673645,
+      "step": 503
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7048951048951049,
+      "grad_norm": 5.774418529053615e-06,
+      "kl": 0.1962890625,
+      "learning_rate": 9.824056986015433e-05,
+      "loss": 0.0002,
+      "step": 504
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 125.76786041259766,
+      "epoch": 0.7062937062937062,
+      "grad_norm": 0.012190639094338367,
+      "kl": 0.158203125,
+      "learning_rate": 9.82244877120851e-05,
+      "loss": -0.0,
+      "num_tokens": 63809.0,
+      "reward": 2.9153220653533936,
+      "reward_std": 0.05151599273085594,
+      "rewards/check_gptzero_func": 0.5,
+      "rewards/check_perplexity_diff_func": 1.9642857313156128,
+      "rewards/check_winston_local_func": 0.45103612542152405,
+      "step": 505
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7076923076923077,
+      "grad_norm": 0.0008706312902749691,
+      "kl": 0.1572265625,
+      "learning_rate": 9.820833372667812e-05,
+      "loss": -0.0001,
+      "step": 506
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 106.67857360839844,
+      "epoch": 0.7090909090909091,
+      "grad_norm": 7.4245877265908374e-06,
+      "kl": 0.171875,
+      "learning_rate": 9.819210792799712e-05,
+      "loss": 0.0002,
+      "num_tokens": 83877.0,
+      "reward": 2.742607831954956,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3571428656578064,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.3854646682739258,
+      "step": 507
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7104895104895105,
+      "grad_norm": 5.9863315776137925e-06,
+      "kl": 0.171875,
+      "learning_rate": 9.817581034021272e-05,
+      "loss": 0.0002,
+      "step": 508
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 112.21429443359375,
+      "epoch": 0.7118881118881119,
+      "grad_norm": 1.5219106062929897e-05,
+      "kl": 0.154296875,
+      "learning_rate": 9.815944098760257e-05,
+      "loss": 0.0002,
+      "num_tokens": 105225.0,
+      "reward": 2.886364698410034,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.5649359226226807,
+      "step": 509
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7132867132867133,
+      "grad_norm": 3.826854189322478e-05,
+      "kl": 0.154296875,
+      "learning_rate": 9.814299989455117e-05,
+      "loss": 0.0002,
+      "step": 510
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 93.50000762939453,
+      "epoch": 0.7146853146853147,
+      "grad_norm": 3.6634275067756628e-06,
+      "kl": 0.18359375,
+      "learning_rate": 9.81264870855499e-05,
+      "loss": 0.0002,
+      "num_tokens": 124237.0,
+      "reward": 2.9750092029571533,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.5,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.5464377403259277,
+      "step": 511
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7160839160839161,
+      "grad_norm": 3.3120330925968346e-06,
+      "kl": 0.18359375,
+      "learning_rate": 9.810990258519699e-05,
+      "loss": 0.0002,
+      "step": 512
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 114.5714340209961,
+      "epoch": 0.7174825174825175,
+      "grad_norm": 1.5887563588226216e-07,
+      "kl": 0.1474609375,
+      "learning_rate": 9.809324641819741e-05,
+      "loss": 0.0001,
+      "num_tokens": 145557.0,
+      "reward": 2.767850637435913,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3571428656578064,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.4107076823711395,
+      "step": 513
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7188811188811188,
+      "grad_norm": 1.5359302478120885e-07,
+      "kl": 0.1474609375,
+      "learning_rate": 9.807651860936297e-05,
+      "loss": 0.0001,
+      "step": 514
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 110.3214340209961,
+      "epoch": 0.7202797202797203,
+      "grad_norm": 1.620447571811333e-07,
+      "kl": 0.1630859375,
+      "learning_rate": 9.805971918361214e-05,
+      "loss": 0.0002,
+      "num_tokens": 165997.0,
+      "reward": 2.9055941104888916,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.5127367377281189,
+      "step": 515
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7216783216783217,
+      "grad_norm": 1.472040658632873e-07,
+      "kl": 0.1630859375,
+      "learning_rate": 9.804284816597008e-05,
+      "loss": 0.0002,
+      "step": 516
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 91.67857360839844,
+      "epoch": 0.7230769230769231,
+      "grad_norm": 2.4623250489986615e-06,
+      "kl": 0.142578125,
+      "learning_rate": 9.802590558156862e-05,
+      "loss": 0.0001,
+      "num_tokens": 184617.0,
+      "reward": 2.853912830352783,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.46105554699897766,
+      "step": 517
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7244755244755244,
+      "grad_norm": 2.0559549189734124e-06,
+      "kl": 0.142578125,
+      "learning_rate": 9.800889145564617e-05,
+      "loss": 0.0001,
+      "step": 518
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 127.21429443359375,
+      "epoch": 0.7258741258741259,
+      "grad_norm": 2.3530034340292295e-07,
+      "kl": 0.1552734375,
+      "learning_rate": 9.799180581354774e-05,
+      "loss": 0.0002,
+      "num_tokens": 207223.0,
+      "reward": 2.7364511489868164,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3571428656578064,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.45073673129081726,
+      "step": 519
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7272727272727273,
+      "grad_norm": 2.2322343800833895e-07,
+      "kl": 0.1552734375,
+      "learning_rate": 9.797464868072488e-05,
+      "loss": 0.0002,
+      "step": 520
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 104.5714340209961,
+      "epoch": 0.7286713286713287,
+      "grad_norm": 2.968346809067501e-07,
+      "kl": 0.177734375,
+      "learning_rate": 9.795742008273558e-05,
+      "loss": 0.0002,
+      "num_tokens": 227059.0,
+      "reward": 2.6932995319366455,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3571428656578064,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.3361565172672272,
+      "step": 521
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.73006993006993,
+      "grad_norm": 2.852980195095417e-07,
+      "kl": 0.177734375,
+      "learning_rate": 9.794012004524434e-05,
+      "loss": 0.0002,
+      "step": 522
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 105.42857360839844,
+      "epoch": 0.7314685314685314,
+      "grad_norm": 3.3796985539550523e-06,
+      "kl": 0.1640625,
+      "learning_rate": 9.792274859402205e-05,
+      "loss": 0.0002,
+      "num_tokens": 246949.0,
+      "reward": 2.6483211517333984,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3214285671710968,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.3983212113380432,
+      "step": 523
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7328671328671329,
+      "grad_norm": 3.1476330793325015e-06,
+      "kl": 0.1640625,
+      "learning_rate": 9.790530575494603e-05,
+      "loss": 0.0002,
+      "step": 524
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 125.03572082519531,
+      "epoch": 0.7342657342657343,
+      "grad_norm": 0.0060337949589422705,
+      "kl": 0.16796875,
+      "learning_rate": 9.788779155399987e-05,
+      "loss": -0.0001,
+      "num_tokens": 269441.0,
+      "reward": 2.768662214279175,
+      "reward_std": 0.049707408994436264,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 1.9642857313156128,
+      "rewards/check_winston_local_func": 0.4115191400051117,
+      "step": 525
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7356643356643356,
+      "grad_norm": 0.0068588182961818464,
+      "kl": 0.16796875,
+      "learning_rate": 9.787020601727352e-05,
+      "loss": -0.0,
+      "step": 526
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 78.39286041259766,
+      "epoch": 0.737062937062937,
+      "grad_norm": 1.3153821115529096e-05,
+      "kl": 0.2353515625,
+      "learning_rate": 9.785254917096318e-05,
+      "loss": 0.0002,
+      "num_tokens": 286345.0,
+      "reward": 3.074557304382324,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.5,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.5745573043823242,
+      "step": 527
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7384615384615385,
+      "grad_norm": 1.256427796379324e-05,
+      "kl": 0.2353515625,
+      "learning_rate": 9.783482104137127e-05,
+      "loss": 0.0002,
+      "step": 528
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 124.39286041259766,
+      "epoch": 0.7398601398601399,
+      "grad_norm": 1.3517164643713064e-05,
+      "kl": 0.12353515625,
+      "learning_rate": 9.781702165490639e-05,
+      "loss": 0.0001,
+      "num_tokens": 309195.0,
+      "reward": 2.4804210662841797,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 1.6428571939468384,
+      "rewards/check_winston_local_func": 0.44470664858818054,
+      "step": 529
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7412587412587412,
+      "grad_norm": 1.3131407312669822e-05,
+      "kl": 0.12353515625,
+      "learning_rate": 9.779915103808328e-05,
+      "loss": 0.0001,
+      "step": 530
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 111.53572082519531,
+      "epoch": 0.7426573426573426,
+      "grad_norm": 7.637416736333039e-07,
+      "kl": 0.15625,
+      "learning_rate": 9.778120921752285e-05,
+      "loss": 0.0002,
+      "num_tokens": 330505.0,
+      "reward": 3.070021390914917,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.5357142686843872,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.5343068838119507,
+      "step": 531
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7440559440559441,
+      "grad_norm": 7.316834780823122e-07,
+      "kl": 0.15625,
+      "learning_rate": 9.776319621995201e-05,
+      "loss": 0.0002,
+      "step": 532
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 106.35714721679688,
+      "epoch": 0.7454545454545455,
+      "grad_norm": 3.663080215251399e-07,
+      "kl": 0.193359375,
+      "learning_rate": 9.77451120722037e-05,
+      "loss": 0.0002,
+      "num_tokens": 350813.0,
+      "reward": 2.63775634765625,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3214285671710968,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.3877563774585724,
+      "step": 533
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7468531468531469,
+      "grad_norm": 3.8678973306591443e-07,
+      "kl": 0.193359375,
+      "learning_rate": 9.77269568012169e-05,
+      "loss": 0.0002,
+      "step": 534
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 106.67857360839844,
+      "epoch": 0.7482517482517482,
+      "grad_norm": 6.123173544504598e-06,
+      "kl": 0.2216796875,
+      "learning_rate": 9.770873043403648e-05,
+      "loss": 0.0002,
+      "num_tokens": 371585.0,
+      "reward": 2.874614953994751,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.5174719095230103,
+      "step": 535
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7496503496503496,
+      "grad_norm": 6.141198966103098e-06,
+      "kl": 0.2216796875,
+      "learning_rate": 9.769043299781327e-05,
+      "loss": 0.0002,
+      "step": 536
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 104.71428680419922,
+      "epoch": 0.7510489510489511,
+      "grad_norm": 6.024782598667822e-07,
+      "kl": 0.212890625,
+      "learning_rate": 9.767206451980394e-05,
+      "loss": 0.0002,
+      "num_tokens": 391201.0,
+      "reward": 2.666369676589966,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.25,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.4163695275783539,
+      "step": 537
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7524475524475525,
+      "grad_norm": 5.651815000751003e-07,
+      "kl": 0.212890625,
+      "learning_rate": 9.765362502737097e-05,
+      "loss": 0.0002,
+      "step": 538
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 107.21428680419922,
+      "epoch": 0.7538461538461538,
+      "grad_norm": 4.5359076717402366e-07,
+      "kl": 0.19140625,
+      "learning_rate": 9.763511454798268e-05,
+      "loss": 0.0002,
+      "num_tokens": 412087.0,
+      "reward": 2.660142660140991,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.25,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.4815710484981537,
+      "step": 539
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7552447552447552,
+      "grad_norm": 7.72545386297213e-07,
+      "kl": 0.19140625,
+      "learning_rate": 9.761653310921307e-05,
+      "loss": 0.0002,
+      "step": 540
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 133.85714721679688,
+      "epoch": 0.7566433566433567,
+      "grad_norm": 8.430058973039463e-08,
+      "kl": 0.146484375,
+      "learning_rate": 9.759788073874189e-05,
+      "loss": 0.0001,
+      "num_tokens": 435535.0,
+      "reward": 2.720244884490967,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.3631021976470947,
+      "step": 541
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7580419580419581,
+      "grad_norm": 8.327469845321351e-08,
+      "kl": 0.146484375,
+      "learning_rate": 9.757915746435453e-05,
+      "loss": 0.0001,
+      "step": 542
+    },
+    {
+      "clip_ratio": 0.0002040816325461492,
+      "completion_length": 102.89286041259766,
+      "epoch": 0.7594405594405594,
+      "grad_norm": 0.000356312443076848,
+      "kl": 0.203125,
+      "learning_rate": 9.756036331394202e-05,
+      "loss": 0.0002,
+      "num_tokens": 455137.0,
+      "reward": 2.6731011867523193,
+      "reward_std": 1.646135569899343e-05,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.351672500371933,
+      "step": 543
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7608391608391608,
+      "grad_norm": 0.00031516713281171804,
+      "kl": 0.203125,
+      "learning_rate": 9.754149831550098e-05,
+      "loss": 0.0002,
+      "step": 544
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 89.35714721679688,
+      "epoch": 0.7622377622377622,
+      "grad_norm": 2.70311309473727e-07,
+      "kl": 0.2080078125,
+      "learning_rate": 9.752256249713351e-05,
+      "loss": 0.0002,
+      "num_tokens": 473297.0,
+      "reward": 2.869070291519165,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3571428656578064,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.5119272470474243,
+      "step": 545
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7636363636363637,
+      "grad_norm": 2.830814999627113e-07,
+      "kl": 0.2080078125,
+      "learning_rate": 9.750355588704727e-05,
+      "loss": 0.0002,
+      "step": 546
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 132.44644165039062,
+      "epoch": 0.765034965034965,
+      "grad_norm": 0.0009688545197921594,
+      "kl": 0.1806640625,
+      "learning_rate": 9.748447851355535e-05,
+      "loss": 0.0002,
+      "num_tokens": 496926.0,
+      "reward": 2.7388832569122314,
+      "reward_std": 0.0023702967446297407,
+      "rewards/check_gptzero_func": 0.5357142686843872,
+      "rewards/check_perplexity_diff_func": 1.7142857313156128,
+      "rewards/check_winston_local_func": 0.4888834059238434,
+      "step": 547
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7664335664335664,
+      "grad_norm": 0.0016734864168510978,
+      "kl": 0.1806640625,
+      "learning_rate": 9.746533040507624e-05,
+      "loss": 0.0002,
+      "step": 548
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 97.03572082519531,
+      "epoch": 0.7678321678321678,
+      "grad_norm": 1.9163803223436476e-06,
+      "kl": 0.177734375,
+      "learning_rate": 9.744611159013381e-05,
+      "loss": 0.0002,
+      "num_tokens": 516080.0,
+      "reward": 2.9889118671417236,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.5,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.48891177773475647,
+      "step": 549
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7692307692307693,
+      "grad_norm": 1.2761616262245509e-06,
+      "kl": 0.177734375,
+      "learning_rate": 9.742682209735727e-05,
+      "loss": 0.0002,
+      "step": 550
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 92.39286041259766,
+      "epoch": 0.7706293706293706,
+      "grad_norm": 1.896094459987994e-07,
+      "kl": 0.1923828125,
+      "learning_rate": 9.740746195548112e-05,
+      "loss": 0.0002,
+      "num_tokens": 534554.0,
+      "reward": 2.943678855895996,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.5151072144508362,
+      "step": 551
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.772027972027972,
+      "grad_norm": 2.1137044966091938e-07,
+      "kl": 0.1923828125,
+      "learning_rate": 9.738803119334506e-05,
+      "loss": 0.0002,
+      "step": 552
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 103.21428680419922,
+      "epoch": 0.7734265734265734,
+      "grad_norm": 4.705318347406964e-07,
+      "kl": 0.1826171875,
+      "learning_rate": 9.736852983989404e-05,
+      "loss": 0.0002,
+      "num_tokens": 554220.0,
+      "reward": 2.9958415031433105,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.5,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.4958415627479553,
+      "step": 553
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7748251748251749,
+      "grad_norm": 5.24917119837944e-07,
+      "kl": 0.1826171875,
+      "learning_rate": 9.734895792417811e-05,
+      "loss": 0.0002,
+      "step": 554
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 102.53572082519531,
+      "epoch": 0.7762237762237763,
+      "grad_norm": 0.0005595837692618245,
+      "kl": 0.1611328125,
+      "learning_rate": 9.73293154753525e-05,
+      "loss": -0.0001,
+      "num_tokens": 574520.0,
+      "reward": 2.660358190536499,
+      "reward_std": 0.05061452463269234,
+      "rewards/check_gptzero_func": 0.2857142984867096,
+      "rewards/check_perplexity_diff_func": 1.75,
+      "rewards/check_winston_local_func": 0.6246438026428223,
+      "step": 555
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7776223776223776,
+      "grad_norm": 0.0005476655329061088,
+      "kl": 0.1611328125,
+      "learning_rate": 9.730960252267743e-05,
+      "loss": -0.0001,
+      "step": 556
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 110.28572082519531,
+      "epoch": 0.779020979020979,
+      "grad_norm": 1.9994241055330475e-07,
+      "kl": 0.162109375,
+      "learning_rate": 9.728981909551824e-05,
+      "loss": 0.0002,
+      "num_tokens": 595056.0,
+      "reward": 2.7659287452697754,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3571428656578064,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.40878555178642273,
+      "step": 557
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7804195804195804,
+      "grad_norm": 2.102720872162732e-07,
+      "kl": 0.162109375,
+      "learning_rate": 9.726996522334516e-05,
+      "loss": 0.0002,
+      "step": 558
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 116.3214340209961,
+      "epoch": 0.7818181818181819,
+      "grad_norm": 1.4418577967265551e-06,
+      "kl": 0.197265625,
+      "learning_rate": 9.725004093573342e-05,
+      "loss": 0.0002,
+      "num_tokens": 616218.0,
+      "reward": 2.90523099899292,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.4766596257686615,
+      "step": 559
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7832167832167832,
+      "grad_norm": 1.4761501812795077e-06,
+      "kl": 0.197265625,
+      "learning_rate": 9.723004626236314e-05,
+      "loss": 0.0002,
+      "step": 560
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 92.28572082519531,
+      "epoch": 0.7846153846153846,
+      "grad_norm": 6.640556191873379e-07,
+      "kl": 0.1904296875,
+      "learning_rate": 9.720998123301923e-05,
+      "loss": 0.0002,
+      "num_tokens": 634450.0,
+      "reward": 2.910466432571411,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4642857015132904,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.4461804926395416,
+      "step": 561
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.786013986013986,
+      "grad_norm": 6.785626739636823e-07,
+      "kl": 0.1904296875,
+      "learning_rate": 9.718984587759148e-05,
+      "loss": 0.0002,
+      "step": 562
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 97.67857360839844,
+      "epoch": 0.7874125874125875,
+      "grad_norm": 0.0002364011375932314,
+      "kl": 0.1953125,
+      "learning_rate": 9.71696402260744e-05,
+      "loss": 0.0002,
+      "num_tokens": 653730.0,
+      "reward": 3.09153151512146,
+      "reward_std": 4.5663102355320007e-05,
+      "rewards/check_gptzero_func": 0.5714285969734192,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.5201030373573303,
+      "step": 563
+    },
+    {
+      "clip_ratio": 0.00014228800137061626,
+      "epoch": 0.7888111888111888,
+      "grad_norm": 0.00020388090420570156,
+      "kl": 0.1953125,
+      "learning_rate": 9.714936430856723e-05,
+      "loss": 0.0002,
+      "step": 564
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 120.66072082519531,
+      "epoch": 0.7902097902097902,
+      "grad_norm": 8.089432221384178e-05,
+      "kl": 0.1357421875,
+      "learning_rate": 9.712901815527386e-05,
+      "loss": 0.0001,
+      "num_tokens": 675657.0,
+      "reward": 2.674098253250122,
+      "reward_std": 0.0001108883589040488,
+      "rewards/check_gptzero_func": 0.3571428656578064,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.38838380575180054,
+      "step": 565
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7916083916083916,
+      "grad_norm": 8.889658368646956e-05,
+      "kl": 0.1357421875,
+      "learning_rate": 9.710860179650287e-05,
+      "loss": 0.0001,
+      "step": 566
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 107.3214340209961,
+      "epoch": 0.793006993006993,
+      "grad_norm": 0.0001037228042730343,
+      "kl": 0.17578125,
+      "learning_rate": 9.70881152626673e-05,
+      "loss": 0.0002,
+      "num_tokens": 695807.0,
+      "reward": 2.615325927734375,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.25,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.4367544949054718,
+      "step": 567
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7944055944055944,
+      "grad_norm": 9.62410190276379e-05,
+      "kl": 0.17578125,
+      "learning_rate": 9.706755858428486e-05,
+      "loss": 0.0002,
+      "step": 568
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 128.3928680419922,
+      "epoch": 0.7958041958041958,
+      "grad_norm": 3.2052372287941215e-05,
+      "kl": 0.12890625,
+      "learning_rate": 9.704693179197767e-05,
+      "loss": 0.0001,
+      "num_tokens": 719201.0,
+      "reward": 2.616921901702881,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 1.7857142686843872,
+      "rewards/check_winston_local_func": 0.40263620018959045,
+      "step": 569
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7972027972027972,
+      "grad_norm": 3.1589569913845654e-05,
+      "kl": 0.12890625,
+      "learning_rate": 9.702623491647233e-05,
+      "loss": 0.0001,
+      "step": 570
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 97.3214340209961,
+      "epoch": 0.7986013986013986,
+      "grad_norm": 2.966365002312992e-07,
+      "kl": 0.1572265625,
+      "learning_rate": 9.70054679885998e-05,
+      "loss": 0.0002,
+      "num_tokens": 738947.0,
+      "reward": 2.835333824157715,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.47819074988365173,
+      "step": 571
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8,
+      "grad_norm": 4.1573650424054196e-07,
+      "kl": 0.1572265625,
+      "learning_rate": 9.698463103929542e-05,
+      "loss": 0.0002,
+      "step": 572
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 105.78572082519531,
+      "epoch": 0.8013986013986014,
+      "grad_norm": 0.000588551803293507,
+      "kl": 0.1611328125,
+      "learning_rate": 9.696372409959886e-05,
+      "loss": 0.0002,
+      "num_tokens": 759359.0,
+      "reward": 2.967468738555908,
+      "reward_std": 0.0020603849552571774,
+      "rewards/check_gptzero_func": 0.5,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.5388973355293274,
+      "step": 573
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8027972027972028,
+      "grad_norm": 0.0006215594211925717,
+      "kl": 0.1611328125,
+      "learning_rate": 9.694274720065399e-05,
+      "loss": 0.0002,
+      "step": 574
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 97.14286041259766,
+      "epoch": 0.8041958041958042,
+      "grad_norm": 3.3967306751168834e-06,
+      "kl": 0.17578125,
+      "learning_rate": 9.692170037370898e-05,
+      "loss": 0.0002,
+      "num_tokens": 778637.0,
+      "reward": 2.9381115436553955,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.509539783000946,
+      "step": 575
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8055944055944056,
+      "grad_norm": 3.2036118908588705e-06,
+      "kl": 0.17578125,
+      "learning_rate": 9.690058365011607e-05,
+      "loss": 0.0002,
+      "step": 576
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 102.64286041259766,
+      "epoch": 0.806993006993007,
+      "grad_norm": 8.125129903430204e-06,
+      "kl": 0.1630859375,
+      "learning_rate": 9.687939706133168e-05,
+      "loss": 0.0002,
+      "num_tokens": 798473.0,
+      "reward": 2.781018018722534,
+      "reward_std": 0.025253813713788986,
+      "rewards/check_gptzero_func": 0.4107142984867096,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.4417320787906647,
+      "step": 577
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8083916083916084,
+      "grad_norm": 7.98774787404577e-06,
+      "kl": 0.1630859375,
+      "learning_rate": 9.685814063891631e-05,
+      "loss": 0.0002,
+      "step": 578
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 126.85714721679688,
+      "epoch": 0.8097902097902098,
+      "grad_norm": 3.679599498892351e-07,
+      "kl": 0.138671875,
+      "learning_rate": 9.683681441453445e-05,
+      "loss": 0.0001,
+      "num_tokens": 821081.0,
+      "reward": 2.8629186153411865,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.43434715270996094,
+      "step": 579
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8111888111888111,
+      "grad_norm": 3.93663211553172e-07,
+      "kl": 0.138671875,
+      "learning_rate": 9.681541841995461e-05,
+      "loss": 0.0001,
+      "step": 580
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 106.00000762939453,
+      "epoch": 0.8125874125874126,
+      "grad_norm": 8.022029708518222e-07,
+      "kl": 0.162109375,
+      "learning_rate": 9.67939526870492e-05,
+      "loss": 0.0002,
+      "num_tokens": 841529.0,
+      "reward": 2.8752760887145996,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.55384761095047,
+      "step": 581
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.813986013986014,
+      "grad_norm": 8.501682817147847e-07,
+      "kl": 0.162109375,
+      "learning_rate": 9.677241724779453e-05,
+      "loss": 0.0002,
+      "step": 582
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 109.85714721679688,
+      "epoch": 0.8153846153846154,
+      "grad_norm": 1.7930012533014129e-06,
+      "kl": 0.1689453125,
+      "learning_rate": 9.675081213427076e-05,
+      "loss": 0.0002,
+      "num_tokens": 862213.0,
+      "reward": 2.9139628410339355,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.48539119958877563,
+      "step": 583
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8167832167832167,
+      "grad_norm": 1.6519465345892093e-06,
+      "kl": 0.1689453125,
+      "learning_rate": 9.672913737866179e-05,
+      "loss": 0.0002,
+      "step": 584
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 98.66072082519531,
+      "epoch": 0.8181818181818182,
+      "grad_norm": 0.002787236207669909,
+      "kl": 0.25,
+      "learning_rate": 9.670739301325534e-05,
+      "loss": 0.0002,
+      "num_tokens": 881842.0,
+      "reward": 2.6230576038360596,
+      "reward_std": 0.006216124631464481,
+      "rewards/check_gptzero_func": 0.3571428656578064,
+      "rewards/check_perplexity_diff_func": 1.7857142686843872,
+      "rewards/check_winston_local_func": 0.48020049929618835,
+      "step": 585
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8195804195804196,
+      "grad_norm": 0.002885109231591189,
+      "kl": 0.2490234375,
+      "learning_rate": 9.668557907044276e-05,
+      "loss": 0.0002,
+      "step": 586
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 79.14286041259766,
+      "epoch": 0.820979020979021,
+      "grad_norm": 4.462250819797621e-06,
+      "kl": 0.1923828125,
+      "learning_rate": 9.666369558271909e-05,
+      "loss": 0.0002,
+      "num_tokens": 898716.0,
+      "reward": 2.900649309158325,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.5077921152114868,
+      "step": 587
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8223776223776224,
+      "grad_norm": 4.488075066816524e-06,
+      "kl": 0.1923828125,
+      "learning_rate": 9.66417425826829e-05,
+      "loss": 0.0002,
+      "step": 588
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 112.03572082519531,
+      "epoch": 0.8237762237762237,
+      "grad_norm": 1.3821264354883407e-07,
+      "kl": 0.14453125,
+      "learning_rate": 9.661972010303641e-05,
+      "loss": 0.0001,
+      "num_tokens": 919636.0,
+      "reward": 2.7668612003326416,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4642857015132904,
+      "rewards/check_perplexity_diff_func": 1.8571428060531616,
+      "rewards/check_winston_local_func": 0.4454323649406433,
+      "step": 589
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8251748251748252,
+      "grad_norm": 1.4174011116816794e-07,
+      "kl": 0.14453125,
+      "learning_rate": 9.659762817658524e-05,
+      "loss": 0.0001,
+      "step": 590
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 107.92857360839844,
+      "epoch": 0.8265734265734266,
+      "grad_norm": 8.773734381605369e-07,
+      "kl": 0.1572265625,
+      "learning_rate": 9.65754668362385e-05,
+      "loss": 0.0002,
+      "num_tokens": 940096.0,
+      "reward": 2.8289589881896973,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4642857015132904,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.4361015856266022,
+      "step": 591
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.827972027972028,
+      "grad_norm": 1.0023469469060766e-06,
+      "kl": 0.1572265625,
+      "learning_rate": 9.655323611500875e-05,
+      "loss": 0.0002,
+      "step": 592
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 113.85714721679688,
+      "epoch": 0.8293706293706293,
+      "grad_norm": 6.1541718025779055e-06,
+      "kl": 0.166015625,
+      "learning_rate": 9.653093604601183e-05,
+      "loss": 0.0002,
+      "num_tokens": 961340.0,
+      "reward": 2.9606730937957764,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.5357142686843872,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.4249587059020996,
+      "step": 593
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8307692307692308,
+      "grad_norm": 6.512813773208274e-06,
+      "kl": 0.166015625,
+      "learning_rate": 9.650856666246693e-05,
+      "loss": 0.0002,
+      "step": 594
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 91.60714721679688,
+      "epoch": 0.8321678321678322,
+      "grad_norm": 3.2190340652283214e-06,
+      "kl": 0.1767578125,
+      "learning_rate": 9.648612799769644e-05,
+      "loss": 0.0002,
+      "num_tokens": 979916.0,
+      "reward": 3.0096070766448975,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.5810357332229614,
+      "step": 595
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8335664335664336,
+      "grad_norm": 3.1137752081162388e-06,
+      "kl": 0.1767578125,
+      "learning_rate": 9.646362008512602e-05,
+      "loss": 0.0002,
+      "step": 596
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 106.53572082519531,
+      "epoch": 0.8349650349650349,
+      "grad_norm": 6.812051267845749e-06,
+      "kl": 0.185546875,
+      "learning_rate": 9.644104295828447e-05,
+      "loss": 0.0002,
+      "num_tokens": 1000300.0,
+      "reward": 2.7739078998565674,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.4167649447917938,
+      "step": 597
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8363636363636363,
+      "grad_norm": 2.398683493432653e-06,
+      "kl": 0.185546875,
+      "learning_rate": 9.641839665080363e-05,
+      "loss": 0.0002,
+      "step": 598
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 107.41072082519531,
+      "epoch": 0.8377622377622378,
+      "grad_norm": 0.0008785473557125962,
+      "kl": 0.1796875,
+      "learning_rate": 9.63956811964185e-05,
+      "loss": 0.0002,
+      "num_tokens": 1021105.0,
+      "reward": 2.91171932220459,
+      "reward_std": 0.00400555832311511,
+      "rewards/check_gptzero_func": 0.5357142686843872,
+      "rewards/check_perplexity_diff_func": 1.8571428060531616,
+      "rewards/check_winston_local_func": 0.518862247467041,
+      "step": 599
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8391608391608392,
+      "grad_norm": 0.000892784560673435,
+      "kl": 0.1796875,
+      "learning_rate": 9.6372896628967e-05,
+      "loss": 0.0002,
+      "step": 600
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 78.8214340209961,
+      "epoch": 0.8405594405594405,
+      "grad_norm": 3.255738764439449e-05,
+      "kl": 0.2236328125,
+      "learning_rate": 9.635004298239004e-05,
+      "loss": 0.0002,
+      "num_tokens": 16776.0,
+      "reward": 2.546457290649414,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 1.0714285373687744,
+      "rewards/check_perplexity_diff_func": 0.9285714030265808,
+      "rewards/check_winston_local_func": 0.5464571118354797,
+      "step": 601
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8419580419580419,
+      "grad_norm": 2.996409313034994e-05,
+      "kl": 0.2236328125,
+      "learning_rate": 9.632712029073141e-05,
+      "loss": 0.0002,
+      "step": 602
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 98.50000762939453,
+      "epoch": 0.8433566433566434,
+      "grad_norm": 6.836714560556308e-07,
+      "kl": 0.1796875,
+      "learning_rate": 9.63041285881378e-05,
+      "loss": 0.0002,
+      "num_tokens": 36316.0,
+      "reward": 2.2952983379364014,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.8571428656578064,
+      "rewards/check_perplexity_diff_func": 0.9642857313156128,
+      "rewards/check_winston_local_func": 0.4738696217536926,
+      "step": 603
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8447552447552448,
+      "grad_norm": 6.51699821327856e-07,
+      "kl": 0.1796875,
+      "learning_rate": 9.628106790885865e-05,
+      "loss": 0.0002,
+      "step": 604
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 108.67857360839844,
+      "epoch": 0.8461538461538461,
+      "grad_norm": 1.0500525475797962e-05,
+      "kl": 0.220703125,
+      "learning_rate": 9.625793828724618e-05,
+      "loss": 0.0002,
+      "num_tokens": 56798.0,
+      "reward": 2.2515501976013184,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.7857142686843872,
+      "rewards/check_perplexity_diff_func": 0.9642857313156128,
+      "rewards/check_winston_local_func": 0.5015498399734497,
+      "step": 605
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8475524475524475,
+      "grad_norm": 9.962219746884584e-06,
+      "kl": 0.220703125,
+      "learning_rate": 9.62347397577553e-05,
+      "loss": 0.0002,
+      "step": 606
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 123.64286041259766,
+      "epoch": 0.848951048951049,
+      "grad_norm": 0.00014103568607896937,
+      "kl": 0.1728515625,
+      "learning_rate": 9.621147235494356e-05,
+      "loss": 0.0002,
+      "num_tokens": 79358.0,
+      "reward": 2.151411771774292,
+      "reward_std": 0.00029218001873232424,
+      "rewards/check_gptzero_func": 0.7142857313156128,
+      "rewards/check_perplexity_diff_func": 0.9285714030265808,
+      "rewards/check_winston_local_func": 0.5085545778274536,
+      "step": 607
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8503496503496504,
+      "grad_norm": 0.00012130440152598759,
+      "kl": 0.171875,
+      "learning_rate": 9.618813611347113e-05,
+      "loss": 0.0002,
+      "step": 608
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 118.71429443359375,
+      "epoch": 0.8517482517482518,
+      "grad_norm": 1.8287469066668684e-07,
+      "kl": 0.1494140625,
+      "learning_rate": 9.61647310681007e-05,
+      "loss": 0.0001,
+      "num_tokens": 101324.0,
+      "reward": 2.4129910469055176,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.9285714030265808,
+      "rewards/check_perplexity_diff_func": 0.9285714030265808,
+      "rewards/check_winston_local_func": 0.5558480024337769,
+      "step": 609
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8531468531468531,
+      "grad_norm": 1.9169052102503996e-07,
+      "kl": 0.1494140625,
+      "learning_rate": 9.614125725369747e-05,
+      "loss": 0.0001,
+      "step": 610
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 84.30357360839844,
+      "epoch": 0.8545454545454545,
+      "grad_norm": 0.0035849846804439827,
+      "kl": 0.2314453125,
+      "learning_rate": 9.611771470522908e-05,
+      "loss": 0.0002,
+      "num_tokens": 118947.0,
+      "reward": 2.3909339904785156,
+      "reward_std": 0.0009617910836823285,
+      "rewards/check_gptzero_func": 0.8571428656578064,
+      "rewards/check_perplexity_diff_func": 1.0,
+      "rewards/check_winston_local_func": 0.533791184425354,
+      "step": 611
+    },
+    {
+      "clip_ratio": 0.00020614307140931487,
+      "epoch": 0.855944055944056,
+      "grad_norm": 0.0024010748168965985,
+      "kl": 0.22265625,
+      "learning_rate": 9.609410345776554e-05,
+      "loss": 0.0002,
+      "step": 612
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 97.92857360839844,
+      "epoch": 0.8573426573426574,
+      "grad_norm": 0.0165782655855821,
+      "kl": 0.220703125,
+      "learning_rate": 9.607042354647924e-05,
+      "loss": -0.0006,
+      "num_tokens": 138117.0,
+      "reward": 2.496706485748291,
+      "reward_std": 0.025315813720226288,
+      "rewards/check_gptzero_func": 0.9642857313156128,
+      "rewards/check_perplexity_diff_func": 0.9464285969734192,
+      "rewards/check_winston_local_func": 0.585992157459259,
+      "step": 613
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8587412587412587,
+      "grad_norm": 0.005231318150303852,
+      "kl": 0.220703125,
+      "learning_rate": 9.60466750066448e-05,
+      "loss": -0.0006,
+      "step": 614
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 116.71429443359375,
+      "epoch": 0.8601398601398601,
+      "grad_norm": 1.4205544083992568e-05,
+      "kl": 0.1533203125,
+      "learning_rate": 9.60228578736391e-05,
+      "loss": 0.0002,
+      "num_tokens": 159543.0,
+      "reward": 2.470890998840332,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 1.0,
+      "rewards/check_perplexity_diff_func": 1.0,
+      "rewards/check_winston_local_func": 0.47089076042175293,
+      "step": 615
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8615384615384616,
+      "grad_norm": 1.6303392641980344e-05,
+      "kl": 0.1533203125,
+      "learning_rate": 9.599897218294122e-05,
+      "loss": 0.0002,
+      "step": 616
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 112.50000762939453,
+      "epoch": 0.862937062937063,
+      "grad_norm": 1.6234044022820597e-07,
+      "kl": 0.2021484375,
+      "learning_rate": 9.597501797013233e-05,
+      "loss": 0.0002,
+      "num_tokens": 180563.0,
+      "reward": 2.09031081199646,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.6428571343421936,
+      "rewards/check_perplexity_diff_func": 0.9285714030265808,
+      "rewards/check_winston_local_func": 0.5188822746276855,
+      "step": 617
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8643356643356643,
+      "grad_norm": 1.666191215120993e-07,
+      "kl": 0.2021484375,
+      "learning_rate": 9.595099527089569e-05,
+      "loss": 0.0002,
+      "step": 618
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 108.78572082519531,
+      "epoch": 0.8657342657342657,
+      "grad_norm": 8.531816841675701e-07,
+      "kl": 0.193359375,
+      "learning_rate": 9.592690412101658e-05,
+      "loss": 0.0002,
+      "num_tokens": 201177.0,
+      "reward": 1.9570515155792236,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.5714285969734192,
+      "rewards/check_perplexity_diff_func": 1.0,
+      "rewards/check_winston_local_func": 0.38562270998954773,
+      "step": 619
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8671328671328671,
+      "grad_norm": 8.842725127740978e-07,
+      "kl": 0.193359375,
+      "learning_rate": 9.590274455638225e-05,
+      "loss": 0.0002,
+      "step": 620
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 138.07144165039062,
+      "epoch": 0.8685314685314686,
+      "grad_norm": 5.1580717328778724e-05,
+      "kl": 0.1640625,
+      "learning_rate": 9.587851661298186e-05,
+      "loss": 0.0002,
+      "num_tokens": 225661.0,
+      "reward": 2.487313747406006,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 1.0714285373687744,
+      "rewards/check_perplexity_diff_func": 0.9642857313156128,
+      "rewards/check_winston_local_func": 0.4515993893146515,
+      "step": 621
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8699300699300699,
+      "grad_norm": 5.097158350926591e-05,
+      "kl": 0.1640625,
+      "learning_rate": 9.585422032690643e-05,
+      "loss": 0.0002,
+      "step": 622
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 126.30357360839844,
+      "epoch": 0.8713286713286713,
+      "grad_norm": 1.5371764428374234e-05,
+      "kl": 0.1494140625,
+      "learning_rate": 9.582985573434877e-05,
+      "loss": 0.0002,
+      "num_tokens": 248756.0,
+      "reward": 2.386916160583496,
+      "reward_std": 0.0005263118073344231,
+      "rewards/check_gptzero_func": 0.8571428656578064,
+      "rewards/check_perplexity_diff_func": 1.0,
+      "rewards/check_winston_local_func": 0.5297732949256897,
+      "step": 623
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8727272727272727,
+      "grad_norm": 1.5410923964835553e-05,
+      "kl": 0.1494140625,
+      "learning_rate": 9.580542287160348e-05,
+      "loss": 0.0002,
+      "step": 624
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 121.85714721679688,
+      "epoch": 0.8741258741258742,
+      "grad_norm": 0.0002966067751134756,
+      "kl": 0.173828125,
+      "learning_rate": 9.578092177506683e-05,
+      "loss": 0.0002,
+      "num_tokens": 270514.0,
+      "reward": 2.7340447902679443,
+      "reward_std": 6.457499694079161e-05,
+      "rewards/check_gptzero_func": 1.2142857313156128,
+      "rewards/check_perplexity_diff_func": 0.9642857313156128,
+      "rewards/check_winston_local_func": 0.555473268032074,
+      "step": 625
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8755244755244755,
+      "grad_norm": 0.0002795708879844016,
+      "kl": 0.173828125,
+      "learning_rate": 9.575635248123675e-05,
+      "loss": 0.0002,
+      "step": 626
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 97.89286041259766,
+      "epoch": 0.8769230769230769,
+      "grad_norm": 1.0675950775787285e-07,
+      "kl": 0.1904296875,
+      "learning_rate": 9.573171502671273e-05,
+      "loss": 0.0002,
+      "num_tokens": 289588.0,
+      "reward": 2.467723846435547,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.9285714030265808,
+      "rewards/check_perplexity_diff_func": 1.0,
+      "rewards/check_winston_local_func": 0.5391523838043213,
+      "step": 627
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8783216783216783,
+      "grad_norm": 1.0758265762547126e-07,
+      "kl": 0.1904296875,
+      "learning_rate": 9.570700944819584e-05,
+      "loss": 0.0002,
+      "step": 628
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 126.96429443359375,
+      "epoch": 0.8797202797202798,
+      "grad_norm": 1.8189249055345895e-06,
+      "kl": 0.150390625,
+      "learning_rate": 9.568223578248858e-05,
+      "loss": 0.0002,
+      "num_tokens": 312792.0,
+      "reward": 2.4109601974487305,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.9285714030265808,
+      "rewards/check_perplexity_diff_func": 0.9285714030265808,
+      "rewards/check_winston_local_func": 0.5538173913955688,
+      "step": 629
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8811188811188811,
+      "grad_norm": 1.817429271106412e-06,
+      "kl": 0.150390625,
+      "learning_rate": 9.565739406649493e-05,
+      "loss": 0.0002,
+      "step": 630
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 109.64286041259766,
+      "epoch": 0.8825174825174825,
+      "grad_norm": 7.209254474831735e-05,
+      "kl": 0.169921875,
+      "learning_rate": 9.563248433722019e-05,
+      "loss": 0.0002,
+      "num_tokens": 333908.0,
+      "reward": 2.1215410232543945,
+      "reward_std": 0.0032710533123463392,
+      "rewards/check_gptzero_func": 0.7142857313156128,
+      "rewards/check_perplexity_diff_func": 0.9285714030265808,
+      "rewards/check_winston_local_func": 0.4786837100982666,
+      "step": 631
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8839160839160839,
+      "grad_norm": 0.005778816018850288,
+      "kl": 0.169921875,
+      "learning_rate": 9.560750663177101e-05,
+      "loss": 0.0002,
+      "step": 632
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 101.0714340209961,
+      "epoch": 0.8853146853146853,
+      "grad_norm": 5.141706041264495e-06,
+      "kl": 0.1689453125,
+      "learning_rate": 9.558246098735528e-05,
+      "loss": 0.0002,
+      "num_tokens": 353808.0,
+      "reward": 2.3030316829681396,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.7857142686843872,
+      "rewards/check_perplexity_diff_func": 0.9642857313156128,
+      "rewards/check_winston_local_func": 0.5530317425727844,
+      "step": 633
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8867132867132868,
+      "grad_norm": 5.171400787940402e-06,
+      "kl": 0.1689453125,
+      "learning_rate": 9.55573474412821e-05,
+      "loss": 0.0002,
+      "step": 634
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 112.0714340209961,
+      "epoch": 0.8881118881118881,
+      "grad_norm": 2.5579574578411833e-07,
+      "kl": 0.16015625,
+      "learning_rate": 9.553216603096176e-05,
+      "loss": 0.0002,
+      "num_tokens": 375024.0,
+      "reward": 2.2100911140441895,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.7857142686843872,
+      "rewards/check_perplexity_diff_func": 1.0,
+      "rewards/check_winston_local_func": 0.42437654733657837,
+      "step": 635
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8895104895104895,
+      "grad_norm": 3.163234894184058e-07,
+      "kl": 0.16015625,
+      "learning_rate": 9.550691679390558e-05,
+      "loss": 0.0002,
+      "step": 636
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 78.39286041259766,
+      "epoch": 0.8909090909090909,
+      "grad_norm": 2.7422578147300963e-06,
+      "kl": 0.19921875,
+      "learning_rate": 9.548159976772592e-05,
+      "loss": 0.0002,
+      "num_tokens": 391944.0,
+      "reward": 2.234971523284912,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.7142857313156128,
+      "rewards/check_perplexity_diff_func": 0.9285714030265808,
+      "rewards/check_winston_local_func": 0.5921144485473633,
+      "step": 637
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8923076923076924,
+      "grad_norm": 2.3669860027714333e-06,
+      "kl": 0.19921875,
+      "learning_rate": 9.54562149901362e-05,
+      "loss": 0.0002,
+      "step": 638
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 120.75000762939453,
+      "epoch": 0.8937062937062937,
+      "grad_norm": 2.3171625543487874e-07,
+      "kl": 0.158203125,
+      "learning_rate": 9.543076249895066e-05,
+      "loss": 0.0002,
+      "num_tokens": 413760.0,
+      "reward": 2.152951955795288,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.7142857313156128,
+      "rewards/check_perplexity_diff_func": 1.0,
+      "rewards/check_winston_local_func": 0.43866607546806335,
+      "step": 639
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8951048951048951,
+      "grad_norm": 2.4228268027328343e-07,
+      "kl": 0.158203125,
+      "learning_rate": 9.540524233208448e-05,
+      "loss": 0.0002,
+      "step": 640
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 87.05357360839844,
+      "epoch": 0.8965034965034965,
+      "grad_norm": 0.008398279317441937,
+      "kl": 0.181640625,
+      "learning_rate": 9.537965452755365e-05,
+      "loss": 0.0094,
+      "num_tokens": 431959.0,
+      "reward": 2.0755865573883057,
+      "reward_std": 0.019971024245023727,
+      "rewards/check_gptzero_func": 0.7142857313156128,
+      "rewards/check_perplexity_diff_func": 0.8571428656578064,
+      "rewards/check_winston_local_func": 0.5041577219963074,
+      "step": 641
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8979020979020979,
+      "grad_norm": 0.0032568280015785474,
+      "kl": 0.181640625,
+      "learning_rate": 9.535399912347489e-05,
+      "loss": 0.0094,
+      "step": 642
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 107.28572082519531,
+      "epoch": 0.8993006993006993,
+      "grad_norm": 4.5580158500525314e-05,
+      "kl": 0.150390625,
+      "learning_rate": 9.53282761580656e-05,
+      "loss": 0.0001,
+      "num_tokens": 452913.0,
+      "reward": 2.6801095008850098,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 1.1428571939468384,
+      "rewards/check_perplexity_diff_func": 0.9285714030265808,
+      "rewards/check_winston_local_func": 0.6086806058883667,
+      "step": 643
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.9006993006993007,
+      "grad_norm": 4.426477450510755e-05,
+      "kl": 0.150390625,
+      "learning_rate": 9.530248566964391e-05,
+      "loss": 0.0001,
+      "step": 644
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 105.25000762939453,
+      "epoch": 0.9020979020979021,
+      "grad_norm": 2.371230122251656e-05,
+      "kl": 0.1513671875,
+      "learning_rate": 9.527662769662841e-05,
+      "loss": 0.0002,
+      "num_tokens": 473141.0,
+      "reward": 2.1837356090545654,
+      "reward_std": 0.00018372414342593402,
+      "rewards/check_gptzero_func": 0.7142857313156128,
+      "rewards/check_perplexity_diff_func": 1.0,
+      "rewards/check_winston_local_func": 0.4694497585296631,
+      "step": 645
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.9034965034965035,
+      "grad_norm": 2.391891150952087e-05,
+      "kl": 0.1513671875,
+      "learning_rate": 9.525070227753834e-05,
+      "loss": 0.0002,
+      "step": 646
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 117.64286041259766,
+      "epoch": 0.9048951048951049,
+      "grad_norm": 1.3380347420537266e-07,
+      "kl": 0.150390625,
+      "learning_rate": 9.522470945099334e-05,
+      "loss": 0.0002,
+      "num_tokens": 494483.0,
+      "reward": 1.8219071626663208,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.5,
+      "rewards/check_perplexity_diff_func": 1.0,
+      "rewards/check_winston_local_func": 0.3219069838523865,
+      "step": 647
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.9062937062937063,
+      "grad_norm": 1.336307086326504e-07,
+      "kl": 0.150390625,
+      "learning_rate": 9.519864925571345e-05,
+      "loss": 0.0002,
+      "step": 648
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 131.0,
+      "epoch": 0.9076923076923077,
+      "grad_norm": 0.00012151914677644382,
+      "kl": 0.11474609375,
+      "learning_rate": 9.517252173051911e-05,
+      "loss": 0.0001,
+      "num_tokens": 518551.0,
+      "reward": 2.6524739265441895,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 1.2142857313156128,
+      "rewards/check_perplexity_diff_func": 0.8928571343421936,
+      "rewards/check_winston_local_func": 0.545330822467804,
+      "step": 649
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.9090909090909091,
+      "grad_norm": 5.8305084675906205e-05,
+      "kl": 0.11474609375,
+      "learning_rate": 9.514632691433107e-05,
+      "loss": 0.0001,
+      "step": 650
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 100.89286041259766,
+      "epoch": 0.9104895104895104,
+      "grad_norm": 5.230987317211898e-07,
+      "kl": 0.2119140625,
+      "learning_rate": 9.512006484617025e-05,
+      "loss": 0.0002,
+      "num_tokens": 538141.0,
+      "reward": 2.5491878986358643,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 1.0,
+      "rewards/check_perplexity_diff_func": 0.9642857313156128,
+      "rewards/check_winston_local_func": 0.5849019885063171,
+      "step": 651
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.9118881118881119,
+      "grad_norm": 5.241995408876847e-07,
+      "kl": 0.2119140625,
+      "learning_rate": 9.509373556515782e-05,
+      "loss": 0.0002,
+      "step": 652
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 87.41072082519531,
+      "epoch": 0.9132867132867133,
+      "grad_norm": 0.00010906936400705836,
+      "kl": 0.201171875,
+      "learning_rate": 9.506733911051502e-05,
+      "loss": 0.0002,
+      "num_tokens": 555810.0,
+      "reward": 2.1223483085632324,
+      "reward_std": 0.00045456583029590547,
+      "rewards/check_gptzero_func": 0.6428571343421936,
+      "rewards/check_perplexity_diff_func": 1.0,
+      "rewards/check_winston_local_func": 0.47949108481407166,
+      "step": 653
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.9146853146853147,
+      "grad_norm": 0.0001088298231648261,
+      "kl": 0.201171875,
+      "learning_rate": 9.50408755215632e-05,
+      "loss": 0.0002,
+      "step": 654
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 108.28572082519531,
+      "epoch": 0.916083916083916,
+      "grad_norm": 4.051397464310879e-07,
+      "kl": 0.16015625,
+      "learning_rate": 9.50143448377237e-05,
+      "loss": 0.0002,
+      "num_tokens": 576240.0,
+      "reward": 2.4261841773986816,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.9285714030265808,
+      "rewards/check_perplexity_diff_func": 0.9642857313156128,
+      "rewards/check_winston_local_func": 0.5333269238471985,
+      "step": 655
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.9174825174825175,
+      "grad_norm": 4.914401353891805e-07,
+      "kl": 0.16015625,
+      "learning_rate": 9.498774709851779e-05,
+      "loss": 0.0002,
+      "step": 656
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 103.35714721679688,
+      "epoch": 0.9188811188811189,
+      "grad_norm": 2.415453772573951e-07,
+      "kl": 0.1845703125,
+      "learning_rate": 9.496108234356666e-05,
+      "loss": 0.0002,
+      "num_tokens": 596032.0,
+      "reward": 2.0929291248321533,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.6428571343421936,
+      "rewards/check_perplexity_diff_func": 1.0,
+      "rewards/check_winston_local_func": 0.4500720500946045,
+      "step": 657
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.9202797202797203,
+      "grad_norm": 2.538209980883487e-07,
+      "kl": 0.1845703125,
+      "learning_rate": 9.493435061259131e-05,
+      "loss": 0.0002,
+      "step": 658
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 108.5714340209961,
+      "epoch": 0.9216783216783216,
+      "grad_norm": 0.0005472818191454855,
+      "kl": 0.15234375,
+      "learning_rate": 9.49075519454125e-05,
+      "loss": 0.0002,
+      "num_tokens": 616806.0,
+      "reward": 2.419811964035034,
+      "reward_std": 0.0021216755267232656,
+      "rewards/check_gptzero_func": 1.0,
+      "rewards/check_perplexity_diff_func": 0.9642857313156128,
+      "rewards/check_winston_local_func": 0.4555261731147766,
+      "step": 659
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.9230769230769231,
+      "grad_norm": 0.000555498769506101,
+      "kl": 0.15234375,
+      "learning_rate": 9.48806863819507e-05,
+      "loss": 0.0002,
+      "step": 660
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 99.50000762939453,
+      "epoch": 0.9244755244755245,
+      "grad_norm": 1.210593242712218e-05,
+      "kl": 0.185546875,
+      "learning_rate": 9.485375396222609e-05,
+      "loss": 0.0002,
+      "num_tokens": 636302.0,
+      "reward": 2.2581381797790527,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.7857142686843872,
+      "rewards/check_perplexity_diff_func": 0.9642857313156128,
+      "rewards/check_winston_local_func": 0.5081380605697632,
+      "step": 661
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.9258741258741259,
+      "grad_norm": 1.2079740491548147e-05,
+      "kl": 0.185546875,
+      "learning_rate": 9.482675472635837e-05,
+      "loss": 0.0002,
+      "step": 662
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 130.8928680419922,
+      "epoch": 0.9272727272727272,
+      "grad_norm": 6.087538896625874e-06,
+      "kl": 0.1640625,
+      "learning_rate": 9.479968871456679e-05,
+      "loss": 0.0002,
+      "num_tokens": 659494.0,
+      "reward": 2.4411208629608154,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 1.0,
+      "rewards/check_perplexity_diff_func": 0.9285714030265808,
+      "rewards/check_winston_local_func": 0.5125490427017212,
+      "step": 663
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.9286713286713286,
+      "grad_norm": 5.91002527093218e-06,
+      "kl": 0.1640625,
+      "learning_rate": 9.477255596717012e-05,
+      "loss": 0.0002,
+      "step": 664
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 106.78572082519531,
+      "epoch": 0.9300699300699301,
+      "grad_norm": 0.0006001697890594295,
+      "kl": 0.1572265625,
+      "learning_rate": 9.474535652458647e-05,
+      "loss": 0.0002,
+      "num_tokens": 679806.0,
+      "reward": 1.9322819709777832,
+      "reward_std": 0.0051432885229587555,
+      "rewards/check_gptzero_func": 0.5714285969734192,
+      "rewards/check_perplexity_diff_func": 0.9642857313156128,
+      "rewards/check_winston_local_func": 0.39656758308410645,
+      "step": 665
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.9314685314685315,
+      "grad_norm": 0.0006059012217339499,
+      "kl": 0.1572265625,
+      "learning_rate": 9.471809042733336e-05,
+      "loss": 0.0002,
+      "step": 666
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 93.46428680419922,
+      "epoch": 0.9328671328671329,
+      "grad_norm": 2.199744146458813e-06,
+      "kl": 0.197265625,
+      "learning_rate": 9.469075771602755e-05,
+      "loss": 0.0002,
+      "num_tokens": 698264.0,
+      "reward": 2.203263282775879,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.7142857313156128,
+      "rewards/check_perplexity_diff_func": 0.9642857313156128,
+      "rewards/check_winston_local_func": 0.524691641330719,
+      "step": 667
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.9342657342657342,
+      "grad_norm": 2.363657066653465e-06,
+      "kl": 0.197265625,
+      "learning_rate": 9.46633584313851e-05,
+      "loss": 0.0002,
+      "step": 668
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 121.28572082519531,
+      "epoch": 0.9356643356643357,
+      "grad_norm": 1.1463012847374795e-06,
+      "kl": 0.138671875,
+      "learning_rate": 9.463589261422116e-05,
+      "loss": 0.0001,
+      "num_tokens": 720488.0,
+      "reward": 2.178800582885742,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.8571428656578064,
+      "rewards/check_perplexity_diff_func": 0.9285714030265808,
+      "rewards/check_winston_local_func": 0.3930862247943878,
+      "step": 669
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.9370629370629371,
+      "grad_norm": 1.142739230420817e-06,
+      "kl": 0.138671875,
+      "learning_rate": 9.460836030545007e-05,
+      "loss": 0.0001,
+      "step": 670
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 110.00000762939453,
+      "epoch": 0.9384615384615385,
+      "grad_norm": 1.5883057563648785e-06,
+      "kl": 0.1748046875,
+      "learning_rate": 9.458076154608515e-05,
+      "loss": 0.0002,
+      "num_tokens": 741268.0,
+      "reward": 2.4054763317108154,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 1.0714285373687744,
+      "rewards/check_perplexity_diff_func": 0.8928571343421936,
+      "rewards/check_winston_local_func": 0.44119033217430115,
+      "step": 671
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.9398601398601398,
+      "grad_norm": 1.485387924830463e-06,
+      "kl": 0.1748046875,
+      "learning_rate": 9.455309637723875e-05,
+      "loss": 0.0002,
+      "step": 672
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 76.0,
+      "epoch": 0.9412587412587412,
+      "grad_norm": 5.476513075735238e-07,
+      "kl": 0.275390625,
+      "learning_rate": 9.452536484012212e-05,
+      "loss": 0.0003,
+      "num_tokens": 757900.0,
+      "reward": 2.1300675868988037,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.5714285969734192,
+      "rewards/check_perplexity_diff_func": 1.0,
+      "rewards/check_winston_local_func": 0.558638870716095,
+      "step": 673
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.9426573426573427,
+      "grad_norm": 5.390848738144119e-07,
+      "kl": 0.275390625,
+      "learning_rate": 9.44975669760454e-05,
+      "loss": 0.0003,
+      "step": 674
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 85.42857360839844,
+      "epoch": 0.9440559440559441,
+      "grad_norm": 0.09102872158611763,
+      "kl": 0.2197265625,
+      "learning_rate": 9.446970282641754e-05,
+      "loss": 0.0018,
+      "num_tokens": 775790.0,
+      "reward": 2.0888733863830566,
+      "reward_std": 0.06202990189194679,
+      "rewards/check_gptzero_func": 0.5357142686843872,
+      "rewards/check_perplexity_diff_func": 1.0,
+      "rewards/check_winston_local_func": 0.5531590580940247,
+      "step": 675
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.9454545454545454,
+      "grad_norm": 0.05645096724429603,
+      "kl": 0.220703125,
+      "learning_rate": 9.444177243274618e-05,
+      "loss": 0.0003,
+      "step": 676
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 109.25000762939453,
+      "epoch": 0.9468531468531468,
+      "grad_norm": 8.293674316919555e-06,
+      "kl": 0.1435546875,
+      "learning_rate": 9.441377583663768e-05,
+      "loss": 0.0001,
+      "num_tokens": 796550.0,
+      "reward": 2.231931209564209,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.8571428656578064,
+      "rewards/check_perplexity_diff_func": 0.9285714030265808,
+      "rewards/check_winston_local_func": 0.44621697068214417,
+      "step": 677
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.9482517482517483,
+      "grad_norm": 1.3063887978182609e-05,
+      "kl": 0.1435546875,
+      "learning_rate": 9.438571307979704e-05,
+      "loss": 0.0001,
+      "step": 678
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 116.10714721679688,
+      "epoch": 0.9496503496503497,
+      "grad_norm": 8.55167943307393e-05,
+      "kl": 0.171875,
+      "learning_rate": 9.435758420402778e-05,
+      "loss": 0.0002,
+      "num_tokens": 818208.0,
+      "reward": 2.5719223022460938,
+      "reward_std": 0.00022740935673937201,
+      "rewards/check_gptzero_func": 1.0714285373687744,
+      "rewards/check_perplexity_diff_func": 0.9642857313156128,
+      "rewards/check_winston_local_func": 0.5362080335617065,
+      "step": 679
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.951048951048951,
+      "grad_norm": 7.003771025309948e-05,
+      "kl": 0.171875,
+      "learning_rate": 9.43293892512319e-05,
+      "loss": 0.0002,
+      "step": 680
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 134.75,
+      "epoch": 0.9524475524475524,
+      "grad_norm": 0.000302463889717664,
+      "kl": 0.1435546875,
+      "learning_rate": 9.430112826340987e-05,
+      "loss": 0.0001,
+      "num_tokens": 841930.0,
+      "reward": 2.399691581726074,
+      "reward_std": 0.0017504910938441753,
+      "rewards/check_gptzero_func": 1.0714285373687744,
+      "rewards/check_perplexity_diff_func": 0.8928571343421936,
+      "rewards/check_winston_local_func": 0.4354057312011719,
+      "step": 681
+    },
+    {
+      "clip_ratio": 9.007386688608676e-05,
+      "epoch": 0.9538461538461539,
+      "grad_norm": 0.00031004535575776196,
+      "kl": 0.1435546875,
+      "learning_rate": 9.42728012826605e-05,
+      "loss": 0.0001,
+      "step": 682
+    },
+    {
+      "clip_ratio": 0.00011586143227759749,
+      "completion_length": 117.26786041259766,
+      "epoch": 0.9552447552447553,
+      "grad_norm": 0.0023255881114160465,
+      "kl": 0.146484375,
+      "learning_rate": 9.424440835118094e-05,
+      "loss": 0.0003,
+      "num_tokens": 863539.0,
+      "reward": 2.1542186737060547,
+      "reward_std": 0.02600925788283348,
+      "rewards/check_gptzero_func": 0.7857142686843872,
+      "rewards/check_perplexity_diff_func": 0.9464285969734192,
+      "rewards/check_winston_local_func": 0.42207565903663635,
+      "step": 683
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.9566433566433566,
+      "grad_norm": 0.003353044551999646,
+      "kl": 0.146484375,
+      "learning_rate": 9.421594951126653e-05,
+      "loss": 0.0003,
+      "step": 684
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 130.75,
+      "epoch": 0.958041958041958,
+      "grad_norm": 3.4853626956300013e-07,
+      "kl": 0.1376953125,
+      "learning_rate": 9.418742480531085e-05,
+      "loss": 0.0001,
+      "num_tokens": 886503.0,
+      "reward": 2.1976943016052246,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.7857142686843872,
+      "rewards/check_perplexity_diff_func": 0.9285714030265808,
+      "rewards/check_winston_local_func": 0.4834086000919342,
+      "step": 685
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.9594405594405594,
+      "grad_norm": 4.041143328184125e-07,
+      "kl": 0.1376953125,
+      "learning_rate": 9.415883427580556e-05,
+      "loss": 0.0001,
+      "step": 686
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 88.14286041259766,
+      "epoch": 0.9608391608391609,
+      "grad_norm": 0.0002925937022236145,
+      "kl": 0.1884765625,
+      "learning_rate": 9.413017796534038e-05,
+      "loss": 0.0002,
+      "num_tokens": 904263.0,
+      "reward": 2.0918164253234863,
+      "reward_std": 0.0009377764072269201,
+      "rewards/check_gptzero_func": 0.6428571343421936,
+      "rewards/check_perplexity_diff_func": 1.0,
+      "rewards/check_winston_local_func": 0.448959082365036,
+      "step": 687
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.9622377622377623,
+      "grad_norm": 0.00027650769429382713,
+      "kl": 0.1884765625,
+      "learning_rate": 9.410145591660301e-05,
+      "loss": 0.0002,
+      "step": 688
+    },
+    {
+      "clip_ratio": 0.00036945813917554915,
+      "completion_length": 111.60714721679688,
+      "epoch": 0.9636363636363636,
+      "grad_norm": 0.00019032746086675872,
+      "kl": 0.154296875,
+      "learning_rate": 9.407266817237911e-05,
+      "loss": 0.0002,
+      "num_tokens": 925343.0,
+      "reward": 2.2334210872650146,
+      "reward_std": 0.00015559412713628262,
+      "rewards/check_gptzero_func": 0.7142857313156128,
+      "rewards/check_perplexity_diff_func": 0.9642857313156128,
+      "rewards/check_winston_local_func": 0.5548495650291443,
+      "step": 689
+    },
+    {
+      "clip_ratio": 0.00024630542611703277,
+      "epoch": 0.965034965034965,
+      "grad_norm": 0.00013578009050933864,
+      "kl": 0.1533203125,
+      "learning_rate": 9.404381477555217e-05,
+      "loss": 0.0002,
+      "step": 690
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 113.14286041259766,
+      "epoch": 0.9664335664335665,
+      "grad_norm": 1.7240378740659336e-05,
+      "kl": 0.16015625,
+      "learning_rate": 9.401489576910349e-05,
+      "loss": 0.0002,
+      "num_tokens": 946805.0,
+      "reward": 2.417461395263672,
+      "reward_std": 0.00010864253272302449,
+      "rewards/check_gptzero_func": 0.9285714030265808,
+      "rewards/check_perplexity_diff_func": 0.9642857313156128,
+      "rewards/check_winston_local_func": 0.5246042609214783,
+      "step": 691
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.9678321678321679,
+      "grad_norm": 1.6290593947397482e-05,
+      "kl": 0.16015625,
+      "learning_rate": 9.398591119611211e-05,
+      "loss": 0.0002,
+      "step": 692
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 89.4464340209961,
+      "epoch": 0.9692307692307692,
+      "grad_norm": 0.0003060987829802528,
+      "kl": 0.19140625,
+      "learning_rate": 9.395686109975474e-05,
+      "loss": 0.0002,
+      "num_tokens": 965064.0,
+      "reward": 2.075838804244995,
+      "reward_std": 0.0029447702690958977,
+      "rewards/check_gptzero_func": 0.6428571343421936,
+      "rewards/check_perplexity_diff_func": 1.0,
+      "rewards/check_winston_local_func": 0.43298134207725525,
+      "step": 693
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.9706293706293706,
+      "grad_norm": 0.001874059602241207,
+      "kl": 0.19140625,
+      "learning_rate": 9.392774552330567e-05,
+      "loss": 0.0002,
+      "step": 694
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 126.10714721679688,
+      "epoch": 0.972027972027972,
+      "grad_norm": 8.509194894061606e-06,
+      "kl": 0.1640625,
+      "learning_rate": 9.38985645101368e-05,
+      "loss": 0.0002,
+      "num_tokens": 987280.0,
+      "reward": 2.1274285316467285,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.7857142686843872,
+      "rewards/check_perplexity_diff_func": 0.9285714030265808,
+      "rewards/check_winston_local_func": 0.41314274072647095,
+      "step": 695
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.9734265734265735,
+      "grad_norm": 8.720610779885554e-06,
+      "kl": 0.1640625,
+      "learning_rate": 9.386931810371742e-05,
+      "loss": 0.0002,
+      "step": 696
+    },
+    {
+      "clip_ratio": 0.00018198363250121474,
+      "completion_length": 104.37500762939453,
+      "epoch": 0.9748251748251748,
+      "grad_norm": 0.006044271030051565,
+      "kl": 0.162109375,
+      "learning_rate": 9.38400063476143e-05,
+      "loss": 0.0001,
+      "num_tokens": 1007329.0,
+      "reward": 2.1394481658935547,
+      "reward_std": 0.008249713107943535,
+      "rewards/check_gptzero_func": 0.7857142686843872,
+      "rewards/check_perplexity_diff_func": 0.9642857313156128,
+      "rewards/check_winston_local_func": 0.38944822549819946,
+      "step": 697
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.9762237762237762,
+      "grad_norm": 0.0001835404667097808,
+      "kl": 0.162109375,
+      "learning_rate": 9.381062928549153e-05,
+      "loss": 0.0001,
+      "step": 698
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 121.28572082519531,
+      "epoch": 0.9776223776223776,
+      "grad_norm": 2.5914502760266237e-06,
+      "kl": 0.1494140625,
+      "learning_rate": 9.378118696111047e-05,
+      "loss": 0.0001,
+      "num_tokens": 1029293.0,
+      "reward": 2.0580966472625732,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.6428571343421936,
+      "rewards/check_perplexity_diff_func": 1.0,
+      "rewards/check_winston_local_func": 0.4152393937110901,
+      "step": 699
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.9790209790209791,
+      "grad_norm": 2.773516504511405e-06,
+      "kl": 0.1494140625,
+      "learning_rate": 9.375167941832973e-05,
+      "loss": 0.0001,
+      "step": 700
+    },
+    {
+      "clip_ratio": 0.00013789300282951444,
+      "completion_length": 93.75000762939453,
+      "epoch": 0.9804195804195804,
+      "grad_norm": 0.013670676452353387,
+      "kl": 0.2041015625,
+      "learning_rate": 9.372210670110505e-05,
+      "loss": -0.0004,
+      "num_tokens": 1048027.0,
+      "reward": 2.084160327911377,
+      "reward_std": 0.011760223656892776,
+      "rewards/check_gptzero_func": 0.6428571343421936,
+      "rewards/check_perplexity_diff_func": 0.9464285969734192,
+      "rewards/check_winston_local_func": 0.49487441778182983,
+      "step": 701
+    },
+    {
+      "clip_ratio": 0.00013789300282951444,
+      "epoch": 0.9818181818181818,
+      "grad_norm": 0.01262284842044345,
+      "kl": 0.2041015625,
+      "learning_rate": 9.369246885348926e-05,
+      "loss": -0.0005,
+      "step": 702
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 113.28572082519531,
+      "epoch": 0.9832167832167832,
+      "grad_norm": 2.9830896158632583e-05,
+      "kl": 0.15625,
+      "learning_rate": 9.366276591963221e-05,
+      "loss": 0.0002,
+      "num_tokens": 1068897.0,
+      "reward": 2.2253122329711914,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.7857142686843872,
+      "rewards/check_perplexity_diff_func": 1.0,
+      "rewards/check_winston_local_func": 0.4395977854728699,
+      "step": 703
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.9846153846153847,
+      "grad_norm": 2.8538602263117158e-05,
+      "kl": 0.15625,
+      "learning_rate": 9.363299794378073e-05,
+      "loss": 0.0002,
+      "step": 704
+    },
+    {
+      "clip_ratio": 0.00022251892369240522,
+      "completion_length": 112.42857360839844,
+      "epoch": 0.986013986013986,
+      "grad_norm": 0.013827349944974443,
+      "kl": 0.1552734375,
+      "learning_rate": 9.360316497027848e-05,
+      "loss": 0.0006,
+      "num_tokens": 1089715.0,
+      "reward": 2.2796030044555664,
+      "reward_std": 0.036380112171173096,
+      "rewards/check_gptzero_func": 0.7857142686843872,
+      "rewards/check_perplexity_diff_func": 0.9642857313156128,
+      "rewards/check_winston_local_func": 0.529603123664856,
+      "step": 705
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.9874125874125874,
+      "grad_norm": 0.013400518475433424,
+      "kl": 0.1552734375,
+      "learning_rate": 9.357326704356602e-05,
+      "loss": 0.0005,
+      "step": 706
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 96.26786041259766,
+      "epoch": 0.9888111888111888,
+      "grad_norm": 0.013500774785978116,
+      "kl": 0.19140625,
+      "learning_rate": 9.35433042081806e-05,
+      "loss": 0.0003,
+      "num_tokens": 1108898.0,
+      "reward": 2.1743431091308594,
+      "reward_std": 0.025075625628232956,
+      "rewards/check_gptzero_func": 0.7857142686843872,
+      "rewards/check_perplexity_diff_func": 0.9464285969734192,
+      "rewards/check_winston_local_func": 0.442200243473053,
+      "step": 707
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.9902097902097902,
+      "grad_norm": 0.004775981640901926,
+      "kl": 0.1904296875,
+      "learning_rate": 9.351327650875624e-05,
+      "loss": 0.0002,
+      "step": 708
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 106.53572082519531,
+      "epoch": 0.9916083916083916,
+      "grad_norm": 5.729650858248644e-07,
+      "kl": 0.1650390625,
+      "learning_rate": 9.348318399002347e-05,
+      "loss": 0.0002,
+      "num_tokens": 1129126.0,
+      "reward": 2.423046588897705,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.9285714030265808,
+      "rewards/check_perplexity_diff_func": 1.0,
+      "rewards/check_winston_local_func": 0.49447494745254517,
+      "step": 709
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.993006993006993,
+      "grad_norm": 5.262172579879396e-07,
+      "kl": 0.1650390625,
+      "learning_rate": 9.345302669680946e-05,
+      "loss": 0.0002,
+      "step": 710
+    },
+    {
+      "clip_ratio": 0.00021829926117789,
+      "completion_length": 120.46429443359375,
+      "epoch": 0.9944055944055944,
+      "grad_norm": 0.05104704112053445,
+      "kl": 0.17578125,
+      "learning_rate": 9.342280467403788e-05,
+      "loss": -0.0017,
+      "num_tokens": 1151220.0,
+      "reward": 2.2311315536499023,
+      "reward_std": 0.04048382490873337,
+      "rewards/check_gptzero_func": 0.8214285969734192,
+      "rewards/check_perplexity_diff_func": 0.9464285969734192,
+      "rewards/check_winston_local_func": 0.4632743299007416,
+      "step": 711
+    },
+    {
+      "clip_ratio": 0.00010004002251662314,
+      "epoch": 0.9958041958041958,
+      "grad_norm": 0.03334333430793214,
+      "kl": 0.169921875,
+      "learning_rate": 9.339251796672877e-05,
+      "loss": -0.0022,
+      "step": 712
+    },
+    {
+      "clip_ratio": 0.00011548677866812795,
+      "completion_length": 106.26786041259766,
+      "epoch": 0.9972027972027973,
+      "grad_norm": 8.312891203940763e-05,
+      "kl": 0.15234375,
+      "learning_rate": 9.336216661999857e-05,
+      "loss": 0.0002,
+      "num_tokens": 1171757.0,
+      "reward": 2.4376866817474365,
+      "reward_std": 0.00044757546856999397,
+      "rewards/check_gptzero_func": 0.9285714030265808,
+      "rewards/check_perplexity_diff_func": 0.9642857313156128,
+      "rewards/check_winston_local_func": 0.5448295474052429,
+      "step": 713
+    },
+    {
+      "clip_ratio": 0.00011548677866812795,
+      "epoch": 0.9986013986013986,
+      "grad_norm": 7.041724475492408e-05,
+      "kl": 0.15234375,
+      "learning_rate": 9.333175067905997e-05,
+      "loss": 0.0002,
+      "step": 714
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 127.00000762939453,
+      "epoch": 1.0013986013986014,
+      "grad_norm": 7.04944892486657e-07,
+      "kl": 0.142578125,
+      "learning_rate": 9.330127018922194e-05,
+      "loss": 0.0001,
+      "num_tokens": 1194243.0,
+      "reward": 1.8460028171539307,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.5,
+      "rewards/check_perplexity_diff_func": 1.0,
+      "rewards/check_winston_local_func": 0.34600263833999634,
+      "step": 715
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 1.0027972027972027,
+      "grad_norm": 6.628698629011792e-07,
+      "kl": 0.142578125,
+      "learning_rate": 9.327072519588954e-05,
+      "loss": 0.0001,
+      "step": 716
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 94.66072082519531,
+      "epoch": 1.0041958041958041,
+      "grad_norm": 0.00044081574153375096,
+      "kl": 0.2099609375,
+      "learning_rate": 9.324011574456397e-05,
+      "loss": 0.0002,
+      "num_tokens": 1213238.0,
+      "reward": 2.0919501781463623,
+      "reward_std": 0.00025218268274329603,
+      "rewards/check_gptzero_func": 0.5714285969734192,
+      "rewards/check_perplexity_diff_func": 0.9285714030265808,
+      "rewards/check_winston_local_func": 0.5919499397277832,
+      "step": 717
+    },
+    {
+      "clip_ratio": 0.00019868866365868598,
+      "epoch": 1.0055944055944055,
+      "grad_norm": 0.000549077840173661,
+      "kl": 0.2109375,
+      "learning_rate": 9.320944188084242e-05,
+      "loss": 0.0002,
+      "step": 718
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 84.85714721679688,
+      "epoch": 1.006993006993007,
+      "grad_norm": 0.005163027828412511,
+      "kl": 0.1884765625,
+      "learning_rate": 9.317870365041804e-05,
+      "loss": 0.0002,
+      "num_tokens": 1230630.0,
+      "reward": 2.2233378887176514,
+      "reward_std": 0.02511097863316536,
+      "rewards/check_gptzero_func": 0.7857142686843872,
+      "rewards/check_perplexity_diff_func": 0.9821428656578064,
+      "rewards/check_winston_local_func": 0.45548057556152344,
+      "step": 719
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 1.0083916083916085,
+      "grad_norm": 0.00444116748229232,
+      "kl": 0.1884765625,
+      "learning_rate": 9.314790109907984e-05,
+      "loss": 0.0002,
+      "step": 720
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 80.75,
+      "epoch": 1.0097902097902098,
+      "grad_norm": 8.668316800887643e-07,
+      "kl": 0.216796875,
+      "learning_rate": 9.31170342727127e-05,
+      "loss": 0.0002,
+      "num_tokens": 1248084.0,
+      "reward": 2.440389394760132,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.9285714030265808,
+      "rewards/check_perplexity_diff_func": 1.0,
+      "rewards/check_winston_local_func": 0.5118179321289062,
+      "step": 721
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 1.0111888111888112,
+      "grad_norm": 7.749843514631877e-07,
+      "kl": 0.216796875,
+      "learning_rate": 9.30861032172972e-05,
+      "loss": 0.0002,
+      "step": 722
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 109.53572082519531,
+      "epoch": 1.0125874125874126,
+      "grad_norm": 1.4806859938034535e-05,
+      "kl": 0.19140625,
+      "learning_rate": 9.305510797890962e-05,
+      "loss": 0.0002,
+      "num_tokens": 1269030.0,
+      "reward": 2.161181688308716,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.7142857313156128,
+      "rewards/check_perplexity_diff_func": 1.0,
+      "rewards/check_winston_local_func": 0.44689592719078064,
+      "step": 723
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 1.013986013986014,
+      "grad_norm": 1.600117516634148e-05,
+      "kl": 0.19140625,
+      "learning_rate": 9.302404860372185e-05,
+      "loss": 0.0002,
+      "step": 724
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 105.42857360839844,
+      "epoch": 1.0153846153846153,
+      "grad_norm": 0.15558383270036835,
+      "kl": 0.21484375,
+      "learning_rate": 9.299292513800129e-05,
+      "loss": 0.05,
+      "num_tokens": 1289470.0,
+      "reward": 2.1427464485168457,
+      "reward_std": 0.07507836818695068,
+      "rewards/check_gptzero_func": 0.6785714030265808,
+      "rewards/check_perplexity_diff_func": 0.9285714030265808,
+      "rewards/check_winston_local_func": 0.535603404045105,
+      "step": 725
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 1.0167832167832167,
+      "grad_norm": 0.06803559596877588,
+      "kl": 0.2353515625,
+      "learning_rate": 9.296173762811085e-05,
+      "loss": 0.0475,
+      "step": 726
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 117.10714721679688,
+      "epoch": 1.018181818181818,
+      "grad_norm": 1.6687299566663977e-05,
+      "kl": 0.14453125,
+      "learning_rate": 9.293048612050883e-05,
+      "loss": 0.0001,
+      "num_tokens": 1311104.0,
+      "reward": 1.9664360284805298,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.7142857313156128,
+      "rewards/check_perplexity_diff_func": 0.8928571343421936,
+      "rewards/check_winston_local_func": 0.35929304361343384,
+      "step": 727
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 1.0195804195804197,
+      "grad_norm": 1.8557887240310947e-05,
+      "kl": 0.14453125,
+      "learning_rate": 9.289917066174886e-05,
+      "loss": 0.0001,
+      "step": 728
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 93.42857360839844,
+      "epoch": 1.020979020979021,
+      "grad_norm": 1.0271883224976578e-06,
+      "kl": 0.2314453125,
+      "learning_rate": 9.286779129847984e-05,
+      "loss": 0.0002,
+      "num_tokens": 1329690.0,
+      "reward": 2.0715172290802,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.5714285969734192,
+      "rewards/check_perplexity_diff_func": 1.0,
+      "rewards/check_winston_local_func": 0.5000885725021362,
+      "step": 729
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 1.0223776223776224,
+      "grad_norm": 1.0026902118997668e-06,
+      "kl": 0.2314453125,
+      "learning_rate": 9.283634807744586e-05,
+      "loss": 0.0002,
+      "step": 730
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 127.00000762939453,
+      "epoch": 1.0237762237762238,
+      "grad_norm": 1.674477493760208e-06,
+      "kl": 0.169921875,
+      "learning_rate": 9.280484104548615e-05,
+      "loss": 0.0002,
+      "num_tokens": 1352330.0,
+      "reward": 2.3377161026000977,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 1.0,
+      "rewards/check_perplexity_diff_func": 0.8928571343421936,
+      "rewards/check_winston_local_func": 0.4448588192462921,
+      "step": 731
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 1.0251748251748252,
+      "grad_norm": 1.927135056596767e-06,
+      "kl": 0.169921875,
+      "learning_rate": 9.277327024953496e-05,
+      "loss": 0.0002,
+      "step": 732
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 96.10714721679688,
+      "epoch": 1.0265734265734265,
+      "grad_norm": 1.21032862101872e-06,
+      "kl": 0.1689453125,
+      "learning_rate": 9.274163573662157e-05,
+      "loss": 0.0002,
+      "num_tokens": 1371384.0,
+      "reward": 2.437558889389038,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.9285714030265808,
+      "rewards/check_perplexity_diff_func": 1.0,
+      "rewards/check_winston_local_func": 0.508987307548523,
+      "step": 733
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 1.027972027972028,
+      "grad_norm": 1.7462301887618669e-06,
+      "kl": 0.1689453125,
+      "learning_rate": 9.270993755387015e-05,
+      "loss": 0.0002,
+      "step": 734
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 89.78572082519531,
+      "epoch": 1.0293706293706293,
+      "grad_norm": 0.0004304814492353597,
+      "kl": 0.2177734375,
+      "learning_rate": 9.26781757484997e-05,
+      "loss": 0.0002,
+      "num_tokens": 1389928.0,
+      "reward": 2.4899582862854004,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.8571428656578064,
+      "rewards/check_perplexity_diff_func": 1.0,
+      "rewards/check_winston_local_func": 0.6328152418136597,
+      "step": 735
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 1.0307692307692307,
+      "grad_norm": 0.00046915841230749046,
+      "kl": 0.2177734375,
+      "learning_rate": 9.264635036782405e-05,
+      "loss": 0.0002,
+      "step": 736
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 117.4464340209961,
+      "epoch": 1.0321678321678323,
+      "grad_norm": 0.003036344469002795,
+      "kl": 0.142578125,
+      "learning_rate": 9.261446145925167e-05,
+      "loss": -0.0041,
+      "num_tokens": 1411475.0,
+      "reward": 2.17564058303833,
+      "reward_std": 0.030071333050727844,
+      "rewards/check_gptzero_func": 0.8571428656578064,
+      "rewards/check_perplexity_diff_func": 0.9821428656578064,
+      "rewards/check_winston_local_func": 0.33635467290878296,
+      "step": 737
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 1.0335664335664336,
+      "grad_norm": 0.0027126180636015186,
+      "kl": 0.1435546875,
+      "learning_rate": 9.258250907028572e-05,
+      "loss": -0.0041,
+      "step": 738
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 104.73214721679688,
+      "epoch": 1.034965034965035,
+      "grad_norm": 0.0544058478741893,
+      "kl": 0.2080078125,
+      "learning_rate": 9.255049324852388e-05,
+      "loss": 0.0002,
+      "num_tokens": 1431764.0,
+      "reward": 2.090238571166992,
+      "reward_std": 0.08559178560972214,
+      "rewards/check_gptzero_func": 0.6785714030265808,
+      "rewards/check_perplexity_diff_func": 0.9821428656578064,
+      "rewards/check_winston_local_func": 0.42952439188957214,
+      "step": 739
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 1.0363636363636364,
+      "grad_norm": 0.006960547191239059,
+      "kl": 0.205078125,
+      "learning_rate": 9.251841404165835e-05,
+      "loss": 0.0001,
+      "step": 740
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 88.33928680419922,
+      "epoch": 1.0377622377622377,
+      "grad_norm": 0.00033693931372504076,
+      "kl": 0.185546875,
+      "learning_rate": 9.248627149747573e-05,
+      "loss": 0.0002,
+      "num_tokens": 1449951.0,
+      "reward": 2.0378341674804688,
+      "reward_std": 0.0007935017929412425,
+      "rewards/check_gptzero_func": 0.6428571343421936,
+      "rewards/check_perplexity_diff_func": 0.9285714030265808,
+      "rewards/check_winston_local_func": 0.4664055407047272,
+      "step": 741
+    },
+    {
+      "clip_ratio": 0.00013215278158895671,
+      "epoch": 1.0391608391608391,
+      "grad_norm": 0.00034490140909119695,
+      "kl": 0.185546875,
+      "learning_rate": 9.245406566385697e-05,
+      "loss": 0.0002,
+      "step": 742
+    },
+    {
+      "clip_ratio": 0.00015926103515084833,
+      "completion_length": 107.80357360839844,
+      "epoch": 1.0405594405594405,
+      "grad_norm": 0.013332167554109667,
+      "kl": 0.1865234375,
+      "learning_rate": 9.242179658877734e-05,
+      "loss": 0.0001,
+      "num_tokens": 1470538.0,
+      "reward": 2.4016566276550293,
+      "reward_std": 0.02674621157348156,
+      "rewards/check_gptzero_func": 0.8571428656578064,
+      "rewards/check_perplexity_diff_func": 0.9821428656578064,
+      "rewards/check_winston_local_func": 0.5623708963394165,
+      "step": 743
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 1.0419580419580419,
+      "grad_norm": 0.0029520677726348676,
+      "kl": 0.2001953125,
+      "learning_rate": 9.238946432030626e-05,
+      "loss": 0.0001,
+      "step": 744
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 83.35714721679688,
+      "epoch": 1.0433566433566432,
+      "grad_norm": 0.0005995187289984499,
+      "kl": 0.2255859375,
+      "learning_rate": 9.235706890660733e-05,
+      "loss": 0.0002,
+      "num_tokens": 1487784.0,
+      "reward": 2.067596435546875,
+      "reward_std": 0.00029761545010842383,
+      "rewards/check_gptzero_func": 0.5714285969734192,
+      "rewards/check_perplexity_diff_func": 0.9642857313156128,
+      "rewards/check_winston_local_func": 0.5318821668624878,
+      "step": 745
+    },
+    {
+      "clip_ratio": 0.0002088554756483063,
+      "epoch": 1.0447552447552448,
+      "grad_norm": 0.0005979581476299124,
+      "kl": 0.2265625,
+      "learning_rate": 9.232461039593819e-05,
+      "loss": 0.0002,
+      "step": 746
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 122.33929443359375,
+      "epoch": 1.0461538461538462,
+      "grad_norm": 1.3373485214002519e-05,
+      "kl": 0.1455078125,
+      "learning_rate": 9.229208883665047e-05,
+      "loss": 0.0001,
+      "num_tokens": 1510355.0,
+      "reward": 1.986084222793579,
+      "reward_std": 0.024681033566594124,
+      "rewards/check_gptzero_func": 0.7142857313156128,
+      "rewards/check_perplexity_diff_func": 0.8392857313156128,
+      "rewards/check_winston_local_func": 0.43251270055770874,
+      "step": 747
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 1.0475524475524476,
+      "grad_norm": 0.005145688764176549,
+      "kl": 0.1455078125,
+      "learning_rate": 9.225950427718975e-05,
+      "loss": 0.0001,
+      "step": 748
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 118.75000762939453,
+      "epoch": 1.048951048951049,
+      "grad_norm": 2.1467569659911603e-06,
+      "kl": 0.142578125,
+      "learning_rate": 9.22268567660954e-05,
+      "loss": 0.0001,
+      "num_tokens": 1532269.0,
+      "reward": 2.1966254711151123,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.7857142686843872,
+      "rewards/check_perplexity_diff_func": 1.0,
+      "rewards/check_winston_local_func": 0.4109109342098236,
+      "step": 749
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 1.0503496503496503,
+      "grad_norm": 2.040002552625244e-06,
+      "kl": 0.142578125,
+      "learning_rate": 9.219414635200065e-05,
+      "loss": 0.0001,
+      "step": 750
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 103.46428680419922,
+      "epoch": 1.0517482517482517,
+      "grad_norm": 0.0015423150610452264,
+      "kl": 0.2177734375,
+      "learning_rate": 9.216137308363233e-05,
+      "loss": 0.0002,
+      "num_tokens": 1552375.0,
+      "reward": 2.8162217140197754,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 1.3571428060531616,
+      "rewards/check_perplexity_diff_func": 1.0,
+      "rewards/check_winston_local_func": 0.45907872915267944,
+      "step": 751
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 1.053146853146853,
+      "grad_norm": 0.000923513262354566,
+      "kl": 0.2158203125,
+      "learning_rate": 9.2128537009811e-05,
+      "loss": 0.0002,
+      "step": 752
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 92.39286041259766,
+      "epoch": 1.0545454545454545,
+      "grad_norm": 1.100033680537667e-06,
+      "kl": 0.1689453125,
+      "learning_rate": 9.209563817945072e-05,
+      "loss": 0.0002,
+      "num_tokens": 1570845.0,
+      "reward": 1.9652879238128662,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.5714285969734192,
+      "rewards/check_perplexity_diff_func": 1.0,
+      "rewards/check_winston_local_func": 0.39385929703712463,
+      "step": 753
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 1.055944055944056,
+      "grad_norm": 1.1228592596360449e-06,
+      "kl": 0.1689453125,
+      "learning_rate": 9.206267664155907e-05,
+      "loss": 0.0002,
+      "step": 754
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 156.42857360839844,
+      "epoch": 1.0573426573426574,
+      "grad_norm": 2.057149136687796e-07,
+      "kl": 0.1171875,
+      "learning_rate": 9.202965244523695e-05,
+      "loss": 0.0001,
+      "num_tokens": 1596971.0,
+      "reward": 2.463409185409546,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 1.0714285373687744,
+      "rewards/check_perplexity_diff_func": 0.9642857313156128,
+      "rewards/check_winston_local_func": 0.4276948571205139,
+      "step": 755
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 1.0587412587412588,
+      "grad_norm": 1.931817225075422e-07,
+      "kl": 0.1171875,
+      "learning_rate": 9.199656563967875e-05,
+      "loss": 0.0001,
+      "step": 756
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 133.60714721679688,
+      "epoch": 1.0601398601398602,
+      "grad_norm": 7.066669417456737e-05,
+      "kl": 0.126953125,
+      "learning_rate": 9.1963416274172e-05,
+      "loss": 0.0001,
+      "num_tokens": 1621103.0,
+      "reward": 2.4794726371765137,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 1.0,
+      "rewards/check_perplexity_diff_func": 0.9285714030265808,
+      "rewards/check_winston_local_func": 0.5509011149406433,
+      "step": 757
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 1.0615384615384615,
+      "grad_norm": 5.5301588065085076e-05,
+      "kl": 0.126953125,
+      "learning_rate": 9.193020439809747e-05,
+      "loss": 0.0001,
+      "step": 758
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 128.57144165039062,
+      "epoch": 1.062937062937063,
+      "grad_norm": 1.0710398657530842e-06,
+      "kl": 0.1494140625,
+      "learning_rate": 9.189693006092907e-05,
+      "loss": 0.0001,
+      "num_tokens": 1644239.0,
+      "reward": 2.3191192150115967,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.9285714030265808,
+      "rewards/check_perplexity_diff_func": 0.9642857313156128,
+      "rewards/check_winston_local_func": 0.426261842250824,
+      "step": 759
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 1.0643356643356643,
+      "grad_norm": 1.0634008914464338e-06,
+      "kl": 0.1494140625,
+      "learning_rate": 9.186359331223369e-05,
+      "loss": 0.0001,
+      "step": 760
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 101.53572082519531,
+      "epoch": 1.0657342657342657,
+      "grad_norm": 1.016780218352338e-06,
+      "kl": 0.1728515625,
+      "learning_rate": 9.183019420167126e-05,
+      "loss": 0.0002,
+      "num_tokens": 1664379.0,
+      "reward": 2.527435779571533,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 1.0,
+      "rewards/check_perplexity_diff_func": 0.9642857313156128,
+      "rewards/check_winston_local_func": 0.5631500482559204,
+      "step": 761
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 1.067132867132867,
+      "grad_norm": 9.957050622798446e-07,
+      "kl": 0.1728515625,
+      "learning_rate": 9.179673277899456e-05,
+      "loss": 0.0002,
+      "step": 762
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 73.91072082519531,
+      "epoch": 1.0685314685314686,
+      "grad_norm": 0.003192954717479896,
+      "kl": 0.251953125,
+      "learning_rate": 9.176320909404924e-05,
+      "loss": 0.0002,
+      "num_tokens": 1680674.0,
+      "reward": 1.9061808586120605,
+      "reward_std": 0.007731279823929071,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 1.0,
+      "rewards/check_winston_local_func": 0.4776093661785126,
+      "step": 763
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 1.06993006993007,
+      "grad_norm": 0.0032145560494452733,
+      "kl": 0.251953125,
+      "learning_rate": 9.172962319677363e-05,
+      "loss": 0.0002,
+      "step": 764
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 114.78572082519531,
+      "epoch": 1.0713286713286714,
+      "grad_norm": 1.061638596684205e-06,
+      "kl": 0.15625,
+      "learning_rate": 9.169597513719881e-05,
+      "loss": 0.0002,
+      "num_tokens": 1702162.0,
+      "reward": 2.3026795387268066,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.8571428656578064,
+      "rewards/check_perplexity_diff_func": 0.9642857313156128,
+      "rewards/check_winston_local_func": 0.4812507629394531,
+      "step": 765
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 1.0727272727272728,
+      "grad_norm": 1.172228963353414e-06,
+      "kl": 0.15625,
+      "learning_rate": 9.166226496544839e-05,
+      "loss": 0.0002,
+      "step": 766
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 93.03572082519531,
+      "epoch": 1.0741258741258741,
+      "grad_norm": 2.1039959611311484e-05,
+      "kl": 0.1611328125,
+      "learning_rate": 9.162849273173857e-05,
+      "loss": 0.0002,
+      "num_tokens": 1721000.0,
+      "reward": 2.4015400409698486,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.8571428656578064,
+      "rewards/check_perplexity_diff_func": 1.0,
+      "rewards/check_winston_local_func": 0.5443968772888184,
+      "step": 767
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 1.0755244755244755,
+      "grad_norm": 1.7217131059909884e-05,
+      "kl": 0.1611328125,
+      "learning_rate": 9.159465848637795e-05,
+      "loss": 0.0002,
+      "step": 768
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 116.53572082519531,
+      "epoch": 1.0769230769230769,
+      "grad_norm": 2.8854421633379664e-07,
+      "kl": 0.15625,
+      "learning_rate": 9.156076227976752e-05,
+      "loss": 0.0002,
+      "num_tokens": 1742074.0,
+      "reward": 2.019984006881714,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.6428571343421936,
+      "rewards/check_perplexity_diff_func": 1.0,
+      "rewards/check_winston_local_func": 0.3771267533302307,
+      "step": 769
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 1.0783216783216782,
+      "grad_norm": 2.958786538763937e-07,
+      "kl": 0.15625,
+      "learning_rate": 9.152680416240059e-05,
+      "loss": 0.0002,
+      "step": 770
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 90.17857360839844,
+      "epoch": 1.0797202797202796,
+      "grad_norm": 1.3818538255152332e-07,
+      "kl": 0.2119140625,
+      "learning_rate": 9.149278418486267e-05,
+      "loss": 0.0002,
+      "num_tokens": 1760270.0,
+      "reward": 2.354346752166748,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 1.0,
+      "rewards/check_perplexity_diff_func": 0.9642857313156128,
+      "rewards/check_winston_local_func": 0.3900607228279114,
+      "step": 771
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 1.0811188811188812,
+      "grad_norm": 1.3653596591447927e-07,
+      "kl": 0.2119140625,
+      "learning_rate": 9.145870239783142e-05,
+      "loss": 0.0002,
+      "step": 772
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 113.96429443359375,
+      "epoch": 1.0825174825174826,
+      "grad_norm": 5.440872620013143e-05,
+      "kl": 0.138671875,
+      "learning_rate": 9.142455885207659e-05,
+      "loss": 0.0001,
+      "num_tokens": 1781844.0,
+      "reward": 1.8641180992126465,
+      "reward_std": 0.0003365546581335366,
+      "rewards/check_gptzero_func": 0.5,
+      "rewards/check_perplexity_diff_func": 0.9285714030265808,
+      "rewards/check_winston_local_func": 0.4355466663837433,
+      "step": 773
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 1.083916083916084,
+      "grad_norm": 5.372773059582913e-05,
+      "kl": 0.138671875,
+      "learning_rate": 9.139035359845994e-05,
+      "loss": 0.0001,
+      "step": 774
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 86.67857360839844,
+      "epoch": 1.0853146853146853,
+      "grad_norm": 2.3746587653962933e-07,
+      "kl": 0.1748046875,
+      "learning_rate": 9.135608668793511e-05,
+      "loss": 0.0002,
+      "num_tokens": 1799902.0,
+      "reward": 2.2892978191375732,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.7857142686843872,
+      "rewards/check_perplexity_diff_func": 1.0,
+      "rewards/check_winston_local_func": 0.5035834908485413,
+      "step": 775
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 1.0867132867132867,
+      "grad_norm": 2.3654979764665925e-07,
+      "kl": 0.1748046875,
+      "learning_rate": 9.132175817154763e-05,
+      "loss": 0.0002,
+      "step": 776
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 106.42857360839844,
+      "epoch": 1.088111888111888,
+      "grad_norm": 1.2723426089566611e-05,
+      "kl": 0.1728515625,
+      "learning_rate": 9.128736810043474e-05,
+      "loss": 0.0002,
+      "num_tokens": 1820052.0,
+      "reward": 2.0230119228363037,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.6428571343421936,
+      "rewards/check_perplexity_diff_func": 0.9642857313156128,
+      "rewards/check_winston_local_func": 0.4158690869808197,
+      "step": 777
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 1.0895104895104895,
+      "grad_norm": 1.0458212078617634e-05,
+      "kl": 0.1728515625,
+      "learning_rate": 9.125291652582549e-05,
+      "loss": 0.0002,
+      "step": 778
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 116.71429443359375,
+      "epoch": 1.0909090909090908,
+      "grad_norm": 3.995811344230737e-07,
+      "kl": 0.189453125,
+      "learning_rate": 9.121840349904039e-05,
+      "loss": 0.0002,
+      "num_tokens": 1841842.0,
+      "reward": 2.676417112350464,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 1.1428571939468384,
+      "rewards/check_perplexity_diff_func": 0.9642857313156128,
+      "rewards/check_winston_local_func": 0.5692740082740784,
+      "step": 779
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 1.0923076923076924,
+      "grad_norm": 4.3767238552948305e-07,
+      "kl": 0.189453125,
+      "learning_rate": 9.118382907149165e-05,
+      "loss": 0.0002,
+      "step": 780
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 89.85714721679688,
+      "epoch": 1.0937062937062938,
+      "grad_norm": 9.02834233189963e-08,
+      "kl": 0.154296875,
+      "learning_rate": 9.114919329468282e-05,
+      "loss": 0.0002,
+      "num_tokens": 1859828.0,
+      "reward": 1.9822884798049927,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.5714285969734192,
+      "rewards/check_perplexity_diff_func": 1.0,
+      "rewards/check_winston_local_func": 0.41085976362228394,
+      "step": 781
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 1.0951048951048952,
+      "grad_norm": 8.932188055681656e-08,
+      "kl": 0.154296875,
+      "learning_rate": 9.111449622020893e-05,
+      "loss": 0.0002,
+      "step": 782
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 136.46429443359375,
+      "epoch": 1.0965034965034965,
+      "grad_norm": 1.7587248831698471e-07,
+      "kl": 0.138671875,
+      "learning_rate": 9.107973789975626e-05,
+      "loss": 0.0001,
+      "num_tokens": 1883558.0,
+      "reward": 2.237004280090332,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.9285714030265808,
+      "rewards/check_perplexity_diff_func": 0.9285714030265808,
+      "rewards/check_winston_local_func": 0.37986141443252563,
+      "step": 783
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 1.097902097902098,
+      "grad_norm": 1.7072205199857748e-07,
+      "kl": 0.138671875,
+      "learning_rate": 9.104491838510235e-05,
+      "loss": 0.0001,
+      "step": 784
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 109.42857360839844,
+      "epoch": 1.0993006993006993,
+      "grad_norm": 4.723811406665928e-07,
+      "kl": 0.185546875,
+      "learning_rate": 9.101003772811589e-05,
+      "loss": 0.0002,
+      "num_tokens": 1904436.0,
+      "reward": 2.0636329650878906,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.5714285969734192,
+      "rewards/check_perplexity_diff_func": 1.0,
+      "rewards/check_winston_local_func": 0.492204487323761,
+      "step": 785
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 1.1006993006993007,
+      "grad_norm": 4.813313652081485e-07,
+      "kl": 0.185546875,
+      "learning_rate": 9.097509598075667e-05,
+      "loss": 0.0002,
+      "step": 786
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 100.75000762939453,
+      "epoch": 1.102097902097902,
+      "grad_norm": 1.2624745647996438e-07,
+      "kl": 0.15625,
+      "learning_rate": 9.094009319507547e-05,
+      "loss": 0.0002,
+      "num_tokens": 1924402.0,
+      "reward": 2.1538825035095215,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.7857142686843872,
+      "rewards/check_perplexity_diff_func": 0.9285714030265808,
+      "rewards/check_winston_local_func": 0.43959662318229675,
+      "step": 787
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 1.1034965034965034,
+      "grad_norm": 1.2855084178721202e-07,
+      "kl": 0.15625,
+      "learning_rate": 9.090502942321397e-05,
+      "loss": 0.0002,
+      "step": 788
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 101.26786041259766,
+      "epoch": 1.104895104895105,
+      "grad_norm": 0.002133418465001678,
+      "kl": 0.2109375,
+      "learning_rate": 9.086990471740473e-05,
+      "loss": -0.0001,
+      "num_tokens": 1944079.0,
+      "reward": 2.209216833114624,
+      "reward_std": 0.006096469704061747,
+      "rewards/check_gptzero_func": 0.7142857313156128,
+      "rewards/check_perplexity_diff_func": 0.9285714030265808,
+      "rewards/check_winston_local_func": 0.5663596987724304,
+      "step": 789
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 1.1062937062937064,
+      "grad_norm": 0.003624393686726291,
+      "kl": 0.2109375,
+      "learning_rate": 9.083471912997108e-05,
+      "loss": -0.0001,
+      "step": 790
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 119.39286041259766,
+      "epoch": 1.1076923076923078,
+      "grad_norm": 2.2118816611478268e-07,
+      "kl": 0.1328125,
+      "learning_rate": 9.079947271332706e-05,
+      "loss": 0.0001,
+      "num_tokens": 1965839.0,
+      "reward": 2.1228208541870117,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.7142857313156128,
+      "rewards/check_perplexity_diff_func": 1.0,
+      "rewards/check_winston_local_func": 0.40853506326675415,
+      "step": 791
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 1.1090909090909091,
+      "grad_norm": 2.2277128985396498e-07,
+      "kl": 0.1328125,
+      "learning_rate": 9.076416551997721e-05,
+      "loss": 0.0001,
+      "step": 792
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 88.03572082519531,
+      "epoch": 1.1104895104895105,
+      "grad_norm": 3.3242813310295524e-07,
+      "kl": 0.1640625,
+      "learning_rate": 9.072879760251679e-05,
+      "loss": 0.0002,
+      "num_tokens": 1983855.0,
+      "reward": 1.911961317062378,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.5714285969734192,
+      "rewards/check_perplexity_diff_func": 0.9642857313156128,
+      "rewards/check_winston_local_func": 0.3762470781803131,
+      "step": 793
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 1.1118881118881119,
+      "grad_norm": 3.866365448835279e-07,
+      "kl": 0.1640625,
+      "learning_rate": 9.069336901363137e-05,
+      "loss": 0.0002,
+      "step": 794
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 101.42857360839844,
+      "epoch": 1.1132867132867132,
+      "grad_norm": 4.186823728733852e-07,
+      "kl": 0.1435546875,
+      "learning_rate": 9.065787980609695e-05,
+      "loss": 0.0001,
+      "num_tokens": 2004327.0,
+      "reward": 2.2043607234954834,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.7857142686843872,
+      "rewards/check_perplexity_diff_func": 0.9285714030265808,
+      "rewards/check_winston_local_func": 0.49007490277290344,
+      "step": 795
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 1.1146853146853146,
+      "grad_norm": 5.030626047983616e-07,
+      "kl": 0.1435546875,
+      "learning_rate": 9.062233003277983e-05,
+      "loss": 0.0001,
+      "step": 796
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 103.03572082519531,
+      "epoch": 1.116083916083916,
+      "grad_norm": 0.0014210803266876898,
+      "kl": 0.166015625,
+      "learning_rate": 9.058671974663655e-05,
+      "loss": 0.0001,
+      "num_tokens": 2024235.0,
+      "reward": 2.1631124019622803,
+      "reward_std": 0.02190079353749752,
+      "rewards/check_gptzero_func": 0.6428571343421936,
+      "rewards/check_perplexity_diff_func": 0.9821428656578064,
+      "rewards/check_winston_local_func": 0.538112461566925,
+      "step": 797
+    },
+    {
+      "clip_ratio": 0.00044923630775883794,
+      "epoch": 1.1174825174825176,
+      "grad_norm": 1.0597804777807007,
+      "kl": 0.1669921875,
+      "learning_rate": 9.055104900071376e-05,
+      "loss": 0.0008,
+      "step": 798
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 112.67857360839844,
+      "epoch": 1.118881118881119,
+      "grad_norm": 2.8951990171744424e-07,
+      "kl": 0.1591796875,
+      "learning_rate": 9.051531784814817e-05,
+      "loss": 0.0002,
+      "num_tokens": 2045053.0,
+      "reward": 2.453437328338623,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 1.0,
+      "rewards/check_perplexity_diff_func": 1.0,
+      "rewards/check_winston_local_func": 0.45343732833862305,
+      "step": 799
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 1.1202797202797203,
+      "grad_norm": 6.334011517220965e-07,
+      "kl": 0.1591796875,
+      "learning_rate": 9.047952634216652e-05,
+      "loss": 0.0002,
+      "step": 800
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 2860,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 4,
+  "save_steps": 100,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 0.0,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/checkpoint-800/training_args.bin b/checkpoint-800/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..14e2d3ed154ee910c6c1698e855fc17a04eb6193
--- /dev/null
+++ b/checkpoint-800/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2c2b399ccd68443f851f396da81b1c9d32709e5be46a59f8f9eb00238e3b493d
+size 7480
diff --git a/checkpoint-800/zero_to_fp32.py b/checkpoint-800/zero_to_fp32.py
new file mode 100644
index 0000000000000000000000000000000000000000..0e759146cadd92ddfefab3680146c2bd6a2b5c04
--- /dev/null
+++ b/checkpoint-800/zero_to_fp32.py
@@ -0,0 +1,760 @@
+#!/usr/bin/env python
+
+# Copyright (c) Microsoft Corporation.
+# SPDX-License-Identifier: Apache-2.0
+
+# DeepSpeed Team
+
+# This script extracts fp32 consolidated weights from a zero 1, 2 and 3 DeepSpeed checkpoints. It gets
+# copied into the top level checkpoint dir, so the user can easily do the conversion at any point in
+# the future. Once extracted, the weights don't require DeepSpeed and can be used in any
+# application.
+#
+# example:
+#   python zero_to_fp32.py . output_dir/
+#   or
+#   python zero_to_fp32.py . output_dir/ --safe_serialization
+
+import argparse
+import torch
+import glob
+import math
+import os
+import re
+import gc
+import json
+import numpy as np
+from tqdm import tqdm
+from collections import OrderedDict
+from dataclasses import dataclass
+
+# while this script doesn't use deepspeed to recover data, since the checkpoints are pickled with
+# DeepSpeed data structures it has to be available in the current python environment.
+from deepspeed.utils import logger
+from deepspeed.checkpoint.constants import (DS_VERSION, OPTIMIZER_STATE_DICT, SINGLE_PARTITION_OF_FP32_GROUPS,
+                                            FP32_FLAT_GROUPS, ZERO_STAGE, PARTITION_COUNT, PARAM_SHAPES, BUFFER_NAMES,
+                                            FROZEN_PARAM_SHAPES, FROZEN_PARAM_FRAGMENTS)
+
+
+@dataclass
+class zero_model_state:
+    buffers: dict()
+    param_shapes: dict()
+    shared_params: list
+    ds_version: int
+    frozen_param_shapes: dict()
+    frozen_param_fragments: dict()
+
+
+debug = 0
+
+# load to cpu
+device = torch.device('cpu')
+
+
+def atoi(text):
+    return int(text) if text.isdigit() else text
+
+
+def natural_keys(text):
+    '''
+    alist.sort(key=natural_keys) sorts in human order
+    http://nedbatchelder.com/blog/200712/human_sorting.html
+    (See Toothy's implementation in the comments)
+    '''
+    return [atoi(c) for c in re.split(r'(\d+)', text)]
+
+
+def get_model_state_file(checkpoint_dir, zero_stage):
+    if not os.path.isdir(checkpoint_dir):
+        raise FileNotFoundError(f"Directory '{checkpoint_dir}' doesn't exist")
+
+    # there should be only one file
+    if zero_stage <= 2:
+        file = os.path.join(checkpoint_dir, "mp_rank_00_model_states.pt")
+    elif zero_stage == 3:
+        file = os.path.join(checkpoint_dir, "zero_pp_rank_0_mp_rank_00_model_states.pt")
+
+    if not os.path.exists(file):
+        raise FileNotFoundError(f"can't find model states file at '{file}'")
+
+    return file
+
+
+def get_checkpoint_files(checkpoint_dir, glob_pattern):
+    # XXX: need to test that this simple glob rule works for multi-node setup too
+    ckpt_files = sorted(glob.glob(os.path.join(checkpoint_dir, glob_pattern)), key=natural_keys)
+
+    if len(ckpt_files) == 0:
+        raise FileNotFoundError(f"can't find {glob_pattern} files in directory '{checkpoint_dir}'")
+
+    return ckpt_files
+
+
+def get_optim_files(checkpoint_dir):
+    return get_checkpoint_files(checkpoint_dir, "*_optim_states.pt")
+
+
+def get_model_state_files(checkpoint_dir):
+    return get_checkpoint_files(checkpoint_dir, "*_model_states.pt")
+
+
+def parse_model_states(files):
+    zero_model_states = []
+    for file in files:
+        state_dict = torch.load(file, map_location=device, weights_only=False)
+
+        if BUFFER_NAMES not in state_dict:
+            raise ValueError(f"{file} is not a model state checkpoint")
+        buffer_names = state_dict[BUFFER_NAMES]
+        if debug:
+            print("Found buffers:", buffer_names)
+
+        # recover just the buffers while restoring them to fp32 if they were saved in fp16
+        buffers = {k: v.float() for k, v in state_dict["module"].items() if k in buffer_names}
+        param_shapes = state_dict[PARAM_SHAPES]
+
+        # collect parameters that are included in param_shapes
+        param_names = []
+        for s in param_shapes:
+            for name in s.keys():
+                param_names.append(name)
+
+        # update with frozen parameters
+        frozen_param_shapes = state_dict.get(FROZEN_PARAM_SHAPES, None)
+        if frozen_param_shapes is not None:
+            if debug:
+                print(f"Found frozen_param_shapes: {frozen_param_shapes}")
+            param_names += list(frozen_param_shapes.keys())
+
+        # handle shared params
+        shared_params = [[k, v] for k, v in state_dict["shared_params"].items()]
+
+        ds_version = state_dict.get(DS_VERSION, None)
+
+        frozen_param_fragments = state_dict.get(FROZEN_PARAM_FRAGMENTS, None)
+
+        z_model_state = zero_model_state(buffers=buffers,
+                                         param_shapes=param_shapes,
+                                         shared_params=shared_params,
+                                         ds_version=ds_version,
+                                         frozen_param_shapes=frozen_param_shapes,
+                                         frozen_param_fragments=frozen_param_fragments)
+        zero_model_states.append(z_model_state)
+
+    return zero_model_states
+
+
+def parse_optim_states(files, ds_checkpoint_dir):
+    total_files = len(files)
+    state_dicts = []
+    for f in tqdm(files, desc='Loading checkpoint shards'):
+        state_dict = torch.load(f, map_location=device, mmap=True, weights_only=False)
+        # immediately discard the potentially huge 2 optimizer states as we only care for fp32 master weights
+        # and also handle the case where it was already removed by another helper script
+        state_dict["optimizer_state_dict"].pop("optimizer_state_dict", None)
+        state_dicts.append(state_dict)
+
+    if not ZERO_STAGE in state_dicts[0][OPTIMIZER_STATE_DICT]:
+        raise ValueError(f"{files[0]} is not a zero checkpoint")
+    zero_stage = state_dicts[0][OPTIMIZER_STATE_DICT][ZERO_STAGE]
+    world_size = state_dicts[0][OPTIMIZER_STATE_DICT][PARTITION_COUNT]
+
+    # For ZeRO-2 each param group can have different partition_count as data parallelism for expert
+    # parameters can be different from data parallelism for non-expert parameters. So we can just
+    # use the max of the partition_count to get the dp world_size.
+
+    if type(world_size) is list:
+        world_size = max(world_size)
+
+    if world_size != total_files:
+        raise ValueError(
+            f"Expected {world_size} of '*_optim_states.pt' under '{ds_checkpoint_dir}' but found {total_files} files. "
+            "Possibly due to an overwrite of an old checkpoint, or a checkpoint didn't get saved by one or more processes."
+        )
+
+    # the groups are named differently in each stage
+    if zero_stage <= 2:
+        fp32_groups_key = SINGLE_PARTITION_OF_FP32_GROUPS
+    elif zero_stage == 3:
+        fp32_groups_key = FP32_FLAT_GROUPS
+    else:
+        raise ValueError(f"unknown zero stage {zero_stage}")
+
+    fp32_flat_groups = [state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key] for i in range(len(state_dicts))]
+    return zero_stage, world_size, fp32_flat_groups
+
+
+def _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters):
+    """
+    Returns fp32 state_dict reconstructed from ds checkpoint
+
+    Args:
+        - ``ds_checkpoint_dir``: path to the deepspeed checkpoint folder (where the optimizer files are)
+
+    """
+    print(f"Processing zero checkpoint '{ds_checkpoint_dir}'")
+
+    optim_files = get_optim_files(ds_checkpoint_dir)
+    zero_stage, world_size, fp32_flat_groups = parse_optim_states(optim_files, ds_checkpoint_dir)
+    print(f"Detected checkpoint of type zero stage {zero_stage}, world_size: {world_size}")
+
+    model_files = get_model_state_files(ds_checkpoint_dir)
+
+    zero_model_states = parse_model_states(model_files)
+    print(f'Parsing checkpoint created by deepspeed=={zero_model_states[0].ds_version}')
+
+    if zero_stage <= 2:
+        return _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states,
+                                                          exclude_frozen_parameters)
+    elif zero_stage == 3:
+        return _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states,
+                                                          exclude_frozen_parameters)
+
+
+def _zero2_merge_frozen_params(state_dict, zero_model_states):
+    if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0:
+        return
+
+    frozen_param_shapes = zero_model_states[0].frozen_param_shapes
+    frozen_param_fragments = zero_model_states[0].frozen_param_fragments
+
+    if debug:
+        num_elem = sum(s.numel() for s in frozen_param_shapes.values())
+        print(f'rank 0: {FROZEN_PARAM_SHAPES}.numel = {num_elem}')
+
+        wanted_params = len(frozen_param_shapes)
+        wanted_numel = sum(s.numel() for s in frozen_param_shapes.values())
+        avail_numel = sum([p.numel() for p in frozen_param_fragments.values()])
+        print(f'Frozen params: Have {avail_numel} numels to process.')
+        print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params')
+
+    total_params = 0
+    total_numel = 0
+    for name, shape in frozen_param_shapes.items():
+        total_params += 1
+        unpartitioned_numel = shape.numel()
+        total_numel += unpartitioned_numel
+
+        state_dict[name] = frozen_param_fragments[name]
+
+        if debug:
+            print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ")
+
+    print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements")
+
+
+def _has_callable(obj, fn):
+    attr = getattr(obj, fn, None)
+    return callable(attr)
+
+
+def _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states):
+    param_shapes = zero_model_states[0].param_shapes
+
+    # Reconstruction protocol:
+    #
+    # XXX: document this
+
+    if debug:
+        for i in range(world_size):
+            for j in range(len(fp32_flat_groups[0])):
+                print(f"{FP32_FLAT_GROUPS}[{i}][{j}].shape={fp32_flat_groups[i][j].shape}")
+
+    # XXX: memory usage doubles here (zero2)
+    num_param_groups = len(fp32_flat_groups[0])
+    merged_single_partition_of_fp32_groups = []
+    for i in range(num_param_groups):
+        merged_partitions = [sd[i] for sd in fp32_flat_groups]
+        full_single_fp32_vector = torch.cat(merged_partitions, 0)
+        merged_single_partition_of_fp32_groups.append(full_single_fp32_vector)
+    avail_numel = sum(
+        [full_single_fp32_vector.numel() for full_single_fp32_vector in merged_single_partition_of_fp32_groups])
+
+    if debug:
+        wanted_params = sum([len(shapes) for shapes in param_shapes])
+        wanted_numel = sum([sum(shape.numel() for shape in shapes.values()) for shapes in param_shapes])
+        # not asserting if there is a mismatch due to possible padding
+        print(f"Have {avail_numel} numels to process.")
+        print(f"Need {wanted_numel} numels in {wanted_params} params.")
+
+    # params
+    # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support
+    # out-of-core computing solution
+    total_numel = 0
+    total_params = 0
+    for shapes, full_single_fp32_vector in zip(param_shapes, merged_single_partition_of_fp32_groups):
+        offset = 0
+        avail_numel = full_single_fp32_vector.numel()
+        for name, shape in shapes.items():
+
+            unpartitioned_numel = shape.numel() if _has_callable(shape, 'numel') else math.prod(shape)
+            total_numel += unpartitioned_numel
+            total_params += 1
+
+            if debug:
+                print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ")
+            state_dict[name] = full_single_fp32_vector.narrow(0, offset, unpartitioned_numel).view(shape)
+            offset += unpartitioned_numel
+
+        # Z2 started to align to 2*world_size to improve nccl performance. Therefore both offset and
+        # avail_numel can differ by anywhere between 0..2*world_size. Due to two unrelated complex
+        # paddings performed in the code it's almost impossible to predict the exact numbers w/o the
+        # live optimizer object, so we are checking that the numbers are within the right range
+        align_to = 2 * world_size
+
+        def zero2_align(x):
+            return align_to * math.ceil(x / align_to)
+
+        if debug:
+            print(f"original offset={offset}, avail_numel={avail_numel}")
+
+        offset = zero2_align(offset)
+        avail_numel = zero2_align(avail_numel)
+
+        if debug:
+            print(f"aligned  offset={offset}, avail_numel={avail_numel}")
+
+        # Sanity check
+        if offset != avail_numel:
+            raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong")
+
+    print(f"Reconstructed fp32 state dict with {total_params} params {total_numel} elements")
+
+
+def _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states,
+                                               exclude_frozen_parameters):
+    state_dict = OrderedDict()
+
+    # buffers
+    buffers = zero_model_states[0].buffers
+    state_dict.update(buffers)
+    if debug:
+        print(f"added {len(buffers)} buffers")
+
+    if not exclude_frozen_parameters:
+        _zero2_merge_frozen_params(state_dict, zero_model_states)
+
+    _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states)
+
+    # recover shared parameters
+    for pair in zero_model_states[0].shared_params:
+        if pair[1] in state_dict:
+            state_dict[pair[0]] = state_dict[pair[1]]
+
+    return state_dict
+
+
+def zero3_partitioned_param_info(unpartitioned_numel, world_size):
+    remainder = unpartitioned_numel % world_size
+    padding_numel = (world_size - remainder) if remainder else 0
+    partitioned_numel = math.ceil(unpartitioned_numel / world_size)
+    return partitioned_numel, padding_numel
+
+
+def _zero3_merge_frozen_params(state_dict, world_size, zero_model_states):
+    if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0:
+        return
+
+    if debug:
+        for i in range(world_size):
+            num_elem = sum(s.numel() for s in zero_model_states[i].frozen_param_fragments.values())
+            print(f'rank {i}: {FROZEN_PARAM_SHAPES}.numel = {num_elem}')
+
+        frozen_param_shapes = zero_model_states[0].frozen_param_shapes
+        wanted_params = len(frozen_param_shapes)
+        wanted_numel = sum(s.numel() for s in frozen_param_shapes.values())
+        avail_numel = sum([p.numel() for p in zero_model_states[0].frozen_param_fragments.values()]) * world_size
+        print(f'Frozen params: Have {avail_numel} numels to process.')
+        print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params')
+
+    total_params = 0
+    total_numel = 0
+    for name, shape in zero_model_states[0].frozen_param_shapes.items():
+        total_params += 1
+        unpartitioned_numel = shape.numel()
+        total_numel += unpartitioned_numel
+
+        param_frags = tuple(model_state.frozen_param_fragments[name] for model_state in zero_model_states)
+        state_dict[name] = torch.cat(param_frags, 0).narrow(0, 0, unpartitioned_numel).view(shape)
+
+        partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size)
+
+        if debug:
+            print(
+                f"Frozen params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}"
+            )
+
+    print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements")
+
+
+class GatheredTensor:
+    """
+    A pseudo tensor that collects partitioned weights.
+    It is more memory efficient when there are multiple groups.
+    """
+
+    def __init__(self, flat_groups, flat_groups_offset, offset, partitioned_numel, shape):
+        self.flat_groups = flat_groups
+        self.flat_groups_offset = flat_groups_offset
+        self.offset = offset
+        self.partitioned_numel = partitioned_numel
+        self.shape = shape
+        self.dtype = self.flat_groups[0][0].dtype
+
+    def contiguous(self):
+        """
+        Merge partitioned weights from flat_groups into a single tensor.
+        """
+        end_idx = self.offset + self.partitioned_numel
+        world_size = len(self.flat_groups)
+        pad_flat_param_chunks = []
+
+        for rank_i in range(world_size):
+            # for each rank, we need to collect weights from related group/groups
+            flat_groups_at_rank_i = self.flat_groups[rank_i]
+            start_group_id = None
+            end_group_id = None
+            for group_id in range(len(self.flat_groups_offset)):
+                if self.flat_groups_offset[group_id] <= self.offset < self.flat_groups_offset[group_id + 1]:
+                    start_group_id = group_id
+                if self.flat_groups_offset[group_id] < end_idx <= self.flat_groups_offset[group_id + 1]:
+                    end_group_id = group_id
+                    break
+            # collect weights from related group/groups
+            for group_id in range(start_group_id, end_group_id + 1):
+                flat_tensor = flat_groups_at_rank_i[group_id]
+                start_offset = self.offset - self.flat_groups_offset[group_id]
+                end_offset = min(end_idx, self.flat_groups_offset[group_id + 1]) - self.flat_groups_offset[group_id]
+                pad_flat_param_chunks.append(flat_tensor[start_offset:end_offset])
+
+        # collect weights from all ranks
+        pad_flat_param = torch.cat(pad_flat_param_chunks, dim=0)
+        param = pad_flat_param[:self.shape.numel()].view(self.shape).contiguous()
+        return param
+
+
+def _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states):
+    param_shapes = zero_model_states[0].param_shapes
+    avail_numel = sum([flat_group.numel() for flat_group in fp32_flat_groups[0]]) * world_size
+
+    # Reconstruction protocol: For zero3 we need to zip the partitions together at boundary of each
+    # param, re-consolidating each param, while dealing with padding if any
+
+    # merge list of dicts, preserving order
+    param_shapes = {k: v for d in param_shapes for k, v in d.items()}
+
+    if debug:
+        for i in range(world_size):
+            print(f"{FP32_FLAT_GROUPS}[{i}].shape={fp32_flat_groups[i].shape}")
+
+        wanted_params = len(param_shapes)
+        wanted_numel = sum(shape.numel() for shape in param_shapes.values())
+        # not asserting if there is a mismatch due to possible padding
+        avail_numel = fp32_flat_groups[0].numel() * world_size
+        print(f"Trainable params: Have {avail_numel} numels to process.")
+        print(f"Trainable params: Need {wanted_numel} numels in {wanted_params} params.")
+
+    # params
+    # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support
+    # out-of-core computing solution
+    offset = 0
+    total_numel = 0
+    total_params = 0
+    flat_groups_offset = [0] + list(np.cumsum([flat_tensor.numel() for flat_tensor in fp32_flat_groups[0]]))
+    for name, shape in tqdm(param_shapes.items(), desc='Gathering sharded weights'):
+        unpartitioned_numel = shape.numel()
+        total_numel += unpartitioned_numel
+        total_params += 1
+        partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size)
+
+        if debug:
+            print(
+                f"Trainable params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}"
+            )
+
+        # memory efficient tensor
+        tensor = GatheredTensor(fp32_flat_groups, flat_groups_offset, offset, partitioned_numel, shape)
+        state_dict[name] = tensor
+        offset += partitioned_numel
+
+    offset *= world_size
+
+    # Sanity check
+    if offset != avail_numel:
+        raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong")
+
+    print(f"Reconstructed Trainable fp32 state dict with {total_params} params {total_numel} elements")
+
+
+def _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states,
+                                               exclude_frozen_parameters):
+    state_dict = OrderedDict()
+
+    # buffers
+    buffers = zero_model_states[0].buffers
+    state_dict.update(buffers)
+    if debug:
+        print(f"added {len(buffers)} buffers")
+
+    if not exclude_frozen_parameters:
+        _zero3_merge_frozen_params(state_dict, world_size, zero_model_states)
+
+    _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states)
+
+    # recover shared parameters
+    for pair in zero_model_states[0].shared_params:
+        if pair[1] in state_dict:
+            state_dict[pair[0]] = state_dict[pair[1]]
+
+    return state_dict
+
+
+def to_torch_tensor(state_dict, return_empty_tensor=False):
+    """
+    Convert state_dict of GatheredTensor to torch tensor
+    """
+    torch_state_dict = {}
+    converted_tensors = {}
+    for name, tensor in state_dict.items():
+        tensor_id = id(tensor)
+        if tensor_id in converted_tensors:  # shared tensors
+            shared_tensor = torch_state_dict[converted_tensors[tensor_id]]
+            torch_state_dict[name] = shared_tensor
+        else:
+            converted_tensors[tensor_id] = name
+            if return_empty_tensor:
+                torch_state_dict[name] = torch.empty(tensor.shape, dtype=tensor.dtype)
+            else:
+                torch_state_dict[name] = tensor.contiguous()
+    return torch_state_dict
+
+
+def get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir,
+                                             tag=None,
+                                             exclude_frozen_parameters=False,
+                                             lazy_mode=False):
+    """
+    Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated state_dict that can be loaded with
+    ``load_state_dict()`` and used for training without DeepSpeed or shared with others, for example
+    via a model hub.
+
+    Args:
+        - ``checkpoint_dir``: path to the desired checkpoint folder
+        - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in 'latest' file. e.g., ``global_step14``
+        - ``exclude_frozen_parameters``: exclude frozen parameters
+        - ``lazy_mode``: get state_dict in lazy mode. It returns a dict of pesduo tensor instead of torch tensor, which is more memory efficient.
+          Convert the pesduo tensor to torch tensor by ``.contiguous()``
+
+    Returns:
+        - pytorch ``state_dict``
+
+    A typical usage might be ::
+
+        from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint
+        # do the training and checkpoint saving
+        state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir) # already on cpu
+        model = model.cpu() # move to cpu
+        model.load_state_dict(state_dict)
+        # submit to model hub or save the model to share with others
+
+    In this example the ``model`` will no longer be usable in the deepspeed context of the same
+    application. i.e. you will need to re-initialize the deepspeed engine, since
+    ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it.
+
+    If you want it all done for you, use ``load_state_dict_from_zero_checkpoint`` instead.
+
+    Note: the above usage may not work if your application doesn't have sufficient free CPU memory.
+    You may need to use the offline approach using the ``zero_to_fp32.py`` script that is saved with
+    the checkpoint. Or you can load state_dict in lazy mode ::
+
+        from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint
+        state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, lazy_mode=True) # not on cpu
+        for name, lazy_tensor in state_dict.item():
+            tensor = lazy_tensor.contiguous()  # to cpu
+            print(name, tensor)
+            # del tensor to release memory if it no longer in use
+    """
+    if tag is None:
+        latest_path = os.path.join(checkpoint_dir, 'latest')
+        if os.path.isfile(latest_path):
+            with open(latest_path, 'r') as fd:
+                tag = fd.read().strip()
+        else:
+            raise ValueError(f"Unable to find 'latest' file at {latest_path}")
+
+    ds_checkpoint_dir = os.path.join(checkpoint_dir, tag)
+
+    if not os.path.isdir(ds_checkpoint_dir):
+        raise FileNotFoundError(f"Directory '{ds_checkpoint_dir}' doesn't exist")
+
+    state_dict = _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters)
+    if lazy_mode:
+        return state_dict
+    else:
+        return to_torch_tensor(state_dict)
+
+
+def convert_zero_checkpoint_to_fp32_state_dict(checkpoint_dir,
+                                               output_dir,
+                                               max_shard_size="5GB",
+                                               safe_serialization=False,
+                                               tag=None,
+                                               exclude_frozen_parameters=False):
+    """
+    Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` file that can be
+    loaded with ``torch.load(file)`` + ``load_state_dict()`` and used for training without DeepSpeed.
+
+    Args:
+        - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``)
+        - ``output_dir``: directory to the pytorch fp32 state_dict output files
+        - ``max_shard_size``: the maximum size for a checkpoint before being sharded, default value is 5GB
+        - ``safe_serialization``:  whether to save the model using `safetensors` or the traditional PyTorch way (that uses `pickle`).
+        - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14``
+        - ``exclude_frozen_parameters``: exclude frozen parameters
+    """
+
+    # Dependency pre-check
+    if safe_serialization:
+        try:
+            from safetensors.torch import save_file
+        except ImportError:
+            print('If you want to use `safe_serialization`, please `pip install safetensors`')
+            raise
+    if max_shard_size is not None:
+        try:
+            from huggingface_hub import split_torch_state_dict_into_shards
+        except ImportError:
+            print('If you want to use `max_shard_size`, please `pip install huggingface_hub`')
+            raise
+
+    # Convert zero checkpoint to state_dict
+    state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir,
+                                                          tag,
+                                                          exclude_frozen_parameters,
+                                                          lazy_mode=True)
+
+    # Shard the model if it is too big.
+    weights_name = "model.safetensors" if safe_serialization else "pytorch_model.bin"
+    if max_shard_size is not None:
+        filename_pattern = weights_name.replace(".bin", "{suffix}.bin").replace(".safetensors", "{suffix}.safetensors")
+        # an memory-efficient approach for sharding
+        empty_state_dict = to_torch_tensor(state_dict, return_empty_tensor=True)
+        state_dict_split = split_torch_state_dict_into_shards(empty_state_dict,
+                                                              filename_pattern=filename_pattern,
+                                                              max_shard_size=max_shard_size)
+    else:
+        from collections import namedtuple
+        StateDictSplit = namedtuple("StateDictSplit", ["is_sharded", "filename_to_tensors"])
+        state_dict_split = StateDictSplit(is_sharded=False,
+                                          filename_to_tensors={weights_name: list(state_dict.keys())})
+
+    # Save the model by shard
+    os.makedirs(output_dir, exist_ok=True)
+    filename_to_tensors = state_dict_split.filename_to_tensors.items()
+    for shard_file, tensors in tqdm(filename_to_tensors, desc="Saving checkpoint shards"):
+        shard_state_dict = {tensor_name: state_dict[tensor_name] for tensor_name in tensors}
+        shard_state_dict = to_torch_tensor(shard_state_dict)
+        output_path = os.path.join(output_dir, shard_file)
+        if safe_serialization:
+            save_file(shard_state_dict, output_path, metadata={"format": "pt"})
+        else:
+            torch.save(shard_state_dict, output_path)
+        # release the memory of current shard
+        for tensor_name in list(shard_state_dict.keys()):
+            del state_dict[tensor_name]
+            del shard_state_dict[tensor_name]
+        del shard_state_dict
+        gc.collect()
+
+    # Save index if sharded
+    if state_dict_split.is_sharded:
+        index = {
+            "metadata": state_dict_split.metadata,
+            "weight_map": state_dict_split.tensor_to_filename,
+        }
+        save_index_file = "model.safetensors.index.json" if safe_serialization else "pytorch_model.bin.index.json"
+        save_index_file = os.path.join(output_dir, save_index_file)
+        with open(save_index_file, "w", encoding="utf-8") as f:
+            content = json.dumps(index, indent=2, sort_keys=True) + "\n"
+            f.write(content)
+
+
+def load_state_dict_from_zero_checkpoint(model, checkpoint_dir, tag=None):
+    """
+    1. Put the provided model to cpu
+    2. Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict``
+    3. Load it into the provided model
+
+    Args:
+        - ``model``: the model object to update
+        - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``)
+        - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14``
+
+    Returns:
+        - ``model`: modified model
+
+    Make sure you have plenty of CPU memory available before you call this function. If you don't
+    have enough use the ``zero_to_fp32.py`` utility to do the conversion. You will find it
+    conveniently placed for you in the checkpoint folder.
+
+    A typical usage might be ::
+
+        from deepspeed.utils.zero_to_fp32 import load_state_dict_from_zero_checkpoint
+        model = load_state_dict_from_zero_checkpoint(trainer.model, checkpoint_dir)
+        # submit to model hub or save the model to share with others
+
+    Note, that once this was run, the ``model`` will no longer be usable in the deepspeed context
+    of the same application. i.e. you will need to re-initialize the deepspeed engine, since
+    ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it.
+
+    """
+    logger.info(f"Extracting fp32 weights")
+    state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag)
+
+    logger.info(f"Overwriting model with fp32 weights")
+    model = model.cpu()
+    model.load_state_dict(state_dict, strict=False)
+
+    return model
+
+
+if __name__ == "__main__":
+    parser = argparse.ArgumentParser()
+    parser.add_argument("checkpoint_dir",
+                        type=str,
+                        help="path to the desired checkpoint folder, e.g., path/checkpoint-12")
+    parser.add_argument("output_dir",
+                        type=str,
+                        help="directory to the pytorch fp32 state_dict output files"
+                        "(e.g. path/checkpoint-12-output/)")
+    parser.add_argument(
+        "--max_shard_size",
+        type=str,
+        default="5GB",
+        help="The maximum size for a checkpoint before being sharded. Checkpoints shard will then be each of size"
+        "lower than this size. If expressed as a string, needs to be digits followed by a unit (like `5MB`"
+        "We default it to 5GB in order for models to be able to run easily on free-tier google colab instances"
+        "without CPU OOM issues.")
+    parser.add_argument(
+        "--safe_serialization",
+        default=False,
+        action='store_true',
+        help="Whether to save the model using `safetensors` or the traditional PyTorch way (that uses `pickle`).")
+    parser.add_argument("-t",
+                        "--tag",
+                        type=str,
+                        default=None,
+                        help="checkpoint tag used as a unique identifier for checkpoint. e.g., global_step1")
+    parser.add_argument("--exclude_frozen_parameters", action='store_true', help="exclude frozen parameters")
+    parser.add_argument("-d", "--debug", action='store_true', help="enable debug")
+    args = parser.parse_args()
+
+    debug = args.debug
+
+    convert_zero_checkpoint_to_fp32_state_dict(args.checkpoint_dir,
+                                               args.output_dir,
+                                               max_shard_size=args.max_shard_size,
+                                               safe_serialization=args.safe_serialization,
+                                               tag=args.tag,
+                                               exclude_frozen_parameters=args.exclude_frozen_parameters)
diff --git a/checkpoint-900/README.md b/checkpoint-900/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..85f3785e3148e4fdfa69be9cd1528c3af1891702
--- /dev/null
+++ b/checkpoint-900/README.md
@@ -0,0 +1,202 @@
+---
+base_model: mistralai/Mistral-Nemo-Instruct-2407
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.14.0
\ No newline at end of file
diff --git a/checkpoint-900/adapter_config.json b/checkpoint-900/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..1a10cbd48cf122bc7508a7cd9c23007d932168d0
--- /dev/null
+++ b/checkpoint-900/adapter_config.json
@@ -0,0 +1,37 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "mistralai/Mistral-Nemo-Instruct-2407",
+  "bias": "none",
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 16,
+  "lora_bias": false,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 16,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "k_proj",
+    "down_proj",
+    "q_proj",
+    "v_proj",
+    "o_proj",
+    "up_proj",
+    "gate_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/checkpoint-900/adapter_model.safetensors b/checkpoint-900/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..79d7dff060ab0d078dadb3d8ff6367287e92c6ff
--- /dev/null
+++ b/checkpoint-900/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9065ee482a52c2d493b49d4bf786722b509d3bd68d23f6ad565219878730c823
+size 114107976
diff --git a/checkpoint-900/global_step900/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt b/checkpoint-900/global_step900/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..9eee13bafe7cc01cbb20562efdbd0fd0a6c97ec2
--- /dev/null
+++ b/checkpoint-900/global_step900/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:eff66ed4c5554719f0ef36a83300685221c51b1059b3006fa72ceb822dc450df
+size 49133072
diff --git a/checkpoint-900/global_step900/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt b/checkpoint-900/global_step900/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..8e96af50fb8e7099413838074fa882386b8bafac
--- /dev/null
+++ b/checkpoint-900/global_step900/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:49c167b9637b47d7e3e2eeb7c0f1997affdbc184a7233dcfe2799d78f080404d
+size 49133072
diff --git a/checkpoint-900/global_step900/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt b/checkpoint-900/global_step900/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..d01aa863fd1512ba4d776251fc9b6902350d8170
--- /dev/null
+++ b/checkpoint-900/global_step900/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:266c7dbe7cad7e500c8ae998f9cf7cf4721d7d3c93fe953c1631b9e2bb6d1c2d
+size 49133072
diff --git a/checkpoint-900/global_step900/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt b/checkpoint-900/global_step900/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..1ac0a6e980b85209d76c2940a1fcd4f275e93e5d
--- /dev/null
+++ b/checkpoint-900/global_step900/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:249fde0ffe6f2786b3743aaee4f79881c3c94c8e9a31b5d616221ae69a2b245c
+size 49133072
diff --git a/checkpoint-900/global_step900/bf16_zero_pp_rank_4_mp_rank_00_optim_states.pt b/checkpoint-900/global_step900/bf16_zero_pp_rank_4_mp_rank_00_optim_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..ffc87faae9085800712a373a6b09e33da91c9d55
--- /dev/null
+++ b/checkpoint-900/global_step900/bf16_zero_pp_rank_4_mp_rank_00_optim_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a5fca8810d21c074730a4a6d5f090680582a0bdbda0d8b254f1b83669c777927
+size 49133072
diff --git a/checkpoint-900/global_step900/bf16_zero_pp_rank_5_mp_rank_00_optim_states.pt b/checkpoint-900/global_step900/bf16_zero_pp_rank_5_mp_rank_00_optim_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..2ba885e3c804c0b1d4bd2cf9c2f0af4f53d07fdc
--- /dev/null
+++ b/checkpoint-900/global_step900/bf16_zero_pp_rank_5_mp_rank_00_optim_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9c95ec39720fc383f5288d0cd15628f282a6f9631af06af5f177a17620dc416b
+size 49133072
diff --git a/checkpoint-900/global_step900/bf16_zero_pp_rank_6_mp_rank_00_optim_states.pt b/checkpoint-900/global_step900/bf16_zero_pp_rank_6_mp_rank_00_optim_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..54907ddc1b4e653ff4174b924d1ce5c75b629fee
--- /dev/null
+++ b/checkpoint-900/global_step900/bf16_zero_pp_rank_6_mp_rank_00_optim_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:25584556b843280daf521ac410f8bccda835c40fef778c26124f7d1d3ac4ad8a
+size 49133072
diff --git a/checkpoint-900/global_step900/zero_pp_rank_0_mp_rank_00_model_states.pt b/checkpoint-900/global_step900/zero_pp_rank_0_mp_rank_00_model_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..a85bfa08e087423b03cf302701bdaa2cc76d77fc
--- /dev/null
+++ b/checkpoint-900/global_step900/zero_pp_rank_0_mp_rank_00_model_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f48c2f0a0da20f710dea2b2c062e060a95998f6d52cbd09a46f7a1951b0fe20d
+size 553886
diff --git a/checkpoint-900/global_step900/zero_pp_rank_1_mp_rank_00_model_states.pt b/checkpoint-900/global_step900/zero_pp_rank_1_mp_rank_00_model_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..8526e1fd4a0fc1c00f815cf4de0df17391b155ed
--- /dev/null
+++ b/checkpoint-900/global_step900/zero_pp_rank_1_mp_rank_00_model_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7bc5ecd8719cefb12f7a1186b8c1ead47d251a4269f9f513e00aa917569e393c
+size 553886
diff --git a/checkpoint-900/global_step900/zero_pp_rank_2_mp_rank_00_model_states.pt b/checkpoint-900/global_step900/zero_pp_rank_2_mp_rank_00_model_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..58c62b412386aa5c01a51ecf45d83f074ea89f00
--- /dev/null
+++ b/checkpoint-900/global_step900/zero_pp_rank_2_mp_rank_00_model_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:458571af2f69d44c8f2c59e041ac6c4e01be74312e96510e932b7a463fb3c7a4
+size 553886
diff --git a/checkpoint-900/global_step900/zero_pp_rank_3_mp_rank_00_model_states.pt b/checkpoint-900/global_step900/zero_pp_rank_3_mp_rank_00_model_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..c19eadf542a1fbd79d5e501681e3bd4c20609e1b
--- /dev/null
+++ b/checkpoint-900/global_step900/zero_pp_rank_3_mp_rank_00_model_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e1e7c0240f40b47702a31b99f30e0c4c9567d485d67ad2ce2d83881d27cd0af1
+size 553886
diff --git a/checkpoint-900/global_step900/zero_pp_rank_4_mp_rank_00_model_states.pt b/checkpoint-900/global_step900/zero_pp_rank_4_mp_rank_00_model_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..034e55fe58cced0cd489ad27f23fc47510931a0f
--- /dev/null
+++ b/checkpoint-900/global_step900/zero_pp_rank_4_mp_rank_00_model_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7efa1d45d7d28967b8b661240472c0aad87e502377f1484a71b2f9e90cf283ba
+size 553886
diff --git a/checkpoint-900/global_step900/zero_pp_rank_5_mp_rank_00_model_states.pt b/checkpoint-900/global_step900/zero_pp_rank_5_mp_rank_00_model_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..428196787f7aec74e4b10c322abb5dd0071bc77a
--- /dev/null
+++ b/checkpoint-900/global_step900/zero_pp_rank_5_mp_rank_00_model_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2be2655afe8c6022c6070215abc8da19e0d25adad637e5a3e8c09d55a0d68736
+size 553886
diff --git a/checkpoint-900/global_step900/zero_pp_rank_6_mp_rank_00_model_states.pt b/checkpoint-900/global_step900/zero_pp_rank_6_mp_rank_00_model_states.pt
new file mode 100644
index 0000000000000000000000000000000000000000..5e6655f6cbf404e5bfa17e05b62d9b241baf4e02
--- /dev/null
+++ b/checkpoint-900/global_step900/zero_pp_rank_6_mp_rank_00_model_states.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:35ad53b55468a923b96d4016faf298dd0badfe45e2a413cbfa26e649e693212f
+size 553886
diff --git a/checkpoint-900/latest b/checkpoint-900/latest
new file mode 100644
index 0000000000000000000000000000000000000000..4b10acccf3e8395339ff8799cea202bbc54d7f7d
--- /dev/null
+++ b/checkpoint-900/latest
@@ -0,0 +1 @@
+global_step900
\ No newline at end of file
diff --git a/checkpoint-900/rng_state_0.pth b/checkpoint-900/rng_state_0.pth
new file mode 100644
index 0000000000000000000000000000000000000000..a34a1616a9e3954343c479fbffef981fa48162e7
--- /dev/null
+++ b/checkpoint-900/rng_state_0.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5713f36ffa627adfeac5a00f1e68b91f6f21347a7f987b965eb5e5815d1c263f
+size 15728
diff --git a/checkpoint-900/rng_state_1.pth b/checkpoint-900/rng_state_1.pth
new file mode 100644
index 0000000000000000000000000000000000000000..a9ece55d08f36cee0149d2db59c6967e65c1fd0e
--- /dev/null
+++ b/checkpoint-900/rng_state_1.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bbae3ac4c10e8b357c2ee5ec241ed21adeddc0abd991b309a465da7094b8f0d8
+size 15728
diff --git a/checkpoint-900/rng_state_2.pth b/checkpoint-900/rng_state_2.pth
new file mode 100644
index 0000000000000000000000000000000000000000..723a7966c5e15ded1ca3e961539b8c8acfa3d36f
--- /dev/null
+++ b/checkpoint-900/rng_state_2.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:49f993e038d439980f54d95da934542f4252b40a6bc18742af779c829e74e6aa
+size 15728
diff --git a/checkpoint-900/rng_state_3.pth b/checkpoint-900/rng_state_3.pth
new file mode 100644
index 0000000000000000000000000000000000000000..496122dcc5aaa1ec176d899f2799ad1569c04f6c
--- /dev/null
+++ b/checkpoint-900/rng_state_3.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4cd2e733ddc6f3706f7a2750006c0ec8aed83cc0bc987f2894ca877a17f3ded3
+size 15792
diff --git a/checkpoint-900/rng_state_4.pth b/checkpoint-900/rng_state_4.pth
new file mode 100644
index 0000000000000000000000000000000000000000..cb66a5ec0a1d31b3c4425a7410428c3167ebcd09
--- /dev/null
+++ b/checkpoint-900/rng_state_4.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c0190bcf7f11b0e86ce4c6aa8169c549662397e03e6a6a3db26911893740f78f
+size 15728
diff --git a/checkpoint-900/rng_state_5.pth b/checkpoint-900/rng_state_5.pth
new file mode 100644
index 0000000000000000000000000000000000000000..8788c4db32d0a151f7708e6e6894e03be73035e2
--- /dev/null
+++ b/checkpoint-900/rng_state_5.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:52686cb6e57a04df54182dff39383473884dad1d1c255059da7a0ad5cdc61e9a
+size 15728
diff --git a/checkpoint-900/rng_state_6.pth b/checkpoint-900/rng_state_6.pth
new file mode 100644
index 0000000000000000000000000000000000000000..84d4cc7102fe39daddd8aba5794d419ffecb9c24
--- /dev/null
+++ b/checkpoint-900/rng_state_6.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:68f69abc5ca5bf5b9583378bb1a7268ad7bd8d59ca5d12c0e2d262a7423ca392
+size 15728
diff --git a/checkpoint-900/scheduler.pt b/checkpoint-900/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..91ba7aa36a9c8a1c40ab97a82aaa30980090b06a
--- /dev/null
+++ b/checkpoint-900/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:14af61686889819f5397d74cc07a362b7e8295f1a40db76049b5499a31dae7f0
+size 1064
diff --git a/checkpoint-900/special_tokens_map.json b/checkpoint-900/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..72ecfeeb7e14d244c936169d2ed139eeae235ef1
--- /dev/null
+++ b/checkpoint-900/special_tokens_map.json
@@ -0,0 +1,24 @@
+{
+  "bos_token": {
+    "content": "<s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "</s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "</s>",
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}
diff --git a/checkpoint-900/tokenizer.json b/checkpoint-900/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..b35b1c0f0bd5007afc7ad5976eeaf44988bd2067
--- /dev/null
+++ b/checkpoint-900/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9a1c103d223ee5dc2dde8307635f7d12581b40855198a38efcfeb6db4e08da69
+size 17078445
diff --git a/checkpoint-900/tokenizer_config.json b/checkpoint-900/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..550f3f7fc9144301cd0f17eeb0ad8746a5fa5354
--- /dev/null
+++ b/checkpoint-900/tokenizer_config.json
@@ -0,0 +1,8020 @@
+{
+  "add_bos_token": true,
+  "add_eos_token": false,
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "[INST]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "4": {
+      "content": "[/INST]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "5": {
+      "content": "[AVAILABLE_TOOLS]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "6": {
+      "content": "[/AVAILABLE_TOOLS]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "7": {
+      "content": "[TOOL_RESULTS]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "8": {
+      "content": "[/TOOL_RESULTS]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "9": {
+      "content": "[TOOL_CALLS]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "10": {
+      "content": "<pad>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "11": {
+      "content": "[PREFIX]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "12": {
+      "content": "[MIDDLE]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "13": {
+      "content": "[SUFFIX]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "14": {
+      "content": "<SPECIAL_14>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "15": {
+      "content": "<SPECIAL_15>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "16": {
+      "content": "<SPECIAL_16>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "17": {
+      "content": "<SPECIAL_17>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "18": {
+      "content": "<SPECIAL_18>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "19": {
+      "content": "<SPECIAL_19>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "20": {
+      "content": "<SPECIAL_20>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "21": {
+      "content": "<SPECIAL_21>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "22": {
+      "content": "<SPECIAL_22>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "23": {
+      "content": "<SPECIAL_23>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "24": {
+      "content": "<SPECIAL_24>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "25": {
+      "content": "<SPECIAL_25>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "26": {
+      "content": "<SPECIAL_26>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "27": {
+      "content": "<SPECIAL_27>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "28": {
+      "content": "<SPECIAL_28>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "29": {
+      "content": "<SPECIAL_29>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "30": {
+      "content": "<SPECIAL_30>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "31": {
+      "content": "<SPECIAL_31>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32": {
+      "content": "<SPECIAL_32>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "33": {
+      "content": "<SPECIAL_33>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "34": {
+      "content": "<SPECIAL_34>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "35": {
+      "content": "<SPECIAL_35>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "36": {
+      "content": "<SPECIAL_36>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "37": {
+      "content": "<SPECIAL_37>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "38": {
+      "content": "<SPECIAL_38>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "39": {
+      "content": "<SPECIAL_39>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "40": {
+      "content": "<SPECIAL_40>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "41": {
+      "content": "<SPECIAL_41>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "42": {
+      "content": "<SPECIAL_42>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "43": {
+      "content": "<SPECIAL_43>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "44": {
+      "content": "<SPECIAL_44>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "45": {
+      "content": "<SPECIAL_45>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "46": {
+      "content": "<SPECIAL_46>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "47": {
+      "content": "<SPECIAL_47>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "48": {
+      "content": "<SPECIAL_48>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "49": {
+      "content": "<SPECIAL_49>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "50": {
+      "content": "<SPECIAL_50>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "51": {
+      "content": "<SPECIAL_51>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "52": {
+      "content": "<SPECIAL_52>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "53": {
+      "content": "<SPECIAL_53>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "54": {
+      "content": "<SPECIAL_54>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "55": {
+      "content": "<SPECIAL_55>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "56": {
+      "content": "<SPECIAL_56>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "57": {
+      "content": "<SPECIAL_57>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "58": {
+      "content": "<SPECIAL_58>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "59": {
+      "content": "<SPECIAL_59>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "60": {
+      "content": "<SPECIAL_60>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "61": {
+      "content": "<SPECIAL_61>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "62": {
+      "content": "<SPECIAL_62>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "63": {
+      "content": "<SPECIAL_63>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "64": {
+      "content": "<SPECIAL_64>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "65": {
+      "content": "<SPECIAL_65>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "66": {
+      "content": "<SPECIAL_66>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "67": {
+      "content": "<SPECIAL_67>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "68": {
+      "content": "<SPECIAL_68>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "69": {
+      "content": "<SPECIAL_69>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "70": {
+      "content": "<SPECIAL_70>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "71": {
+      "content": "<SPECIAL_71>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "72": {
+      "content": "<SPECIAL_72>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "73": {
+      "content": "<SPECIAL_73>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "74": {
+      "content": "<SPECIAL_74>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "75": {
+      "content": "<SPECIAL_75>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "76": {
+      "content": "<SPECIAL_76>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "77": {
+      "content": "<SPECIAL_77>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "78": {
+      "content": "<SPECIAL_78>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "79": {
+      "content": "<SPECIAL_79>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "80": {
+      "content": "<SPECIAL_80>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "81": {
+      "content": "<SPECIAL_81>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "82": {
+      "content": "<SPECIAL_82>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "83": {
+      "content": "<SPECIAL_83>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "84": {
+      "content": "<SPECIAL_84>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "85": {
+      "content": "<SPECIAL_85>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "86": {
+      "content": "<SPECIAL_86>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "87": {
+      "content": "<SPECIAL_87>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "88": {
+      "content": "<SPECIAL_88>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "89": {
+      "content": "<SPECIAL_89>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "90": {
+      "content": "<SPECIAL_90>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "91": {
+      "content": "<SPECIAL_91>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "92": {
+      "content": "<SPECIAL_92>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "93": {
+      "content": "<SPECIAL_93>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "94": {
+      "content": "<SPECIAL_94>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "95": {
+      "content": "<SPECIAL_95>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "96": {
+      "content": "<SPECIAL_96>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "97": {
+      "content": "<SPECIAL_97>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "98": {
+      "content": "<SPECIAL_98>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "99": {
+      "content": "<SPECIAL_99>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "100": {
+      "content": "<SPECIAL_100>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "101": {
+      "content": "<SPECIAL_101>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "102": {
+      "content": "<SPECIAL_102>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "103": {
+      "content": "<SPECIAL_103>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "104": {
+      "content": "<SPECIAL_104>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "105": {
+      "content": "<SPECIAL_105>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "106": {
+      "content": "<SPECIAL_106>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "107": {
+      "content": "<SPECIAL_107>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "108": {
+      "content": "<SPECIAL_108>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "109": {
+      "content": "<SPECIAL_109>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "110": {
+      "content": "<SPECIAL_110>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "111": {
+      "content": "<SPECIAL_111>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "112": {
+      "content": "<SPECIAL_112>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "113": {
+      "content": "<SPECIAL_113>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "114": {
+      "content": "<SPECIAL_114>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "115": {
+      "content": "<SPECIAL_115>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "116": {
+      "content": "<SPECIAL_116>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "117": {
+      "content": "<SPECIAL_117>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "118": {
+      "content": "<SPECIAL_118>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "119": {
+      "content": "<SPECIAL_119>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "120": {
+      "content": "<SPECIAL_120>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "121": {
+      "content": "<SPECIAL_121>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "122": {
+      "content": "<SPECIAL_122>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "123": {
+      "content": "<SPECIAL_123>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "124": {
+      "content": "<SPECIAL_124>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "125": {
+      "content": "<SPECIAL_125>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "126": {
+      "content": "<SPECIAL_126>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "127": {
+      "content": "<SPECIAL_127>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128": {
+      "content": "<SPECIAL_128>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "129": {
+      "content": "<SPECIAL_129>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "130": {
+      "content": "<SPECIAL_130>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "131": {
+      "content": "<SPECIAL_131>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "132": {
+      "content": "<SPECIAL_132>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "133": {
+      "content": "<SPECIAL_133>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "134": {
+      "content": "<SPECIAL_134>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "135": {
+      "content": "<SPECIAL_135>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "136": {
+      "content": "<SPECIAL_136>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "137": {
+      "content": "<SPECIAL_137>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "138": {
+      "content": "<SPECIAL_138>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "139": {
+      "content": "<SPECIAL_139>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "140": {
+      "content": "<SPECIAL_140>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "141": {
+      "content": "<SPECIAL_141>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "142": {
+      "content": "<SPECIAL_142>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "143": {
+      "content": "<SPECIAL_143>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "144": {
+      "content": "<SPECIAL_144>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "145": {
+      "content": "<SPECIAL_145>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "146": {
+      "content": "<SPECIAL_146>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "147": {
+      "content": "<SPECIAL_147>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "148": {
+      "content": "<SPECIAL_148>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "149": {
+      "content": "<SPECIAL_149>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "150": {
+      "content": "<SPECIAL_150>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151": {
+      "content": "<SPECIAL_151>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "152": {
+      "content": "<SPECIAL_152>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "153": {
+      "content": "<SPECIAL_153>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "154": {
+      "content": "<SPECIAL_154>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "155": {
+      "content": "<SPECIAL_155>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "156": {
+      "content": "<SPECIAL_156>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "157": {
+      "content": "<SPECIAL_157>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "158": {
+      "content": "<SPECIAL_158>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "159": {
+      "content": "<SPECIAL_159>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "160": {
+      "content": "<SPECIAL_160>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "161": {
+      "content": "<SPECIAL_161>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "162": {
+      "content": "<SPECIAL_162>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "163": {
+      "content": "<SPECIAL_163>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "164": {
+      "content": "<SPECIAL_164>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "165": {
+      "content": "<SPECIAL_165>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "166": {
+      "content": "<SPECIAL_166>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "167": {
+      "content": "<SPECIAL_167>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "168": {
+      "content": "<SPECIAL_168>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "169": {
+      "content": "<SPECIAL_169>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "170": {
+      "content": "<SPECIAL_170>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "171": {
+      "content": "<SPECIAL_171>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "172": {
+      "content": "<SPECIAL_172>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "173": {
+      "content": "<SPECIAL_173>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "174": {
+      "content": "<SPECIAL_174>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "175": {
+      "content": "<SPECIAL_175>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "176": {
+      "content": "<SPECIAL_176>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "177": {
+      "content": "<SPECIAL_177>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "178": {
+      "content": "<SPECIAL_178>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "179": {
+      "content": "<SPECIAL_179>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "180": {
+      "content": "<SPECIAL_180>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "181": {
+      "content": "<SPECIAL_181>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "182": {
+      "content": "<SPECIAL_182>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "183": {
+      "content": "<SPECIAL_183>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "184": {
+      "content": "<SPECIAL_184>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "185": {
+      "content": "<SPECIAL_185>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "186": {
+      "content": "<SPECIAL_186>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "187": {
+      "content": "<SPECIAL_187>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "188": {
+      "content": "<SPECIAL_188>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "189": {
+      "content": "<SPECIAL_189>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "190": {
+      "content": "<SPECIAL_190>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "191": {
+      "content": "<SPECIAL_191>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "192": {
+      "content": "<SPECIAL_192>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "193": {
+      "content": "<SPECIAL_193>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "194": {
+      "content": "<SPECIAL_194>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "195": {
+      "content": "<SPECIAL_195>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "196": {
+      "content": "<SPECIAL_196>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "197": {
+      "content": "<SPECIAL_197>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "198": {
+      "content": "<SPECIAL_198>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "199": {
+      "content": "<SPECIAL_199>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "200": {
+      "content": "<SPECIAL_200>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "201": {
+      "content": "<SPECIAL_201>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "202": {
+      "content": "<SPECIAL_202>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "203": {
+      "content": "<SPECIAL_203>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "204": {
+      "content": "<SPECIAL_204>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "205": {
+      "content": "<SPECIAL_205>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "206": {
+      "content": "<SPECIAL_206>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "207": {
+      "content": "<SPECIAL_207>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "208": {
+      "content": "<SPECIAL_208>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "209": {
+      "content": "<SPECIAL_209>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "210": {
+      "content": "<SPECIAL_210>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "211": {
+      "content": "<SPECIAL_211>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "212": {
+      "content": "<SPECIAL_212>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "213": {
+      "content": "<SPECIAL_213>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "214": {
+      "content": "<SPECIAL_214>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "215": {
+      "content": "<SPECIAL_215>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "216": {
+      "content": "<SPECIAL_216>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "217": {
+      "content": "<SPECIAL_217>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "218": {
+      "content": "<SPECIAL_218>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "219": {
+      "content": "<SPECIAL_219>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "220": {
+      "content": "<SPECIAL_220>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "221": {
+      "content": "<SPECIAL_221>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "222": {
+      "content": "<SPECIAL_222>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "223": {
+      "content": "<SPECIAL_223>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "224": {
+      "content": "<SPECIAL_224>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "225": {
+      "content": "<SPECIAL_225>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "226": {
+      "content": "<SPECIAL_226>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "227": {
+      "content": "<SPECIAL_227>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "228": {
+      "content": "<SPECIAL_228>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "229": {
+      "content": "<SPECIAL_229>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "230": {
+      "content": "<SPECIAL_230>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "231": {
+      "content": "<SPECIAL_231>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "232": {
+      "content": "<SPECIAL_232>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "233": {
+      "content": "<SPECIAL_233>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "234": {
+      "content": "<SPECIAL_234>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "235": {
+      "content": "<SPECIAL_235>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "236": {
+      "content": "<SPECIAL_236>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "237": {
+      "content": "<SPECIAL_237>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "238": {
+      "content": "<SPECIAL_238>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "239": {
+      "content": "<SPECIAL_239>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "240": {
+      "content": "<SPECIAL_240>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "241": {
+      "content": "<SPECIAL_241>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "242": {
+      "content": "<SPECIAL_242>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "243": {
+      "content": "<SPECIAL_243>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "244": {
+      "content": "<SPECIAL_244>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "245": {
+      "content": "<SPECIAL_245>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "246": {
+      "content": "<SPECIAL_246>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "247": {
+      "content": "<SPECIAL_247>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "248": {
+      "content": "<SPECIAL_248>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "249": {
+      "content": "<SPECIAL_249>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250": {
+      "content": "<SPECIAL_250>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "251": {
+      "content": "<SPECIAL_251>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "252": {
+      "content": "<SPECIAL_252>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "253": {
+      "content": "<SPECIAL_253>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "254": {
+      "content": "<SPECIAL_254>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "255": {
+      "content": "<SPECIAL_255>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "256": {
+      "content": "<SPECIAL_256>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "257": {
+      "content": "<SPECIAL_257>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "258": {
+      "content": "<SPECIAL_258>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "259": {
+      "content": "<SPECIAL_259>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "260": {
+      "content": "<SPECIAL_260>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "261": {
+      "content": "<SPECIAL_261>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "262": {
+      "content": "<SPECIAL_262>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "263": {
+      "content": "<SPECIAL_263>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "264": {
+      "content": "<SPECIAL_264>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "265": {
+      "content": "<SPECIAL_265>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "266": {
+      "content": "<SPECIAL_266>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "267": {
+      "content": "<SPECIAL_267>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "268": {
+      "content": "<SPECIAL_268>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "269": {
+      "content": "<SPECIAL_269>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "270": {
+      "content": "<SPECIAL_270>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "271": {
+      "content": "<SPECIAL_271>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "272": {
+      "content": "<SPECIAL_272>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "273": {
+      "content": "<SPECIAL_273>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "274": {
+      "content": "<SPECIAL_274>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "275": {
+      "content": "<SPECIAL_275>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "276": {
+      "content": "<SPECIAL_276>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "277": {
+      "content": "<SPECIAL_277>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "278": {
+      "content": "<SPECIAL_278>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "279": {
+      "content": "<SPECIAL_279>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "280": {
+      "content": "<SPECIAL_280>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "281": {
+      "content": "<SPECIAL_281>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "282": {
+      "content": "<SPECIAL_282>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "283": {
+      "content": "<SPECIAL_283>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "284": {
+      "content": "<SPECIAL_284>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "285": {
+      "content": "<SPECIAL_285>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "286": {
+      "content": "<SPECIAL_286>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "287": {
+      "content": "<SPECIAL_287>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "288": {
+      "content": "<SPECIAL_288>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "289": {
+      "content": "<SPECIAL_289>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "290": {
+      "content": "<SPECIAL_290>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "291": {
+      "content": "<SPECIAL_291>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "292": {
+      "content": "<SPECIAL_292>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "293": {
+      "content": "<SPECIAL_293>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "294": {
+      "content": "<SPECIAL_294>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "295": {
+      "content": "<SPECIAL_295>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "296": {
+      "content": "<SPECIAL_296>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "297": {
+      "content": "<SPECIAL_297>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "298": {
+      "content": "<SPECIAL_298>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "299": {
+      "content": "<SPECIAL_299>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "300": {
+      "content": "<SPECIAL_300>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "301": {
+      "content": "<SPECIAL_301>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "302": {
+      "content": "<SPECIAL_302>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "303": {
+      "content": "<SPECIAL_303>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "304": {
+      "content": "<SPECIAL_304>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "305": {
+      "content": "<SPECIAL_305>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "306": {
+      "content": "<SPECIAL_306>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "307": {
+      "content": "<SPECIAL_307>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "308": {
+      "content": "<SPECIAL_308>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "309": {
+      "content": "<SPECIAL_309>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "310": {
+      "content": "<SPECIAL_310>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "311": {
+      "content": "<SPECIAL_311>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "312": {
+      "content": "<SPECIAL_312>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "313": {
+      "content": "<SPECIAL_313>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "314": {
+      "content": "<SPECIAL_314>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "315": {
+      "content": "<SPECIAL_315>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "316": {
+      "content": "<SPECIAL_316>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "317": {
+      "content": "<SPECIAL_317>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "318": {
+      "content": "<SPECIAL_318>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "319": {
+      "content": "<SPECIAL_319>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "320": {
+      "content": "<SPECIAL_320>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "321": {
+      "content": "<SPECIAL_321>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "322": {
+      "content": "<SPECIAL_322>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "323": {
+      "content": "<SPECIAL_323>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "324": {
+      "content": "<SPECIAL_324>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "325": {
+      "content": "<SPECIAL_325>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "326": {
+      "content": "<SPECIAL_326>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "327": {
+      "content": "<SPECIAL_327>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "328": {
+      "content": "<SPECIAL_328>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "329": {
+      "content": "<SPECIAL_329>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "330": {
+      "content": "<SPECIAL_330>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "331": {
+      "content": "<SPECIAL_331>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "332": {
+      "content": "<SPECIAL_332>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "333": {
+      "content": "<SPECIAL_333>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "334": {
+      "content": "<SPECIAL_334>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "335": {
+      "content": "<SPECIAL_335>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "336": {
+      "content": "<SPECIAL_336>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "337": {
+      "content": "<SPECIAL_337>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "338": {
+      "content": "<SPECIAL_338>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "339": {
+      "content": "<SPECIAL_339>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "340": {
+      "content": "<SPECIAL_340>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "341": {
+      "content": "<SPECIAL_341>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "342": {
+      "content": "<SPECIAL_342>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "343": {
+      "content": "<SPECIAL_343>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "344": {
+      "content": "<SPECIAL_344>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "345": {
+      "content": "<SPECIAL_345>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "346": {
+      "content": "<SPECIAL_346>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "347": {
+      "content": "<SPECIAL_347>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "348": {
+      "content": "<SPECIAL_348>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "349": {
+      "content": "<SPECIAL_349>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "350": {
+      "content": "<SPECIAL_350>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "351": {
+      "content": "<SPECIAL_351>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "352": {
+      "content": "<SPECIAL_352>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "353": {
+      "content": "<SPECIAL_353>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "354": {
+      "content": "<SPECIAL_354>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "355": {
+      "content": "<SPECIAL_355>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "356": {
+      "content": "<SPECIAL_356>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "357": {
+      "content": "<SPECIAL_357>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "358": {
+      "content": "<SPECIAL_358>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "359": {
+      "content": "<SPECIAL_359>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "360": {
+      "content": "<SPECIAL_360>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "361": {
+      "content": "<SPECIAL_361>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "362": {
+      "content": "<SPECIAL_362>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "363": {
+      "content": "<SPECIAL_363>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "364": {
+      "content": "<SPECIAL_364>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "365": {
+      "content": "<SPECIAL_365>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "366": {
+      "content": "<SPECIAL_366>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "367": {
+      "content": "<SPECIAL_367>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "368": {
+      "content": "<SPECIAL_368>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "369": {
+      "content": "<SPECIAL_369>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "370": {
+      "content": "<SPECIAL_370>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "371": {
+      "content": "<SPECIAL_371>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "372": {
+      "content": "<SPECIAL_372>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "373": {
+      "content": "<SPECIAL_373>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "374": {
+      "content": "<SPECIAL_374>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "375": {
+      "content": "<SPECIAL_375>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "376": {
+      "content": "<SPECIAL_376>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "377": {
+      "content": "<SPECIAL_377>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "378": {
+      "content": "<SPECIAL_378>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "379": {
+      "content": "<SPECIAL_379>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "380": {
+      "content": "<SPECIAL_380>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "381": {
+      "content": "<SPECIAL_381>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "382": {
+      "content": "<SPECIAL_382>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "383": {
+      "content": "<SPECIAL_383>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "384": {
+      "content": "<SPECIAL_384>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "385": {
+      "content": "<SPECIAL_385>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "386": {
+      "content": "<SPECIAL_386>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "387": {
+      "content": "<SPECIAL_387>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "388": {
+      "content": "<SPECIAL_388>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "389": {
+      "content": "<SPECIAL_389>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "390": {
+      "content": "<SPECIAL_390>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "391": {
+      "content": "<SPECIAL_391>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "392": {
+      "content": "<SPECIAL_392>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "393": {
+      "content": "<SPECIAL_393>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "394": {
+      "content": "<SPECIAL_394>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "395": {
+      "content": "<SPECIAL_395>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "396": {
+      "content": "<SPECIAL_396>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "397": {
+      "content": "<SPECIAL_397>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "398": {
+      "content": "<SPECIAL_398>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "399": {
+      "content": "<SPECIAL_399>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "400": {
+      "content": "<SPECIAL_400>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "401": {
+      "content": "<SPECIAL_401>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "402": {
+      "content": "<SPECIAL_402>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "403": {
+      "content": "<SPECIAL_403>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "404": {
+      "content": "<SPECIAL_404>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "405": {
+      "content": "<SPECIAL_405>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "406": {
+      "content": "<SPECIAL_406>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "407": {
+      "content": "<SPECIAL_407>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "408": {
+      "content": "<SPECIAL_408>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "409": {
+      "content": "<SPECIAL_409>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "410": {
+      "content": "<SPECIAL_410>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "411": {
+      "content": "<SPECIAL_411>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "412": {
+      "content": "<SPECIAL_412>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "413": {
+      "content": "<SPECIAL_413>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "414": {
+      "content": "<SPECIAL_414>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "415": {
+      "content": "<SPECIAL_415>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "416": {
+      "content": "<SPECIAL_416>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "417": {
+      "content": "<SPECIAL_417>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "418": {
+      "content": "<SPECIAL_418>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "419": {
+      "content": "<SPECIAL_419>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "420": {
+      "content": "<SPECIAL_420>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "421": {
+      "content": "<SPECIAL_421>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "422": {
+      "content": "<SPECIAL_422>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "423": {
+      "content": "<SPECIAL_423>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "424": {
+      "content": "<SPECIAL_424>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "425": {
+      "content": "<SPECIAL_425>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "426": {
+      "content": "<SPECIAL_426>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "427": {
+      "content": "<SPECIAL_427>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "428": {
+      "content": "<SPECIAL_428>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "429": {
+      "content": "<SPECIAL_429>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "430": {
+      "content": "<SPECIAL_430>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "431": {
+      "content": "<SPECIAL_431>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "432": {
+      "content": "<SPECIAL_432>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "433": {
+      "content": "<SPECIAL_433>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "434": {
+      "content": "<SPECIAL_434>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "435": {
+      "content": "<SPECIAL_435>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "436": {
+      "content": "<SPECIAL_436>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "437": {
+      "content": "<SPECIAL_437>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "438": {
+      "content": "<SPECIAL_438>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "439": {
+      "content": "<SPECIAL_439>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "440": {
+      "content": "<SPECIAL_440>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "441": {
+      "content": "<SPECIAL_441>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "442": {
+      "content": "<SPECIAL_442>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "443": {
+      "content": "<SPECIAL_443>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "444": {
+      "content": "<SPECIAL_444>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "445": {
+      "content": "<SPECIAL_445>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "446": {
+      "content": "<SPECIAL_446>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "447": {
+      "content": "<SPECIAL_447>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "448": {
+      "content": "<SPECIAL_448>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "449": {
+      "content": "<SPECIAL_449>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "450": {
+      "content": "<SPECIAL_450>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "451": {
+      "content": "<SPECIAL_451>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "452": {
+      "content": "<SPECIAL_452>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "453": {
+      "content": "<SPECIAL_453>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "454": {
+      "content": "<SPECIAL_454>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "455": {
+      "content": "<SPECIAL_455>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "456": {
+      "content": "<SPECIAL_456>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "457": {
+      "content": "<SPECIAL_457>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "458": {
+      "content": "<SPECIAL_458>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "459": {
+      "content": "<SPECIAL_459>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "460": {
+      "content": "<SPECIAL_460>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "461": {
+      "content": "<SPECIAL_461>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "462": {
+      "content": "<SPECIAL_462>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "463": {
+      "content": "<SPECIAL_463>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "464": {
+      "content": "<SPECIAL_464>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "465": {
+      "content": "<SPECIAL_465>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "466": {
+      "content": "<SPECIAL_466>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "467": {
+      "content": "<SPECIAL_467>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "468": {
+      "content": "<SPECIAL_468>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "469": {
+      "content": "<SPECIAL_469>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "470": {
+      "content": "<SPECIAL_470>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "471": {
+      "content": "<SPECIAL_471>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "472": {
+      "content": "<SPECIAL_472>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "473": {
+      "content": "<SPECIAL_473>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "474": {
+      "content": "<SPECIAL_474>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "475": {
+      "content": "<SPECIAL_475>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "476": {
+      "content": "<SPECIAL_476>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "477": {
+      "content": "<SPECIAL_477>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "478": {
+      "content": "<SPECIAL_478>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "479": {
+      "content": "<SPECIAL_479>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "480": {
+      "content": "<SPECIAL_480>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "481": {
+      "content": "<SPECIAL_481>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "482": {
+      "content": "<SPECIAL_482>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "483": {
+      "content": "<SPECIAL_483>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "484": {
+      "content": "<SPECIAL_484>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "485": {
+      "content": "<SPECIAL_485>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "486": {
+      "content": "<SPECIAL_486>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "487": {
+      "content": "<SPECIAL_487>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "488": {
+      "content": "<SPECIAL_488>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "489": {
+      "content": "<SPECIAL_489>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "490": {
+      "content": "<SPECIAL_490>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "491": {
+      "content": "<SPECIAL_491>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "492": {
+      "content": "<SPECIAL_492>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "493": {
+      "content": "<SPECIAL_493>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "494": {
+      "content": "<SPECIAL_494>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "495": {
+      "content": "<SPECIAL_495>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "496": {
+      "content": "<SPECIAL_496>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "497": {
+      "content": "<SPECIAL_497>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "498": {
+      "content": "<SPECIAL_498>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "499": {
+      "content": "<SPECIAL_499>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "500": {
+      "content": "<SPECIAL_500>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "501": {
+      "content": "<SPECIAL_501>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "502": {
+      "content": "<SPECIAL_502>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "503": {
+      "content": "<SPECIAL_503>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "504": {
+      "content": "<SPECIAL_504>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "505": {
+      "content": "<SPECIAL_505>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "506": {
+      "content": "<SPECIAL_506>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "507": {
+      "content": "<SPECIAL_507>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "508": {
+      "content": "<SPECIAL_508>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "509": {
+      "content": "<SPECIAL_509>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "510": {
+      "content": "<SPECIAL_510>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "511": {
+      "content": "<SPECIAL_511>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "512": {
+      "content": "<SPECIAL_512>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "513": {
+      "content": "<SPECIAL_513>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "514": {
+      "content": "<SPECIAL_514>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "515": {
+      "content": "<SPECIAL_515>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "516": {
+      "content": "<SPECIAL_516>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "517": {
+      "content": "<SPECIAL_517>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "518": {
+      "content": "<SPECIAL_518>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "519": {
+      "content": "<SPECIAL_519>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "520": {
+      "content": "<SPECIAL_520>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "521": {
+      "content": "<SPECIAL_521>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "522": {
+      "content": "<SPECIAL_522>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "523": {
+      "content": "<SPECIAL_523>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "524": {
+      "content": "<SPECIAL_524>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "525": {
+      "content": "<SPECIAL_525>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "526": {
+      "content": "<SPECIAL_526>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "527": {
+      "content": "<SPECIAL_527>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "528": {
+      "content": "<SPECIAL_528>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "529": {
+      "content": "<SPECIAL_529>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "530": {
+      "content": "<SPECIAL_530>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "531": {
+      "content": "<SPECIAL_531>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "532": {
+      "content": "<SPECIAL_532>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "533": {
+      "content": "<SPECIAL_533>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "534": {
+      "content": "<SPECIAL_534>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "535": {
+      "content": "<SPECIAL_535>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "536": {
+      "content": "<SPECIAL_536>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "537": {
+      "content": "<SPECIAL_537>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "538": {
+      "content": "<SPECIAL_538>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "539": {
+      "content": "<SPECIAL_539>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "540": {
+      "content": "<SPECIAL_540>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "541": {
+      "content": "<SPECIAL_541>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "542": {
+      "content": "<SPECIAL_542>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "543": {
+      "content": "<SPECIAL_543>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "544": {
+      "content": "<SPECIAL_544>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "545": {
+      "content": "<SPECIAL_545>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "546": {
+      "content": "<SPECIAL_546>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "547": {
+      "content": "<SPECIAL_547>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "548": {
+      "content": "<SPECIAL_548>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "549": {
+      "content": "<SPECIAL_549>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "550": {
+      "content": "<SPECIAL_550>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "551": {
+      "content": "<SPECIAL_551>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "552": {
+      "content": "<SPECIAL_552>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "553": {
+      "content": "<SPECIAL_553>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "554": {
+      "content": "<SPECIAL_554>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "555": {
+      "content": "<SPECIAL_555>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "556": {
+      "content": "<SPECIAL_556>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "557": {
+      "content": "<SPECIAL_557>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "558": {
+      "content": "<SPECIAL_558>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "559": {
+      "content": "<SPECIAL_559>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "560": {
+      "content": "<SPECIAL_560>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "561": {
+      "content": "<SPECIAL_561>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "562": {
+      "content": "<SPECIAL_562>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "563": {
+      "content": "<SPECIAL_563>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "564": {
+      "content": "<SPECIAL_564>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "565": {
+      "content": "<SPECIAL_565>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "566": {
+      "content": "<SPECIAL_566>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "567": {
+      "content": "<SPECIAL_567>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "568": {
+      "content": "<SPECIAL_568>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "569": {
+      "content": "<SPECIAL_569>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "570": {
+      "content": "<SPECIAL_570>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "571": {
+      "content": "<SPECIAL_571>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "572": {
+      "content": "<SPECIAL_572>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "573": {
+      "content": "<SPECIAL_573>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "574": {
+      "content": "<SPECIAL_574>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "575": {
+      "content": "<SPECIAL_575>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "576": {
+      "content": "<SPECIAL_576>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "577": {
+      "content": "<SPECIAL_577>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "578": {
+      "content": "<SPECIAL_578>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "579": {
+      "content": "<SPECIAL_579>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "580": {
+      "content": "<SPECIAL_580>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "581": {
+      "content": "<SPECIAL_581>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "582": {
+      "content": "<SPECIAL_582>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "583": {
+      "content": "<SPECIAL_583>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "584": {
+      "content": "<SPECIAL_584>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "585": {
+      "content": "<SPECIAL_585>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "586": {
+      "content": "<SPECIAL_586>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "587": {
+      "content": "<SPECIAL_587>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "588": {
+      "content": "<SPECIAL_588>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "589": {
+      "content": "<SPECIAL_589>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "590": {
+      "content": "<SPECIAL_590>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "591": {
+      "content": "<SPECIAL_591>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "592": {
+      "content": "<SPECIAL_592>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "593": {
+      "content": "<SPECIAL_593>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "594": {
+      "content": "<SPECIAL_594>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "595": {
+      "content": "<SPECIAL_595>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "596": {
+      "content": "<SPECIAL_596>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "597": {
+      "content": "<SPECIAL_597>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "598": {
+      "content": "<SPECIAL_598>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "599": {
+      "content": "<SPECIAL_599>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "600": {
+      "content": "<SPECIAL_600>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "601": {
+      "content": "<SPECIAL_601>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "602": {
+      "content": "<SPECIAL_602>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "603": {
+      "content": "<SPECIAL_603>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "604": {
+      "content": "<SPECIAL_604>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "605": {
+      "content": "<SPECIAL_605>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "606": {
+      "content": "<SPECIAL_606>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "607": {
+      "content": "<SPECIAL_607>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "608": {
+      "content": "<SPECIAL_608>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "609": {
+      "content": "<SPECIAL_609>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "610": {
+      "content": "<SPECIAL_610>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "611": {
+      "content": "<SPECIAL_611>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "612": {
+      "content": "<SPECIAL_612>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "613": {
+      "content": "<SPECIAL_613>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "614": {
+      "content": "<SPECIAL_614>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "615": {
+      "content": "<SPECIAL_615>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "616": {
+      "content": "<SPECIAL_616>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "617": {
+      "content": "<SPECIAL_617>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "618": {
+      "content": "<SPECIAL_618>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "619": {
+      "content": "<SPECIAL_619>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "620": {
+      "content": "<SPECIAL_620>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "621": {
+      "content": "<SPECIAL_621>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "622": {
+      "content": "<SPECIAL_622>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "623": {
+      "content": "<SPECIAL_623>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "624": {
+      "content": "<SPECIAL_624>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "625": {
+      "content": "<SPECIAL_625>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "626": {
+      "content": "<SPECIAL_626>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "627": {
+      "content": "<SPECIAL_627>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "628": {
+      "content": "<SPECIAL_628>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "629": {
+      "content": "<SPECIAL_629>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "630": {
+      "content": "<SPECIAL_630>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "631": {
+      "content": "<SPECIAL_631>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "632": {
+      "content": "<SPECIAL_632>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "633": {
+      "content": "<SPECIAL_633>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "634": {
+      "content": "<SPECIAL_634>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "635": {
+      "content": "<SPECIAL_635>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "636": {
+      "content": "<SPECIAL_636>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "637": {
+      "content": "<SPECIAL_637>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "638": {
+      "content": "<SPECIAL_638>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "639": {
+      "content": "<SPECIAL_639>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "640": {
+      "content": "<SPECIAL_640>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "641": {
+      "content": "<SPECIAL_641>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "642": {
+      "content": "<SPECIAL_642>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "643": {
+      "content": "<SPECIAL_643>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "644": {
+      "content": "<SPECIAL_644>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "645": {
+      "content": "<SPECIAL_645>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "646": {
+      "content": "<SPECIAL_646>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "647": {
+      "content": "<SPECIAL_647>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "648": {
+      "content": "<SPECIAL_648>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "649": {
+      "content": "<SPECIAL_649>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "650": {
+      "content": "<SPECIAL_650>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "651": {
+      "content": "<SPECIAL_651>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "652": {
+      "content": "<SPECIAL_652>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "653": {
+      "content": "<SPECIAL_653>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "654": {
+      "content": "<SPECIAL_654>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "655": {
+      "content": "<SPECIAL_655>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "656": {
+      "content": "<SPECIAL_656>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "657": {
+      "content": "<SPECIAL_657>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "658": {
+      "content": "<SPECIAL_658>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "659": {
+      "content": "<SPECIAL_659>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "660": {
+      "content": "<SPECIAL_660>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "661": {
+      "content": "<SPECIAL_661>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "662": {
+      "content": "<SPECIAL_662>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "663": {
+      "content": "<SPECIAL_663>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "664": {
+      "content": "<SPECIAL_664>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "665": {
+      "content": "<SPECIAL_665>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "666": {
+      "content": "<SPECIAL_666>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "667": {
+      "content": "<SPECIAL_667>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "668": {
+      "content": "<SPECIAL_668>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "669": {
+      "content": "<SPECIAL_669>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "670": {
+      "content": "<SPECIAL_670>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "671": {
+      "content": "<SPECIAL_671>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "672": {
+      "content": "<SPECIAL_672>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "673": {
+      "content": "<SPECIAL_673>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "674": {
+      "content": "<SPECIAL_674>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "675": {
+      "content": "<SPECIAL_675>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "676": {
+      "content": "<SPECIAL_676>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "677": {
+      "content": "<SPECIAL_677>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "678": {
+      "content": "<SPECIAL_678>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "679": {
+      "content": "<SPECIAL_679>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "680": {
+      "content": "<SPECIAL_680>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "681": {
+      "content": "<SPECIAL_681>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "682": {
+      "content": "<SPECIAL_682>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "683": {
+      "content": "<SPECIAL_683>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "684": {
+      "content": "<SPECIAL_684>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "685": {
+      "content": "<SPECIAL_685>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "686": {
+      "content": "<SPECIAL_686>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "687": {
+      "content": "<SPECIAL_687>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "688": {
+      "content": "<SPECIAL_688>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "689": {
+      "content": "<SPECIAL_689>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "690": {
+      "content": "<SPECIAL_690>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "691": {
+      "content": "<SPECIAL_691>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "692": {
+      "content": "<SPECIAL_692>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "693": {
+      "content": "<SPECIAL_693>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "694": {
+      "content": "<SPECIAL_694>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "695": {
+      "content": "<SPECIAL_695>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "696": {
+      "content": "<SPECIAL_696>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "697": {
+      "content": "<SPECIAL_697>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "698": {
+      "content": "<SPECIAL_698>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "699": {
+      "content": "<SPECIAL_699>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "700": {
+      "content": "<SPECIAL_700>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "701": {
+      "content": "<SPECIAL_701>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "702": {
+      "content": "<SPECIAL_702>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "703": {
+      "content": "<SPECIAL_703>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "704": {
+      "content": "<SPECIAL_704>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "705": {
+      "content": "<SPECIAL_705>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "706": {
+      "content": "<SPECIAL_706>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "707": {
+      "content": "<SPECIAL_707>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "708": {
+      "content": "<SPECIAL_708>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "709": {
+      "content": "<SPECIAL_709>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "710": {
+      "content": "<SPECIAL_710>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "711": {
+      "content": "<SPECIAL_711>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "712": {
+      "content": "<SPECIAL_712>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "713": {
+      "content": "<SPECIAL_713>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "714": {
+      "content": "<SPECIAL_714>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "715": {
+      "content": "<SPECIAL_715>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "716": {
+      "content": "<SPECIAL_716>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "717": {
+      "content": "<SPECIAL_717>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "718": {
+      "content": "<SPECIAL_718>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "719": {
+      "content": "<SPECIAL_719>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "720": {
+      "content": "<SPECIAL_720>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "721": {
+      "content": "<SPECIAL_721>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "722": {
+      "content": "<SPECIAL_722>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "723": {
+      "content": "<SPECIAL_723>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "724": {
+      "content": "<SPECIAL_724>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "725": {
+      "content": "<SPECIAL_725>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "726": {
+      "content": "<SPECIAL_726>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "727": {
+      "content": "<SPECIAL_727>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "728": {
+      "content": "<SPECIAL_728>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "729": {
+      "content": "<SPECIAL_729>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "730": {
+      "content": "<SPECIAL_730>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "731": {
+      "content": "<SPECIAL_731>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "732": {
+      "content": "<SPECIAL_732>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "733": {
+      "content": "<SPECIAL_733>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "734": {
+      "content": "<SPECIAL_734>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "735": {
+      "content": "<SPECIAL_735>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "736": {
+      "content": "<SPECIAL_736>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "737": {
+      "content": "<SPECIAL_737>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "738": {
+      "content": "<SPECIAL_738>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "739": {
+      "content": "<SPECIAL_739>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "740": {
+      "content": "<SPECIAL_740>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "741": {
+      "content": "<SPECIAL_741>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "742": {
+      "content": "<SPECIAL_742>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "743": {
+      "content": "<SPECIAL_743>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "744": {
+      "content": "<SPECIAL_744>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "745": {
+      "content": "<SPECIAL_745>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "746": {
+      "content": "<SPECIAL_746>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "747": {
+      "content": "<SPECIAL_747>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "748": {
+      "content": "<SPECIAL_748>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "749": {
+      "content": "<SPECIAL_749>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "750": {
+      "content": "<SPECIAL_750>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "751": {
+      "content": "<SPECIAL_751>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "752": {
+      "content": "<SPECIAL_752>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "753": {
+      "content": "<SPECIAL_753>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "754": {
+      "content": "<SPECIAL_754>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "755": {
+      "content": "<SPECIAL_755>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "756": {
+      "content": "<SPECIAL_756>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "757": {
+      "content": "<SPECIAL_757>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "758": {
+      "content": "<SPECIAL_758>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "759": {
+      "content": "<SPECIAL_759>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "760": {
+      "content": "<SPECIAL_760>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "761": {
+      "content": "<SPECIAL_761>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "762": {
+      "content": "<SPECIAL_762>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "763": {
+      "content": "<SPECIAL_763>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "764": {
+      "content": "<SPECIAL_764>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "765": {
+      "content": "<SPECIAL_765>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "766": {
+      "content": "<SPECIAL_766>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "767": {
+      "content": "<SPECIAL_767>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "768": {
+      "content": "<SPECIAL_768>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "769": {
+      "content": "<SPECIAL_769>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "770": {
+      "content": "<SPECIAL_770>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "771": {
+      "content": "<SPECIAL_771>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "772": {
+      "content": "<SPECIAL_772>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "773": {
+      "content": "<SPECIAL_773>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "774": {
+      "content": "<SPECIAL_774>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "775": {
+      "content": "<SPECIAL_775>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "776": {
+      "content": "<SPECIAL_776>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "777": {
+      "content": "<SPECIAL_777>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "778": {
+      "content": "<SPECIAL_778>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "779": {
+      "content": "<SPECIAL_779>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "780": {
+      "content": "<SPECIAL_780>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "781": {
+      "content": "<SPECIAL_781>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "782": {
+      "content": "<SPECIAL_782>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "783": {
+      "content": "<SPECIAL_783>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "784": {
+      "content": "<SPECIAL_784>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "785": {
+      "content": "<SPECIAL_785>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "786": {
+      "content": "<SPECIAL_786>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "787": {
+      "content": "<SPECIAL_787>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "788": {
+      "content": "<SPECIAL_788>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "789": {
+      "content": "<SPECIAL_789>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "790": {
+      "content": "<SPECIAL_790>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "791": {
+      "content": "<SPECIAL_791>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "792": {
+      "content": "<SPECIAL_792>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "793": {
+      "content": "<SPECIAL_793>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "794": {
+      "content": "<SPECIAL_794>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "795": {
+      "content": "<SPECIAL_795>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "796": {
+      "content": "<SPECIAL_796>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "797": {
+      "content": "<SPECIAL_797>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "798": {
+      "content": "<SPECIAL_798>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "799": {
+      "content": "<SPECIAL_799>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "800": {
+      "content": "<SPECIAL_800>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "801": {
+      "content": "<SPECIAL_801>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "802": {
+      "content": "<SPECIAL_802>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "803": {
+      "content": "<SPECIAL_803>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "804": {
+      "content": "<SPECIAL_804>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "805": {
+      "content": "<SPECIAL_805>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "806": {
+      "content": "<SPECIAL_806>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "807": {
+      "content": "<SPECIAL_807>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "808": {
+      "content": "<SPECIAL_808>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "809": {
+      "content": "<SPECIAL_809>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "810": {
+      "content": "<SPECIAL_810>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "811": {
+      "content": "<SPECIAL_811>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "812": {
+      "content": "<SPECIAL_812>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "813": {
+      "content": "<SPECIAL_813>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "814": {
+      "content": "<SPECIAL_814>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "815": {
+      "content": "<SPECIAL_815>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "816": {
+      "content": "<SPECIAL_816>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "817": {
+      "content": "<SPECIAL_817>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "818": {
+      "content": "<SPECIAL_818>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "819": {
+      "content": "<SPECIAL_819>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "820": {
+      "content": "<SPECIAL_820>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "821": {
+      "content": "<SPECIAL_821>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "822": {
+      "content": "<SPECIAL_822>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "823": {
+      "content": "<SPECIAL_823>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "824": {
+      "content": "<SPECIAL_824>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "825": {
+      "content": "<SPECIAL_825>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "826": {
+      "content": "<SPECIAL_826>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "827": {
+      "content": "<SPECIAL_827>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "828": {
+      "content": "<SPECIAL_828>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "829": {
+      "content": "<SPECIAL_829>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "830": {
+      "content": "<SPECIAL_830>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "831": {
+      "content": "<SPECIAL_831>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "832": {
+      "content": "<SPECIAL_832>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "833": {
+      "content": "<SPECIAL_833>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "834": {
+      "content": "<SPECIAL_834>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "835": {
+      "content": "<SPECIAL_835>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "836": {
+      "content": "<SPECIAL_836>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "837": {
+      "content": "<SPECIAL_837>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "838": {
+      "content": "<SPECIAL_838>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "839": {
+      "content": "<SPECIAL_839>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "840": {
+      "content": "<SPECIAL_840>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "841": {
+      "content": "<SPECIAL_841>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "842": {
+      "content": "<SPECIAL_842>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "843": {
+      "content": "<SPECIAL_843>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "844": {
+      "content": "<SPECIAL_844>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "845": {
+      "content": "<SPECIAL_845>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "846": {
+      "content": "<SPECIAL_846>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "847": {
+      "content": "<SPECIAL_847>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "848": {
+      "content": "<SPECIAL_848>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "849": {
+      "content": "<SPECIAL_849>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "850": {
+      "content": "<SPECIAL_850>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "851": {
+      "content": "<SPECIAL_851>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "852": {
+      "content": "<SPECIAL_852>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "853": {
+      "content": "<SPECIAL_853>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "854": {
+      "content": "<SPECIAL_854>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "855": {
+      "content": "<SPECIAL_855>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "856": {
+      "content": "<SPECIAL_856>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "857": {
+      "content": "<SPECIAL_857>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "858": {
+      "content": "<SPECIAL_858>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "859": {
+      "content": "<SPECIAL_859>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "860": {
+      "content": "<SPECIAL_860>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "861": {
+      "content": "<SPECIAL_861>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "862": {
+      "content": "<SPECIAL_862>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "863": {
+      "content": "<SPECIAL_863>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "864": {
+      "content": "<SPECIAL_864>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "865": {
+      "content": "<SPECIAL_865>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "866": {
+      "content": "<SPECIAL_866>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "867": {
+      "content": "<SPECIAL_867>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "868": {
+      "content": "<SPECIAL_868>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "869": {
+      "content": "<SPECIAL_869>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "870": {
+      "content": "<SPECIAL_870>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "871": {
+      "content": "<SPECIAL_871>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "872": {
+      "content": "<SPECIAL_872>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "873": {
+      "content": "<SPECIAL_873>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "874": {
+      "content": "<SPECIAL_874>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "875": {
+      "content": "<SPECIAL_875>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "876": {
+      "content": "<SPECIAL_876>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "877": {
+      "content": "<SPECIAL_877>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "878": {
+      "content": "<SPECIAL_878>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "879": {
+      "content": "<SPECIAL_879>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "880": {
+      "content": "<SPECIAL_880>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "881": {
+      "content": "<SPECIAL_881>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "882": {
+      "content": "<SPECIAL_882>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "883": {
+      "content": "<SPECIAL_883>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "884": {
+      "content": "<SPECIAL_884>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "885": {
+      "content": "<SPECIAL_885>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "886": {
+      "content": "<SPECIAL_886>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "887": {
+      "content": "<SPECIAL_887>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "888": {
+      "content": "<SPECIAL_888>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "889": {
+      "content": "<SPECIAL_889>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "890": {
+      "content": "<SPECIAL_890>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "891": {
+      "content": "<SPECIAL_891>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "892": {
+      "content": "<SPECIAL_892>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "893": {
+      "content": "<SPECIAL_893>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "894": {
+      "content": "<SPECIAL_894>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "895": {
+      "content": "<SPECIAL_895>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "896": {
+      "content": "<SPECIAL_896>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "897": {
+      "content": "<SPECIAL_897>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "898": {
+      "content": "<SPECIAL_898>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "899": {
+      "content": "<SPECIAL_899>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "900": {
+      "content": "<SPECIAL_900>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "901": {
+      "content": "<SPECIAL_901>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "902": {
+      "content": "<SPECIAL_902>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "903": {
+      "content": "<SPECIAL_903>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "904": {
+      "content": "<SPECIAL_904>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "905": {
+      "content": "<SPECIAL_905>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "906": {
+      "content": "<SPECIAL_906>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "907": {
+      "content": "<SPECIAL_907>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "908": {
+      "content": "<SPECIAL_908>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "909": {
+      "content": "<SPECIAL_909>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "910": {
+      "content": "<SPECIAL_910>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "911": {
+      "content": "<SPECIAL_911>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "912": {
+      "content": "<SPECIAL_912>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "913": {
+      "content": "<SPECIAL_913>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "914": {
+      "content": "<SPECIAL_914>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "915": {
+      "content": "<SPECIAL_915>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "916": {
+      "content": "<SPECIAL_916>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "917": {
+      "content": "<SPECIAL_917>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "918": {
+      "content": "<SPECIAL_918>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "919": {
+      "content": "<SPECIAL_919>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "920": {
+      "content": "<SPECIAL_920>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "921": {
+      "content": "<SPECIAL_921>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "922": {
+      "content": "<SPECIAL_922>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "923": {
+      "content": "<SPECIAL_923>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "924": {
+      "content": "<SPECIAL_924>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "925": {
+      "content": "<SPECIAL_925>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "926": {
+      "content": "<SPECIAL_926>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "927": {
+      "content": "<SPECIAL_927>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "928": {
+      "content": "<SPECIAL_928>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "929": {
+      "content": "<SPECIAL_929>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "930": {
+      "content": "<SPECIAL_930>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "931": {
+      "content": "<SPECIAL_931>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "932": {
+      "content": "<SPECIAL_932>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "933": {
+      "content": "<SPECIAL_933>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "934": {
+      "content": "<SPECIAL_934>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "935": {
+      "content": "<SPECIAL_935>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "936": {
+      "content": "<SPECIAL_936>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "937": {
+      "content": "<SPECIAL_937>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "938": {
+      "content": "<SPECIAL_938>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "939": {
+      "content": "<SPECIAL_939>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "940": {
+      "content": "<SPECIAL_940>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "941": {
+      "content": "<SPECIAL_941>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "942": {
+      "content": "<SPECIAL_942>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "943": {
+      "content": "<SPECIAL_943>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "944": {
+      "content": "<SPECIAL_944>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "945": {
+      "content": "<SPECIAL_945>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "946": {
+      "content": "<SPECIAL_946>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "947": {
+      "content": "<SPECIAL_947>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "948": {
+      "content": "<SPECIAL_948>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "949": {
+      "content": "<SPECIAL_949>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "950": {
+      "content": "<SPECIAL_950>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "951": {
+      "content": "<SPECIAL_951>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "952": {
+      "content": "<SPECIAL_952>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "953": {
+      "content": "<SPECIAL_953>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "954": {
+      "content": "<SPECIAL_954>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "955": {
+      "content": "<SPECIAL_955>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "956": {
+      "content": "<SPECIAL_956>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "957": {
+      "content": "<SPECIAL_957>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "958": {
+      "content": "<SPECIAL_958>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "959": {
+      "content": "<SPECIAL_959>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "960": {
+      "content": "<SPECIAL_960>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "961": {
+      "content": "<SPECIAL_961>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "962": {
+      "content": "<SPECIAL_962>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "963": {
+      "content": "<SPECIAL_963>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "964": {
+      "content": "<SPECIAL_964>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "965": {
+      "content": "<SPECIAL_965>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "966": {
+      "content": "<SPECIAL_966>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "967": {
+      "content": "<SPECIAL_967>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "968": {
+      "content": "<SPECIAL_968>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "969": {
+      "content": "<SPECIAL_969>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "970": {
+      "content": "<SPECIAL_970>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "971": {
+      "content": "<SPECIAL_971>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "972": {
+      "content": "<SPECIAL_972>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "973": {
+      "content": "<SPECIAL_973>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "974": {
+      "content": "<SPECIAL_974>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "975": {
+      "content": "<SPECIAL_975>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "976": {
+      "content": "<SPECIAL_976>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "977": {
+      "content": "<SPECIAL_977>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "978": {
+      "content": "<SPECIAL_978>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "979": {
+      "content": "<SPECIAL_979>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "980": {
+      "content": "<SPECIAL_980>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "981": {
+      "content": "<SPECIAL_981>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "982": {
+      "content": "<SPECIAL_982>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "983": {
+      "content": "<SPECIAL_983>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "984": {
+      "content": "<SPECIAL_984>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "985": {
+      "content": "<SPECIAL_985>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "986": {
+      "content": "<SPECIAL_986>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "987": {
+      "content": "<SPECIAL_987>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "988": {
+      "content": "<SPECIAL_988>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "989": {
+      "content": "<SPECIAL_989>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "990": {
+      "content": "<SPECIAL_990>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "991": {
+      "content": "<SPECIAL_991>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "992": {
+      "content": "<SPECIAL_992>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "993": {
+      "content": "<SPECIAL_993>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "994": {
+      "content": "<SPECIAL_994>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "995": {
+      "content": "<SPECIAL_995>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "996": {
+      "content": "<SPECIAL_996>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "997": {
+      "content": "<SPECIAL_997>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "998": {
+      "content": "<SPECIAL_998>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "999": {
+      "content": "<SPECIAL_999>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<s>",
+  "chat_template": "{%- if messages[0][\"role\"] == \"system\" %}\n    {%- set system_message = messages[0][\"content\"] %}\n    {%- set loop_messages = messages[1:] %}\n{%- else %}\n    {%- set loop_messages = messages %}\n{%- endif %}\n{%- if not tools is defined %}\n    {%- set tools = none %}\n{%- endif %}\n{%- set user_messages = loop_messages | selectattr(\"role\", \"equalto\", \"user\") | list %}\n\n{#- This block checks for alternating user/assistant messages, skipping tool calling messages #}\n{%- set ns = namespace() %}\n{%- set ns.index = 0 %}\n{%- for message in loop_messages %}\n    {%- if not (message.role == \"tool\" or message.role == \"tool_results\" or (message.tool_calls is defined and message.tool_calls is not none)) %}\n        {%- if (message[\"role\"] == \"user\") != (ns.index % 2 == 0) %}\n            {{- raise_exception(\"After the optional system message, conversation roles must alternate user/assistant/user/assistant/...\") }}\n        {%- endif %}\n        {%- set ns.index = ns.index + 1 %}\n    {%- endif %}\n{%- endfor %}\n\n{{- bos_token }}\n{%- for message in loop_messages %}\n    {%- if message[\"role\"] == \"user\" %}\n        {%- if tools is not none and (message == user_messages[-1]) %}\n            {{- \"[AVAILABLE_TOOLS][\" }}\n            {%- for tool in tools %}\n                {%- set tool = tool.function %}\n                {{- '{\"type\": \"function\", \"function\": {' }}\n                {%- for key, val in tool.items() if key != \"return\" %}\n                    {%- if val is string %}\n                        {{- '\"' + key + '\": \"' + val + '\"' }}\n                    {%- else %}\n                        {{- '\"' + key + '\": ' + val|tojson }}\n                    {%- endif %}\n                    {%- if not loop.last %}\n                        {{- \", \" }}\n                    {%- endif %}\n                {%- endfor %}\n                {{- \"}}\" }}\n                {%- if not loop.last %}\n                    {{- \", \" }}\n                {%- else %}\n                    {{- \"]\" }}\n                {%- endif %}\n            {%- endfor %}\n            {{- \"[/AVAILABLE_TOOLS]\" }}\n            {%- endif %}\n        {%- if loop.last and system_message is defined %}\n            {{- \"[INST]\" + system_message + \"\\n\\n\" + message[\"content\"] + \"[/INST]\" }}\n        {%- else %}\n            {{- \"[INST]\" + message[\"content\"] + \"[/INST]\" }}\n        {%- endif %}\n    {%- elif (message.tool_calls is defined and message.tool_calls is not none) %}\n        {{- \"[TOOL_CALLS][\" }}\n        {%- for tool_call in message.tool_calls %}\n            {%- set out = tool_call.function|tojson %}\n            {{- out[:-1] }}\n            {%- if not tool_call.id is defined or tool_call.id|length != 9 %}\n                {{- raise_exception(\"Tool call IDs should be alphanumeric strings with length 9!\") }}\n            {%- endif %}\n            {{- ', \"id\": \"' + tool_call.id + '\"}' }}\n            {%- if not loop.last %}\n                {{- \", \" }}\n            {%- else %}\n                {{- \"]\" + eos_token }}\n            {%- endif %}\n        {%- endfor %}\n    {%- elif message[\"role\"] == \"assistant\" %}\n        {{- message[\"content\"] + eos_token}}\n    {%- elif message[\"role\"] == \"tool_results\" or message[\"role\"] == \"tool\" %}\n        {%- if message.content is defined and message.content.content is defined %}\n            {%- set content = message.content.content %}\n        {%- else %}\n            {%- set content = message.content %}\n        {%- endif %}\n        {{- '[TOOL_RESULTS]{\"content\": ' + content|string + \", \" }}\n        {%- if not message.tool_call_id is defined or message.tool_call_id|length != 9 %}\n            {{- raise_exception(\"Tool call IDs should be alphanumeric strings with length 9!\") }}\n        {%- endif %}\n        {{- '\"call_id\": \"' + message.tool_call_id + '\"}[/TOOL_RESULTS]' }}\n    {%- else %}\n        {{- raise_exception(\"Only user and assistant roles are supported, with the exception of an initial optional system message!\") }}\n    {%- endif %}\n{%- endfor %}\n",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "</s>",
+  "extra_special_tokens": {},
+  "model_input_names": [
+    "input_ids",
+    "attention_mask"
+  ],
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "</s>",
+  "tokenizer_class": "PreTrainedTokenizer",
+  "unk_token": "<unk>"
+}
diff --git a/checkpoint-900/trainer_state.json b/checkpoint-900/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..76d83fd0cf1308476c0e8585b071b799011d0b19
--- /dev/null
+++ b/checkpoint-900/trainer_state.json
@@ -0,0 +1,11284 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 1.2601398601398601,
+  "eval_steps": 500,
+  "global_step": 900,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 94.14286041259766,
+      "epoch": 0.0013986013986013986,
+      "grad_norm": 0.09098726750251449,
+      "kl": 0.0003147125244140625,
+      "learning_rate": 3.496503496503497e-07,
+      "loss": -0.0001,
+      "num_tokens": 19338.0,
+      "reward": 1.3853150606155396,
+      "reward_std": 0.4827312231063843,
+      "rewards/check_gptzero_func": 0.3214285671710968,
+      "rewards/check_perplexity_diff_func": 0.6428571343421936,
+      "rewards/check_winston_local_func": 0.42102929949760437,
+      "step": 1
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.002797202797202797,
+      "grad_norm": 0.09100260462019598,
+      "kl": 0.0003147125244140625,
+      "learning_rate": 6.993006993006994e-07,
+      "loss": -0.0001,
+      "step": 2
+    },
+    {
+      "clip_ratio": 0.0031990089919418097,
+      "completion_length": 97.73214721679688,
+      "epoch": 0.004195804195804196,
+      "grad_norm": 0.16978190242842756,
+      "kl": 0.0003662109375,
+      "learning_rate": 1.0489510489510491e-06,
+      "loss": -0.0017,
+      "num_tokens": 39487.0,
+      "reward": 0.9710169434547424,
+      "reward_std": 0.6376833319664001,
+      "rewards/check_gptzero_func": 0.125,
+      "rewards/check_perplexity_diff_func": 0.5,
+      "rewards/check_winston_local_func": 0.34601688385009766,
+      "step": 3
+    },
+    {
+      "clip_ratio": 0.0018373647471889853,
+      "epoch": 0.005594405594405594,
+      "grad_norm": 0.17663660241053314,
+      "kl": 0.000423431396484375,
+      "learning_rate": 1.3986013986013987e-06,
+      "loss": -0.0013,
+      "step": 4
+    },
+    {
+      "clip_ratio": 0.0016871786210685968,
+      "completion_length": 72.83928680419922,
+      "epoch": 0.006993006993006993,
+      "grad_norm": 0.09466520543424979,
+      "kl": 0.00034332275390625,
+      "learning_rate": 1.7482517482517485e-06,
+      "loss": 0.0008,
+      "num_tokens": 56082.0,
+      "reward": 1.0736479759216309,
+      "reward_std": 0.6862176656723022,
+      "rewards/check_gptzero_func": 0.2321428507566452,
+      "rewards/check_perplexity_diff_func": 0.3928571343421936,
+      "rewards/check_winston_local_func": 0.4486479163169861,
+      "step": 5
+    },
+    {
+      "clip_ratio": 0.0010098000057041645,
+      "epoch": 0.008391608391608392,
+      "grad_norm": 0.09539550614656178,
+      "kl": 0.0003376007080078125,
+      "learning_rate": 2.0979020979020983e-06,
+      "loss": 0.001,
+      "step": 6
+    },
+    {
+      "clip_ratio": 0.00184189947322011,
+      "completion_length": 95.92857360839844,
+      "epoch": 0.009790209790209791,
+      "grad_norm": 0.10573631695411624,
+      "kl": 0.0003757476806640625,
+      "learning_rate": 2.4475524475524477e-06,
+      "loss": 0.001,
+      "num_tokens": 75906.0,
+      "reward": 1.135968804359436,
+      "reward_std": 0.48974505066871643,
+      "rewards/check_gptzero_func": 0.2142857164144516,
+      "rewards/check_perplexity_diff_func": 0.5,
+      "rewards/check_winston_local_func": 0.42168304324150085,
+      "step": 7
+    },
+    {
+      "clip_ratio": 0.0018530809320509434,
+      "epoch": 0.011188811188811189,
+      "grad_norm": 0.10614532740192534,
+      "kl": 0.0004119873046875,
+      "learning_rate": 2.7972027972027974e-06,
+      "loss": 0.0011,
+      "step": 8
+    },
+    {
+      "clip_ratio": 0.0017129909247159958,
+      "completion_length": 106.76786041259766,
+      "epoch": 0.012587412587412588,
+      "grad_norm": 0.07916450383171113,
+      "kl": 0.000377655029296875,
+      "learning_rate": 3.1468531468531472e-06,
+      "loss": -0.0012,
+      "num_tokens": 96989.0,
+      "reward": 0.9590713977813721,
+      "reward_std": 0.5206417441368103,
+      "rewards/check_gptzero_func": 0.2321428507566452,
+      "rewards/check_perplexity_diff_func": 0.3928571343421936,
+      "rewards/check_winston_local_func": 0.3340713679790497,
+      "step": 9
+    },
+    {
+      "clip_ratio": 0.0034840807784348726,
+      "epoch": 0.013986013986013986,
+      "grad_norm": 0.08061029037170311,
+      "kl": 0.00037384033203125,
+      "learning_rate": 3.496503496503497e-06,
+      "loss": -0.0008,
+      "step": 10
+    },
+    {
+      "clip_ratio": 0.0017933619674295187,
+      "completion_length": 102.85714721679688,
+      "epoch": 0.015384615384615385,
+      "grad_norm": 0.07930375667041344,
+      "kl": 0.00030517578125,
+      "learning_rate": 3.846153846153847e-06,
+      "loss": -0.0038,
+      "num_tokens": 117271.0,
+      "reward": 1.1903148889541626,
+      "reward_std": 0.4722224473953247,
+      "rewards/check_gptzero_func": 0.2857142984867096,
+      "rewards/check_perplexity_diff_func": 0.5357142686843872,
+      "rewards/check_winston_local_func": 0.368886262178421,
+      "step": 11
+    },
+    {
+      "clip_ratio": 0.0015239757485687733,
+      "epoch": 0.016783216783216783,
+      "grad_norm": 0.07898151410343454,
+      "kl": 0.0003871917724609375,
+      "learning_rate": 4.195804195804197e-06,
+      "loss": -0.0041,
+      "step": 12
+    },
+    {
+      "clip_ratio": 0.0029670600779354572,
+      "completion_length": 104.37500762939453,
+      "epoch": 0.01818181818181818,
+      "grad_norm": 0.09247854491588844,
+      "kl": 0.0003986358642578125,
+      "learning_rate": 4.5454545454545455e-06,
+      "loss": -0.0049,
+      "num_tokens": 138174.0,
+      "reward": 0.9787324070930481,
+      "reward_std": 0.63657546043396,
+      "rewards/check_gptzero_func": 0.1964285671710968,
+      "rewards/check_perplexity_diff_func": 0.5714285969734192,
+      "rewards/check_winston_local_func": 0.21087531745433807,
+      "step": 13
+    },
+    {
+      "clip_ratio": 0.003217793768271804,
+      "epoch": 0.019580419580419582,
+      "grad_norm": 0.09293749933034297,
+      "kl": 0.0003204345703125,
+      "learning_rate": 4.895104895104895e-06,
+      "loss": -0.0045,
+      "step": 14
+    },
+    {
+      "clip_ratio": 0.002677519340068102,
+      "completion_length": 67.41072082519531,
+      "epoch": 0.02097902097902098,
+      "grad_norm": 0.14020673218962584,
+      "kl": 0.000499725341796875,
+      "learning_rate": 5.244755244755245e-06,
+      "loss": -0.008,
+      "num_tokens": 154385.0,
+      "reward": 1.0573337078094482,
+      "reward_std": 0.5388314127922058,
+      "rewards/check_gptzero_func": 0.2857142984867096,
+      "rewards/check_perplexity_diff_func": 0.3571428656578064,
+      "rewards/check_winston_local_func": 0.41447654366493225,
+      "step": 15
+    },
+    {
+      "clip_ratio": 0.0021408216562122107,
+      "epoch": 0.022377622377622378,
+      "grad_norm": 0.16110555979669544,
+      "kl": 0.000530242919921875,
+      "learning_rate": 5.594405594405595e-06,
+      "loss": -0.0081,
+      "step": 16
+    },
+    {
+      "clip_ratio": 0.002852825215086341,
+      "completion_length": 89.83928680419922,
+      "epoch": 0.023776223776223775,
+      "grad_norm": 0.1066932843002634,
+      "kl": 0.000415802001953125,
+      "learning_rate": 5.944055944055945e-06,
+      "loss": 0.0166,
+      "num_tokens": 173158.0,
+      "reward": 1.2226747274398804,
+      "reward_std": 0.45578521490097046,
+      "rewards/check_gptzero_func": 0.3392857015132904,
+      "rewards/check_perplexity_diff_func": 0.5,
+      "rewards/check_winston_local_func": 0.3833889663219452,
+      "step": 17
+    },
+    {
+      "clip_ratio": 0.002018069615587592,
+      "epoch": 0.025174825174825177,
+      "grad_norm": 0.10578254537451186,
+      "kl": 0.0003833770751953125,
+      "learning_rate": 6.2937062937062944e-06,
+      "loss": 0.0164,
+      "step": 18
+    },
+    {
+      "clip_ratio": 0.0032663617748767138,
+      "completion_length": 105.83928680419922,
+      "epoch": 0.026573426573426574,
+      "grad_norm": 0.11554707163597037,
+      "kl": 0.0003719329833984375,
+      "learning_rate": 6.643356643356643e-06,
+      "loss": -0.0005,
+      "num_tokens": 194649.0,
+      "reward": 0.8957912921905518,
+      "reward_std": 0.47702187299728394,
+      "rewards/check_gptzero_func": 0.1785714328289032,
+      "rewards/check_perplexity_diff_func": 0.3571428656578064,
+      "rewards/check_winston_local_func": 0.3600769340991974,
+      "step": 19
+    },
+    {
+      "clip_ratio": 0.0024422537535429,
+      "epoch": 0.027972027972027972,
+      "grad_norm": 0.11417666172406592,
+      "kl": 0.0003814697265625,
+      "learning_rate": 6.993006993006994e-06,
+      "loss": -0.0008,
+      "step": 20
+    },
+    {
+      "clip_ratio": 0.0013639701064676046,
+      "completion_length": 107.39286041259766,
+      "epoch": 0.02937062937062937,
+      "grad_norm": 0.09634196153318929,
+      "kl": 0.000396728515625,
+      "learning_rate": 7.342657342657343e-06,
+      "loss": -0.0011,
+      "num_tokens": 215987.0,
+      "reward": 0.9197577238082886,
+      "reward_std": 0.39100831747055054,
+      "rewards/check_gptzero_func": 0.2321428507566452,
+      "rewards/check_perplexity_diff_func": 0.3571428656578064,
+      "rewards/check_winston_local_func": 0.33047202229499817,
+      "step": 21
+    },
+    {
+      "clip_ratio": 0.003595889313146472,
+      "epoch": 0.03076923076923077,
+      "grad_norm": 0.09779857975187953,
+      "kl": 0.00052642822265625,
+      "learning_rate": 7.692307692307694e-06,
+      "loss": -0.0013,
+      "step": 22
+    },
+    {
+      "clip_ratio": 0.0024316231720149517,
+      "completion_length": 89.08928680419922,
+      "epoch": 0.032167832167832165,
+      "grad_norm": 0.13189477135999747,
+      "kl": 0.0003414154052734375,
+      "learning_rate": 8.041958041958042e-06,
+      "loss": -0.0009,
+      "num_tokens": 234678.0,
+      "reward": 0.9593304395675659,
+      "reward_std": 0.5972030758857727,
+      "rewards/check_gptzero_func": 0.1964285671710968,
+      "rewards/check_perplexity_diff_func": 0.4285714328289032,
+      "rewards/check_winston_local_func": 0.33433040976524353,
+      "step": 23
+    },
+    {
+      "clip_ratio": 0.000728283659555018,
+      "epoch": 0.033566433566433566,
+      "grad_norm": 0.12953291972672934,
+      "kl": 0.000335693359375,
+      "learning_rate": 8.391608391608393e-06,
+      "loss": -0.0011,
+      "step": 24
+    },
+    {
+      "clip_ratio": 0.0021546650677919388,
+      "completion_length": 90.98214721679688,
+      "epoch": 0.03496503496503497,
+      "grad_norm": 0.07682474826469807,
+      "kl": 0.0004558563232421875,
+      "learning_rate": 8.741258741258741e-06,
+      "loss": 0.0077,
+      "num_tokens": 253833.0,
+      "reward": 0.8255766034126282,
+      "reward_std": 0.3988816440105438,
+      "rewards/check_gptzero_func": 0.2678571343421936,
+      "rewards/check_perplexity_diff_func": 0.2857142984867096,
+      "rewards/check_winston_local_func": 0.272005170583725,
+      "step": 25
+    },
+    {
+      "clip_ratio": 0.0019250252516940236,
+      "epoch": 0.03636363636363636,
+      "grad_norm": 0.07873511079065725,
+      "kl": 0.0003490447998046875,
+      "learning_rate": 9.090909090909091e-06,
+      "loss": 0.0075,
+      "step": 26
+    },
+    {
+      "clip_ratio": 0.001886485842987895,
+      "completion_length": 74.4464340209961,
+      "epoch": 0.03776223776223776,
+      "grad_norm": 0.06765277577538054,
+      "kl": 0.000354766845703125,
+      "learning_rate": 9.44055944055944e-06,
+      "loss": -0.0003,
+      "num_tokens": 270908.0,
+      "reward": 1.3086735010147095,
+      "reward_std": 0.3881511986255646,
+      "rewards/check_gptzero_func": 0.4107142984867096,
+      "rewards/check_perplexity_diff_func": 0.4642857015132904,
+      "rewards/check_winston_local_func": 0.4336733818054199,
+      "step": 27
+    },
+    {
+      "clip_ratio": 0.0020898371003568172,
+      "epoch": 0.039160839160839164,
+      "grad_norm": 0.06722581226675682,
+      "kl": 0.000453948974609375,
+      "learning_rate": 9.79020979020979e-06,
+      "loss": -0.0006,
+      "step": 28
+    },
+    {
+      "clip_ratio": 0.002855924190953374,
+      "completion_length": 103.64286041259766,
+      "epoch": 0.04055944055944056,
+      "grad_norm": 0.16301390061045865,
+      "kl": 0.0003948211669921875,
+      "learning_rate": 1.013986013986014e-05,
+      "loss": 0.0063,
+      "num_tokens": 292060.0,
+      "reward": 1.22488272190094,
+      "reward_std": 0.8093323111534119,
+      "rewards/check_gptzero_func": 0.2678571343421936,
+      "rewards/check_perplexity_diff_func": 0.6071428656578064,
+      "rewards/check_winston_local_func": 0.34988275170326233,
+      "step": 29
+    },
+    {
+      "clip_ratio": 0.0016208746237680316,
+      "epoch": 0.04195804195804196,
+      "grad_norm": 0.17335125351231856,
+      "kl": 0.000507354736328125,
+      "learning_rate": 1.048951048951049e-05,
+      "loss": 0.0059,
+      "step": 30
+    },
+    {
+      "clip_ratio": 0.0011572305811569095,
+      "completion_length": 82.91072082519531,
+      "epoch": 0.043356643356643354,
+      "grad_norm": 0.11830097508590764,
+      "kl": 0.0006256103515625,
+      "learning_rate": 1.0839160839160838e-05,
+      "loss": 0.002,
+      "num_tokens": 310163.0,
+      "reward": 0.9069231748580933,
+      "reward_std": 0.4621748924255371,
+      "rewards/check_gptzero_func": 0.25,
+      "rewards/check_perplexity_diff_func": 0.3571428656578064,
+      "rewards/check_winston_local_func": 0.29978030920028687,
+      "step": 31
+    },
+    {
+      "clip_ratio": 0.00402362085878849,
+      "epoch": 0.044755244755244755,
+      "grad_norm": 0.11861334422671065,
+      "kl": 0.000762939453125,
+      "learning_rate": 1.118881118881119e-05,
+      "loss": 0.0019,
+      "step": 32
+    },
+    {
+      "clip_ratio": 0.0017474278574809432,
+      "completion_length": 90.9464340209961,
+      "epoch": 0.046153846153846156,
+      "grad_norm": 0.1025496804751534,
+      "kl": 0.000701904296875,
+      "learning_rate": 1.153846153846154e-05,
+      "loss": -0.0014,
+      "num_tokens": 328922.0,
+      "reward": 1.1198338270187378,
+      "reward_std": 0.5045111775398254,
+      "rewards/check_gptzero_func": 0.0892857164144516,
+      "rewards/check_perplexity_diff_func": 0.7857142686843872,
+      "rewards/check_winston_local_func": 0.24483375251293182,
+      "step": 33
+    },
+    {
+      "clip_ratio": 0.0012307984288781881,
+      "epoch": 0.04755244755244755,
+      "grad_norm": 0.10274008519956918,
+      "kl": 0.00087738037109375,
+      "learning_rate": 1.188811188811189e-05,
+      "loss": -0.0015,
+      "step": 34
+    },
+    {
+      "clip_ratio": 0.0029886537231504917,
+      "completion_length": 93.00000762939453,
+      "epoch": 0.04895104895104895,
+      "grad_norm": 0.1311802256012859,
+      "kl": 0.001190185546875,
+      "learning_rate": 1.2237762237762239e-05,
+      "loss": -0.0141,
+      "num_tokens": 348852.0,
+      "reward": 0.9456299543380737,
+      "reward_std": 0.6374980807304382,
+      "rewards/check_gptzero_func": 0.2321428507566452,
+      "rewards/check_perplexity_diff_func": 0.4285714328289032,
+      "rewards/check_winston_local_func": 0.28491565585136414,
+      "step": 35
+    },
+    {
+      "clip_ratio": 0.0028984802775084972,
+      "epoch": 0.05034965034965035,
+      "grad_norm": 0.13267305607921692,
+      "kl": 0.00145721435546875,
+      "learning_rate": 1.2587412587412589e-05,
+      "loss": -0.0147,
+      "step": 36
+    },
+    {
+      "clip_ratio": 0.0018746532732620835,
+      "completion_length": 89.85714721679688,
+      "epoch": 0.05174825174825175,
+      "grad_norm": 0.1497182815907709,
+      "kl": 0.002044677734375,
+      "learning_rate": 1.2937062937062939e-05,
+      "loss": -0.0253,
+      "num_tokens": 367482.0,
+      "reward": 1.2605940103530884,
+      "reward_std": 0.66156405210495,
+      "rewards/check_gptzero_func": 0.1785714328289032,
+      "rewards/check_perplexity_diff_func": 0.75,
+      "rewards/check_winston_local_func": 0.3320225179195404,
+      "step": 37
+    },
+    {
+      "clip_ratio": 0.0020914783235639334,
+      "epoch": 0.05314685314685315,
+      "grad_norm": 0.15027543310666272,
+      "kl": 0.0026092529296875,
+      "learning_rate": 1.3286713286713287e-05,
+      "loss": -0.0257,
+      "step": 38
+    },
+    {
+      "clip_ratio": 0.0017028081929311156,
+      "completion_length": 101.58928680419922,
+      "epoch": 0.05454545454545454,
+      "grad_norm": 0.1306748742984494,
+      "kl": 0.0027008056640625,
+      "learning_rate": 1.3636363636363637e-05,
+      "loss": -0.0091,
+      "num_tokens": 388365.0,
+      "reward": 1.1187876462936401,
+      "reward_std": 0.4044211804866791,
+      "rewards/check_gptzero_func": 0.2321428507566452,
+      "rewards/check_perplexity_diff_func": 0.5357142686843872,
+      "rewards/check_winston_local_func": 0.3509305417537689,
+      "step": 39
+    },
+    {
+      "clip_ratio": 0.003386714030057192,
+      "epoch": 0.055944055944055944,
+      "grad_norm": 0.12880289464580882,
+      "kl": 0.003265380859375,
+      "learning_rate": 1.3986013986013988e-05,
+      "loss": -0.0086,
+      "step": 40
+    },
+    {
+      "clip_ratio": 0.0024523374158889055,
+      "completion_length": 100.41072082519531,
+      "epoch": 0.057342657342657345,
+      "grad_norm": 0.12385816200639352,
+      "kl": 0.004547119140625,
+      "learning_rate": 1.4335664335664336e-05,
+      "loss": -0.0161,
+      "num_tokens": 409074.0,
+      "reward": 0.804709792137146,
+      "reward_std": 0.5129754543304443,
+      "rewards/check_gptzero_func": 0.1428571492433548,
+      "rewards/check_perplexity_diff_func": 0.3928571343421936,
+      "rewards/check_winston_local_func": 0.2689954936504364,
+      "step": 41
+    },
+    {
+      "clip_ratio": 0.004267544951289892,
+      "epoch": 0.05874125874125874,
+      "grad_norm": 0.12301160152684054,
+      "kl": 0.005218505859375,
+      "learning_rate": 1.4685314685314686e-05,
+      "loss": -0.0161,
+      "step": 42
+    },
+    {
+      "clip_ratio": 0.0019341636216267943,
+      "completion_length": 90.46428680419922,
+      "epoch": 0.06013986013986014,
+      "grad_norm": 0.1390908713431379,
+      "kl": 0.0048828125,
+      "learning_rate": 1.5034965034965034e-05,
+      "loss": 0.0068,
+      "num_tokens": 428262.0,
+      "reward": 1.087609887123108,
+      "reward_std": 0.6448312997817993,
+      "rewards/check_gptzero_func": 0.2321428507566452,
+      "rewards/check_perplexity_diff_func": 0.3928571343421936,
+      "rewards/check_winston_local_func": 0.46260982751846313,
+      "step": 43
+    },
+    {
+      "clip_ratio": 0.0016832423862069845,
+      "epoch": 0.06153846153846154,
+      "grad_norm": 0.13873080950458352,
+      "kl": 0.00567626953125,
+      "learning_rate": 1.5384615384615387e-05,
+      "loss": 0.0065,
+      "step": 44
+    },
+    {
+      "clip_ratio": 0.0019954824820160866,
+      "completion_length": 96.92857360839844,
+      "epoch": 0.06293706293706294,
+      "grad_norm": 0.11630124487143968,
+      "kl": 0.00897216796875,
+      "learning_rate": 1.5734265734265734e-05,
+      "loss": 0.008,
+      "num_tokens": 448480.0,
+      "reward": 1.0604994297027588,
+      "reward_std": 0.5378546118736267,
+      "rewards/check_gptzero_func": 0.2321428507566452,
+      "rewards/check_perplexity_diff_func": 0.4642857015132904,
+      "rewards/check_winston_local_func": 0.36407074332237244,
+      "step": 45
+    },
+    {
+      "clip_ratio": 0.002663462422788143,
+      "epoch": 0.06433566433566433,
+      "grad_norm": 0.1149796219701467,
+      "kl": 0.01007080078125,
+      "learning_rate": 1.6083916083916083e-05,
+      "loss": 0.0081,
+      "step": 46
+    },
+    {
+      "clip_ratio": 0.0018039485439658165,
+      "completion_length": 79.51786041259766,
+      "epoch": 0.06573426573426573,
+      "grad_norm": 0.12721969778213826,
+      "kl": 0.01123046875,
+      "learning_rate": 1.6433566433566433e-05,
+      "loss": 0.0112,
+      "num_tokens": 465671.0,
+      "reward": 0.9921315312385559,
+      "reward_std": 0.38579052686691284,
+      "rewards/check_gptzero_func": 0.1607142835855484,
+      "rewards/check_perplexity_diff_func": 0.3928571343421936,
+      "rewards/check_winston_local_func": 0.43856000900268555,
+      "step": 47
+    },
+    {
+      "clip_ratio": 0.0027954555116593838,
+      "epoch": 0.06713286713286713,
+      "grad_norm": 0.12930272448754576,
+      "kl": 0.01300048828125,
+      "learning_rate": 1.6783216783216786e-05,
+      "loss": 0.0107,
+      "step": 48
+    },
+    {
+      "clip_ratio": 0.0028269642498344183,
+      "completion_length": 116.5714340209961,
+      "epoch": 0.06853146853146853,
+      "grad_norm": 0.12430052891449103,
+      "kl": 0.01226806640625,
+      "learning_rate": 1.7132867132867133e-05,
+      "loss": 0.025,
+      "num_tokens": 488367.0,
+      "reward": 0.8830849528312683,
+      "reward_std": 0.5276607871055603,
+      "rewards/check_gptzero_func": 0.25,
+      "rewards/check_perplexity_diff_func": 0.3214285671710968,
+      "rewards/check_winston_local_func": 0.31165632605552673,
+      "step": 49
+    },
+    {
+      "clip_ratio": 0.0023628019262105227,
+      "epoch": 0.06993006993006994,
+      "grad_norm": 0.12293590613717971,
+      "kl": 0.0142822265625,
+      "learning_rate": 1.7482517482517483e-05,
+      "loss": 0.0249,
+      "step": 50
+    },
+    {
+      "clip_ratio": 0.0033425339497625828,
+      "completion_length": 83.39286041259766,
+      "epoch": 0.07132867132867132,
+      "grad_norm": 0.11106847343390151,
+      "kl": 0.01177978515625,
+      "learning_rate": 1.7832167832167836e-05,
+      "loss": -0.0041,
+      "num_tokens": 506551.0,
+      "reward": 1.029296875,
+      "reward_std": 0.4901208281517029,
+      "rewards/check_gptzero_func": 0.2857142984867096,
+      "rewards/check_perplexity_diff_func": 0.3214285671710968,
+      "rewards/check_winston_local_func": 0.42215394973754883,
+      "step": 51
+    },
+    {
+      "clip_ratio": 0.0022109400015324354,
+      "epoch": 0.07272727272727272,
+      "grad_norm": 0.11017795132415774,
+      "kl": 0.013671875,
+      "learning_rate": 1.8181818181818182e-05,
+      "loss": -0.0043,
+      "step": 52
+    },
+    {
+      "clip_ratio": 0.0019256824161857367,
+      "completion_length": 86.5714340209961,
+      "epoch": 0.07412587412587412,
+      "grad_norm": 0.11010712395827207,
+      "kl": 0.0208740234375,
+      "learning_rate": 1.8531468531468532e-05,
+      "loss": -0.0032,
+      "num_tokens": 524863.0,
+      "reward": 0.962996244430542,
+      "reward_std": 0.6607940196990967,
+      "rewards/check_gptzero_func": 0.1071428582072258,
+      "rewards/check_perplexity_diff_func": 0.5,
+      "rewards/check_winston_local_func": 0.355853408575058,
+      "step": 53
+    },
+    {
+      "clip_ratio": 0.0025832760147750378,
+      "epoch": 0.07552447552447553,
+      "grad_norm": 0.10695378216267973,
+      "kl": 0.0250244140625,
+      "learning_rate": 1.888111888111888e-05,
+      "loss": -0.0034,
+      "step": 54
+    },
+    {
+      "clip_ratio": 0.0017639577854424715,
+      "completion_length": 96.08928680419922,
+      "epoch": 0.07692307692307693,
+      "grad_norm": 0.17472941592839772,
+      "kl": 0.0322265625,
+      "learning_rate": 1.923076923076923e-05,
+      "loss": -0.0141,
+      "num_tokens": 544736.0,
+      "reward": 1.3478963375091553,
+      "reward_std": 0.8002303242683411,
+      "rewards/check_gptzero_func": 0.2857142984867096,
+      "rewards/check_perplexity_diff_func": 0.7142857313156128,
+      "rewards/check_winston_local_func": 0.34789615869522095,
+      "step": 55
+    },
+    {
+      "clip_ratio": 0.0034089265391230583,
+      "epoch": 0.07832167832167833,
+      "grad_norm": 0.14666884899599142,
+      "kl": 0.041748046875,
+      "learning_rate": 1.958041958041958e-05,
+      "loss": -0.0155,
+      "step": 56
+    },
+    {
+      "clip_ratio": 0.0026615143287926912,
+      "completion_length": 96.91072082519531,
+      "epoch": 0.07972027972027972,
+      "grad_norm": 0.1400337476406121,
+      "kl": 0.035400390625,
+      "learning_rate": 1.993006993006993e-05,
+      "loss": -0.027,
+      "num_tokens": 564803.0,
+      "reward": 1.0594887733459473,
+      "reward_std": 0.6912091970443726,
+      "rewards/check_gptzero_func": 0.2857142984867096,
+      "rewards/check_perplexity_diff_func": 0.4285714328289032,
+      "rewards/check_winston_local_func": 0.34520307183265686,
+      "step": 57
+    },
+    {
+      "clip_ratio": 0.003153084311634302,
+      "epoch": 0.08111888111888112,
+      "grad_norm": 0.138816359725178,
+      "kl": 0.042236328125,
+      "learning_rate": 2.027972027972028e-05,
+      "loss": -0.0274,
+      "step": 58
+    },
+    {
+      "clip_ratio": 0.0012626759707927704,
+      "completion_length": 76.35714721679688,
+      "epoch": 0.08251748251748252,
+      "grad_norm": 0.15940476459351663,
+      "kl": 0.0556640625,
+      "learning_rate": 2.062937062937063e-05,
+      "loss": -0.0057,
+      "num_tokens": 582265.0,
+      "reward": 0.992591381072998,
+      "reward_std": 0.5938137769699097,
+      "rewards/check_gptzero_func": 0.125,
+      "rewards/check_perplexity_diff_func": 0.4642857015132904,
+      "rewards/check_winston_local_func": 0.4033055901527405,
+      "step": 59
+    },
+    {
+      "clip_ratio": 0.004643784370273352,
+      "epoch": 0.08391608391608392,
+      "grad_norm": 0.15305834024157844,
+      "kl": 0.07421875,
+      "learning_rate": 2.097902097902098e-05,
+      "loss": -0.0064,
+      "step": 60
+    },
+    {
+      "clip_ratio": 0.0032869603019207716,
+      "completion_length": 103.33928680419922,
+      "epoch": 0.08531468531468532,
+      "grad_norm": 0.12260760902895611,
+      "kl": 0.0810546875,
+      "learning_rate": 2.132867132867133e-05,
+      "loss": 0.0066,
+      "num_tokens": 603454.0,
+      "reward": 1.191124439239502,
+      "reward_std": 0.5557684898376465,
+      "rewards/check_gptzero_func": 0.2321428507566452,
+      "rewards/check_perplexity_diff_func": 0.5357142686843872,
+      "rewards/check_winston_local_func": 0.4232672154903412,
+      "step": 61
+    },
+    {
+      "clip_ratio": 0.008058370091021061,
+      "epoch": 0.08671328671328671,
+      "grad_norm": 0.10224499878092017,
+      "kl": 0.1162109375,
+      "learning_rate": 2.1678321678321677e-05,
+      "loss": 0.0058,
+      "step": 62
+    },
+    {
+      "clip_ratio": 0.002202474046498537,
+      "completion_length": 91.4464340209961,
+      "epoch": 0.08811188811188811,
+      "grad_norm": 0.09776882919444682,
+      "kl": 0.2138671875,
+      "learning_rate": 2.202797202797203e-05,
+      "loss": 0.0021,
+      "num_tokens": 622705.0,
+      "reward": 1.1320958137512207,
+      "reward_std": 0.4060821831226349,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 0.3214285671710968,
+      "rewards/check_winston_local_func": 0.41781002283096313,
+      "step": 63
+    },
+    {
+      "clip_ratio": 0.0025903189089149237,
+      "epoch": 0.08951048951048951,
+      "grad_norm": 0.0963149975780195,
+      "kl": 0.2109375,
+      "learning_rate": 2.237762237762238e-05,
+      "loss": 0.0014,
+      "step": 64
+    },
+    {
+      "clip_ratio": 0.0016666523879393935,
+      "completion_length": 100.00000762939453,
+      "epoch": 0.09090909090909091,
+      "grad_norm": 0.12490001498293682,
+      "kl": 0.1220703125,
+      "learning_rate": 2.272727272727273e-05,
+      "loss": 0.0081,
+      "num_tokens": 643191.0,
+      "reward": 1.274580478668213,
+      "reward_std": 0.5277792811393738,
+      "rewards/check_gptzero_func": 0.2321428507566452,
+      "rewards/check_perplexity_diff_func": 0.5357142686843872,
+      "rewards/check_winston_local_func": 0.5067232847213745,
+      "step": 65
+    },
+    {
+      "clip_ratio": 0.0032784033101052046,
+      "epoch": 0.09230769230769231,
+      "grad_norm": 0.12114304348028067,
+      "kl": 0.134765625,
+      "learning_rate": 2.307692307692308e-05,
+      "loss": 0.0074,
+      "step": 66
+    },
+    {
+      "clip_ratio": 0.0017520035617053509,
+      "completion_length": 95.16072082519531,
+      "epoch": 0.0937062937062937,
+      "grad_norm": 0.14292853522837679,
+      "kl": 0.216796875,
+      "learning_rate": 2.342657342657343e-05,
+      "loss": -0.0018,
+      "num_tokens": 662736.0,
+      "reward": 1.1007850170135498,
+      "reward_std": 0.45985397696495056,
+      "rewards/check_gptzero_func": 0.25,
+      "rewards/check_perplexity_diff_func": 0.4642857015132904,
+      "rewards/check_winston_local_func": 0.38649922609329224,
+      "step": 67
+    },
+    {
+      "clip_ratio": 0.00579434959217906,
+      "epoch": 0.0951048951048951,
+      "grad_norm": 0.14340586759573087,
+      "kl": 0.2431640625,
+      "learning_rate": 2.377622377622378e-05,
+      "loss": -0.0029,
+      "step": 68
+    },
+    {
+      "clip_ratio": 0.0022546499967575073,
+      "completion_length": 86.17857360839844,
+      "epoch": 0.0965034965034965,
+      "grad_norm": 0.17287042335415018,
+      "kl": 0.26171875,
+      "learning_rate": 2.4125874125874125e-05,
+      "loss": 0.0059,
+      "num_tokens": 681030.0,
+      "reward": 1.4391711950302124,
+      "reward_std": 0.6516181230545044,
+      "rewards/check_gptzero_func": 0.5178571343421936,
+      "rewards/check_perplexity_diff_func": 0.4642857015132904,
+      "rewards/check_winston_local_func": 0.45702821016311646,
+      "step": 69
+    },
+    {
+      "clip_ratio": 0.004240955226123333,
+      "epoch": 0.0979020979020979,
+      "grad_norm": 0.16371494613893736,
+      "kl": 0.28125,
+      "learning_rate": 2.4475524475524478e-05,
+      "loss": 0.0048,
+      "step": 70
+    },
+    {
+      "clip_ratio": 0.003548440057784319,
+      "completion_length": 102.42857360839844,
+      "epoch": 0.0993006993006993,
+      "grad_norm": 0.1615418523026911,
+      "kl": 0.1708984375,
+      "learning_rate": 2.4825174825174828e-05,
+      "loss": 0.0015,
+      "num_tokens": 701874.0,
+      "reward": 1.4929367303848267,
+      "reward_std": 0.6083499193191528,
+      "rewards/check_gptzero_func": 0.3214285671710968,
+      "rewards/check_perplexity_diff_func": 0.7142857313156128,
+      "rewards/check_winston_local_func": 0.4572224020957947,
+      "step": 71
+    },
+    {
+      "clip_ratio": 0.004331245087087154,
+      "epoch": 0.1006993006993007,
+      "grad_norm": 0.12876017762018707,
+      "kl": 0.216796875,
+      "learning_rate": 2.5174825174825178e-05,
+      "loss": 0.0,
+      "step": 72
+    },
+    {
+      "clip_ratio": 0.003951852675527334,
+      "completion_length": 73.46428680419922,
+      "epoch": 0.1020979020979021,
+      "grad_norm": 0.20791748542634808,
+      "kl": 0.365234375,
+      "learning_rate": 2.5524475524475528e-05,
+      "loss": -0.0037,
+      "num_tokens": 718488.0,
+      "reward": 1.7953797578811646,
+      "reward_std": 0.5393027067184448,
+      "rewards/check_gptzero_func": 0.4464285671710968,
+      "rewards/check_perplexity_diff_func": 0.8571428656578064,
+      "rewards/check_winston_local_func": 0.49180838465690613,
+      "step": 73
+    },
+    {
+      "clip_ratio": 0.008312534540891647,
+      "epoch": 0.1034965034965035,
+      "grad_norm": 0.19232465302245624,
+      "kl": 0.5390625,
+      "learning_rate": 2.5874125874125877e-05,
+      "loss": -0.0053,
+      "step": 74
+    },
+    {
+      "clip_ratio": 0.0021736263297498226,
+      "completion_length": 97.73214721679688,
+      "epoch": 0.1048951048951049,
+      "grad_norm": 0.18504738647098173,
+      "kl": 0.2421875,
+      "learning_rate": 2.6223776223776224e-05,
+      "loss": -0.0023,
+      "num_tokens": 738619.0,
+      "reward": 1.284185767173767,
+      "reward_std": 0.6959513425827026,
+      "rewards/check_gptzero_func": 0.4464285671710968,
+      "rewards/check_perplexity_diff_func": 0.3928571343421936,
+      "rewards/check_winston_local_func": 0.4449000358581543,
+      "step": 75
+    },
+    {
+      "clip_ratio": 0.0033756059128791094,
+      "epoch": 0.1062937062937063,
+      "grad_norm": 0.17585639043073228,
+      "kl": 0.255859375,
+      "learning_rate": 2.6573426573426574e-05,
+      "loss": -0.0043,
+      "step": 76
+    },
+    {
+      "clip_ratio": 0.003740633837878704,
+      "completion_length": 95.67857360839844,
+      "epoch": 0.1076923076923077,
+      "grad_norm": 0.21573775826044048,
+      "kl": 0.2451171875,
+      "learning_rate": 2.6923076923076923e-05,
+      "loss": 0.0105,
+      "num_tokens": 758111.0,
+      "reward": 1.4672702550888062,
+      "reward_std": 0.85318922996521,
+      "rewards/check_gptzero_func": 0.25,
+      "rewards/check_perplexity_diff_func": 0.7857142686843872,
+      "rewards/check_winston_local_func": 0.43155592679977417,
+      "step": 77
+    },
+    {
+      "clip_ratio": 0.004517109598964453,
+      "epoch": 0.10909090909090909,
+      "grad_norm": 0.2071235680791843,
+      "kl": 0.302734375,
+      "learning_rate": 2.7272727272727273e-05,
+      "loss": 0.009,
+      "step": 78
+    },
+    {
+      "clip_ratio": 0.004641580395400524,
+      "completion_length": 117.75000762939453,
+      "epoch": 0.11048951048951049,
+      "grad_norm": 0.19278645376555428,
+      "kl": 0.259765625,
+      "learning_rate": 2.762237762237762e-05,
+      "loss": -0.0004,
+      "num_tokens": 780403.0,
+      "reward": 1.4974778890609741,
+      "reward_std": 0.59557044506073,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 0.6785714030265808,
+      "rewards/check_winston_local_func": 0.39033493399620056,
+      "step": 79
+    },
+    {
+      "clip_ratio": 0.006305322516709566,
+      "epoch": 0.11188811188811189,
+      "grad_norm": 0.13786850726930885,
+      "kl": 0.3125,
+      "learning_rate": 2.7972027972027976e-05,
+      "loss": -0.0019,
+      "step": 80
+    },
+    {
+      "clip_ratio": 0.0017778041074052453,
+      "completion_length": 101.03572082519531,
+      "epoch": 0.11328671328671329,
+      "grad_norm": 0.18099911272050398,
+      "kl": 1.7578125,
+      "learning_rate": 2.8321678321678326e-05,
+      "loss": 0.0053,
+      "num_tokens": 800231.0,
+      "reward": 1.420699954032898,
+      "reward_std": 0.5104399919509888,
+      "rewards/check_gptzero_func": 0.2678571343421936,
+      "rewards/check_perplexity_diff_func": 0.6428571343421936,
+      "rewards/check_winston_local_func": 0.5099858045578003,
+      "step": 81
+    },
+    {
+      "clip_ratio": 0.0030425102449953556,
+      "epoch": 0.11468531468531469,
+      "grad_norm": 0.10725025403922342,
+      "kl": 0.9921875,
+      "learning_rate": 2.8671328671328672e-05,
+      "loss": 0.0043,
+      "step": 82
+    },
+    {
+      "clip_ratio": 0.0030319676734507084,
+      "completion_length": 93.3214340209961,
+      "epoch": 0.11608391608391608,
+      "grad_norm": 0.22858957476463732,
+      "kl": 0.35546875,
+      "learning_rate": 2.9020979020979022e-05,
+      "loss": 0.0021,
+      "num_tokens": 819773.0,
+      "reward": 1.4201573133468628,
+      "reward_std": 0.761705756187439,
+      "rewards/check_gptzero_func": 0.3392857015132904,
+      "rewards/check_perplexity_diff_func": 0.5714285969734192,
+      "rewards/check_winston_local_func": 0.5094431042671204,
+      "step": 83
+    },
+    {
+      "clip_ratio": 0.0027519434224814177,
+      "epoch": 0.11748251748251748,
+      "grad_norm": 0.180287994009811,
+      "kl": 0.392578125,
+      "learning_rate": 2.9370629370629372e-05,
+      "loss": -0.0002,
+      "step": 84
+    },
+    {
+      "clip_ratio": 0.0034523813519626856,
+      "completion_length": 102.66072082519531,
+      "epoch": 0.11888111888111888,
+      "grad_norm": 0.25571068617285075,
+      "kl": 0.494140625,
+      "learning_rate": 2.972027972027972e-05,
+      "loss": -0.0103,
+      "num_tokens": 840280.0,
+      "reward": 1.7226576805114746,
+      "reward_std": 0.7545516490936279,
+      "rewards/check_gptzero_func": 0.5714285969734192,
+      "rewards/check_perplexity_diff_func": 0.6785714030265808,
+      "rewards/check_winston_local_func": 0.4726576805114746,
+      "step": 85
+    },
+    {
+      "clip_ratio": 0.00493992306292057,
+      "epoch": 0.12027972027972028,
+      "grad_norm": 0.2172932526081326,
+      "kl": 0.490234375,
+      "learning_rate": 3.0069930069930068e-05,
+      "loss": -0.013,
+      "step": 86
+    },
+    {
+      "clip_ratio": 0.003154533449560404,
+      "completion_length": 98.96428680419922,
+      "epoch": 0.12167832167832168,
+      "grad_norm": 0.3651883135439141,
+      "kl": 0.5859375,
+      "learning_rate": 3.0419580419580425e-05,
+      "loss": -0.0129,
+      "num_tokens": 860422.0,
+      "reward": 1.6449967622756958,
+      "reward_std": 0.7163441777229309,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 0.7142857313156128,
+      "rewards/check_winston_local_func": 0.5378537178039551,
+      "step": 87
+    },
+    {
+      "clip_ratio": 0.007247431669384241,
+      "epoch": 0.12307692307692308,
+      "grad_norm": 0.2167770735292921,
+      "kl": 0.61328125,
+      "learning_rate": 3.0769230769230774e-05,
+      "loss": -0.0168,
+      "step": 88
+    },
+    {
+      "clip_ratio": 0.0027036736719310284,
+      "completion_length": 108.03572082519531,
+      "epoch": 0.12447552447552447,
+      "grad_norm": 0.12604084880729077,
+      "kl": 0.291015625,
+      "learning_rate": 3.111888111888112e-05,
+      "loss": -0.0039,
+      "num_tokens": 881364.0,
+      "reward": 1.3799673318862915,
+      "reward_std": 0.6881024837493896,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 0.5714285969734192,
+      "rewards/check_winston_local_func": 0.4156815707683563,
+      "step": 89
+    },
+    {
+      "clip_ratio": 0.005238416139036417,
+      "epoch": 0.1258741258741259,
+      "grad_norm": 0.12103303780018655,
+      "kl": 0.30859375,
+      "learning_rate": 3.146853146853147e-05,
+      "loss": -0.005,
+      "step": 90
+    },
+    {
+      "clip_ratio": 0.003768512513488531,
+      "completion_length": 122.80357360839844,
+      "epoch": 0.12727272727272726,
+      "grad_norm": 0.18423815092289067,
+      "kl": 0.67578125,
+      "learning_rate": 3.181818181818182e-05,
+      "loss": -0.0171,
+      "num_tokens": 904487.0,
+      "reward": 1.6101170778274536,
+      "reward_std": 0.6241902709007263,
+      "rewards/check_gptzero_func": 0.3392857015132904,
+      "rewards/check_perplexity_diff_func": 0.7857142686843872,
+      "rewards/check_winston_local_func": 0.48511695861816406,
+      "step": 91
+    },
+    {
+      "clip_ratio": 0.005407070741057396,
+      "epoch": 0.12867132867132866,
+      "grad_norm": 0.15457101650134836,
+      "kl": 0.66796875,
+      "learning_rate": 3.216783216783217e-05,
+      "loss": -0.0182,
+      "step": 92
+    },
+    {
+      "clip_ratio": 0.002042532665655017,
+      "completion_length": 108.17857360839844,
+      "epoch": 0.13006993006993006,
+      "grad_norm": 0.19588538280583437,
+      "kl": 1.75,
+      "learning_rate": 3.251748251748252e-05,
+      "loss": -0.0023,
+      "num_tokens": 925633.0,
+      "reward": 1.3932582139968872,
+      "reward_std": 0.5087055563926697,
+      "rewards/check_gptzero_func": 0.4107142984867096,
+      "rewards/check_perplexity_diff_func": 0.4642857015132904,
+      "rewards/check_winston_local_func": 0.5182580351829529,
+      "step": 93
+    },
+    {
+      "clip_ratio": 0.00876330491155386,
+      "epoch": 0.13146853146853146,
+      "grad_norm": 0.1506256243597478,
+      "kl": 0.80078125,
+      "learning_rate": 3.2867132867132866e-05,
+      "loss": -0.0035,
+      "step": 94
+    },
+    {
+      "clip_ratio": 0.003118924330919981,
+      "completion_length": 77.4464340209961,
+      "epoch": 0.13286713286713286,
+      "grad_norm": 0.34784170699852607,
+      "kl": 0.72265625,
+      "learning_rate": 3.321678321678322e-05,
+      "loss": -0.0134,
+      "num_tokens": 942892.0,
+      "reward": 1.832355260848999,
+      "reward_std": 1.064102292060852,
+      "rewards/check_gptzero_func": 0.4821428656578064,
+      "rewards/check_perplexity_diff_func": 0.8214285969734192,
+      "rewards/check_winston_local_func": 0.5287837982177734,
+      "step": 95
+    },
+    {
+      "clip_ratio": 0.00509096821770072,
+      "epoch": 0.13426573426573427,
+      "grad_norm": 0.25968424971577175,
+      "kl": 0.96875,
+      "learning_rate": 3.356643356643357e-05,
+      "loss": -0.0176,
+      "step": 96
+    },
+    {
+      "clip_ratio": 0.0037272910121828318,
+      "completion_length": 95.26786041259766,
+      "epoch": 0.13566433566433567,
+      "grad_norm": 0.16914051199341906,
+      "kl": 0.349609375,
+      "learning_rate": 3.391608391608392e-05,
+      "loss": 0.0037,
+      "num_tokens": 962317.0,
+      "reward": 1.805726170539856,
+      "reward_std": 0.831261396408081,
+      "rewards/check_gptzero_func": 0.4107142984867096,
+      "rewards/check_perplexity_diff_func": 0.9285714030265808,
+      "rewards/check_winston_local_func": 0.46644020080566406,
+      "step": 97
+    },
+    {
+      "clip_ratio": 0.0034601751249283552,
+      "epoch": 0.13706293706293707,
+      "grad_norm": 0.16538870434198547,
+      "kl": 0.388671875,
+      "learning_rate": 3.4265734265734265e-05,
+      "loss": 0.0019,
+      "step": 98
+    },
+    {
+      "clip_ratio": 0.004487877711653709,
+      "completion_length": 111.8214340209961,
+      "epoch": 0.13846153846153847,
+      "grad_norm": 0.12415078751808917,
+      "kl": 0.384765625,
+      "learning_rate": 3.461538461538462e-05,
+      "loss": -0.001,
+      "num_tokens": 983513.0,
+      "reward": 1.2062333822250366,
+      "reward_std": 0.591974139213562,
+      "rewards/check_gptzero_func": 0.3214285671710968,
+      "rewards/check_perplexity_diff_func": 0.3571428656578064,
+      "rewards/check_winston_local_func": 0.5276618599891663,
+      "step": 99
+    },
+    {
+      "clip_ratio": 0.00284260674379766,
+      "epoch": 0.13986013986013987,
+      "grad_norm": 0.1197068572237121,
+      "kl": 0.376953125,
+      "learning_rate": 3.4965034965034965e-05,
+      "loss": -0.0017,
+      "step": 100
+    },
+    {
+      "clip_ratio": 0.0023206709884107113,
+      "completion_length": 101.08928680419922,
+      "epoch": 0.14125874125874127,
+      "grad_norm": 0.16400691003183243,
+      "kl": 0.30078125,
+      "learning_rate": 3.531468531468531e-05,
+      "loss": 0.0033,
+      "num_tokens": 1003878.0,
+      "reward": 1.631854772567749,
+      "reward_std": 0.5223387479782104,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 0.7142857313156128,
+      "rewards/check_winston_local_func": 0.5247119665145874,
+      "step": 101
+    },
+    {
+      "clip_ratio": 0.0016618981026113033,
+      "epoch": 0.14265734265734265,
+      "grad_norm": 0.1573929247787687,
+      "kl": 0.326171875,
+      "learning_rate": 3.566433566433567e-05,
+      "loss": 0.0012,
+      "step": 102
+    },
+    {
+      "clip_ratio": 0.0022201864048838615,
+      "completion_length": 124.46429443359375,
+      "epoch": 0.14405594405594405,
+      "grad_norm": 0.1353861256020412,
+      "kl": 0.3828125,
+      "learning_rate": 3.601398601398602e-05,
+      "loss": 0.005,
+      "num_tokens": 1027066.0,
+      "reward": 1.6920486688613892,
+      "reward_std": 0.48560601472854614,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 0.75,
+      "rewards/check_winston_local_func": 0.5491914749145508,
+      "step": 103
+    },
+    {
+      "clip_ratio": 0.0029075751081109047,
+      "epoch": 0.14545454545454545,
+      "grad_norm": 0.13328591024866146,
+      "kl": 0.39453125,
+      "learning_rate": 3.6363636363636364e-05,
+      "loss": 0.0034,
+      "step": 104
+    },
+    {
+      "clip_ratio": 0.0019614642951637506,
+      "completion_length": 85.26786041259766,
+      "epoch": 0.14685314685314685,
+      "grad_norm": 0.19674478002774554,
+      "kl": 0.333984375,
+      "learning_rate": 3.671328671328672e-05,
+      "loss": -0.0188,
+      "num_tokens": 1045567.0,
+      "reward": 1.6822034120559692,
+      "reward_std": 0.7559517621994019,
+      "rewards/check_gptzero_func": 0.5357142686843872,
+      "rewards/check_perplexity_diff_func": 0.6071428656578064,
+      "rewards/check_winston_local_func": 0.5393460988998413,
+      "step": 105
+    },
+    {
+      "clip_ratio": 0.004534203093498945,
+      "epoch": 0.14825174825174825,
+      "grad_norm": 0.16787611299305724,
+      "kl": 0.3671875,
+      "learning_rate": 3.7062937062937064e-05,
+      "loss": -0.0217,
+      "step": 106
+    },
+    {
+      "clip_ratio": 0.003009920008480549,
+      "completion_length": 106.55357360839844,
+      "epoch": 0.14965034965034965,
+      "grad_norm": 0.19751132540015684,
+      "kl": 0.455078125,
+      "learning_rate": 3.741258741258741e-05,
+      "loss": 0.0111,
+      "num_tokens": 1066466.0,
+      "reward": 2.178619861602783,
+      "reward_std": 0.7016831040382385,
+      "rewards/check_gptzero_func": 0.6071428656578064,
+      "rewards/check_perplexity_diff_func": 0.8928571343421936,
+      "rewards/check_winston_local_func": 0.6786197423934937,
+      "step": 107
+    },
+    {
+      "clip_ratio": 0.004473494831472635,
+      "epoch": 0.15104895104895105,
+      "grad_norm": 0.1431470008572649,
+      "kl": 0.4375,
+      "learning_rate": 3.776223776223776e-05,
+      "loss": 0.0095,
+      "step": 108
+    },
+    {
+      "clip_ratio": 0.0026541000697761774,
+      "completion_length": 100.6964340209961,
+      "epoch": 0.15244755244755245,
+      "grad_norm": 0.33135604940901486,
+      "kl": 0.310546875,
+      "learning_rate": 3.811188811188811e-05,
+      "loss": 0.0138,
+      "num_tokens": 1086751.0,
+      "reward": 1.6315226554870605,
+      "reward_std": 0.7806248068809509,
+      "rewards/check_gptzero_func": 0.4642857015132904,
+      "rewards/check_perplexity_diff_func": 0.6071428656578064,
+      "rewards/check_winston_local_func": 0.5600939393043518,
+      "step": 109
+    },
+    {
+      "clip_ratio": 0.0028892713598906994,
+      "epoch": 0.15384615384615385,
+      "grad_norm": 0.24217207616877234,
+      "kl": 0.361328125,
+      "learning_rate": 3.846153846153846e-05,
+      "loss": 0.0074,
+      "step": 110
+    },
+    {
+      "clip_ratio": 0.0021747422870248556,
+      "completion_length": 84.76786041259766,
+      "epoch": 0.15524475524475526,
+      "grad_norm": 0.15803622964815373,
+      "kl": 0.66796875,
+      "learning_rate": 3.8811188811188816e-05,
+      "loss": -0.0045,
+      "num_tokens": 1104518.0,
+      "reward": 1.6692500114440918,
+      "reward_std": 0.5965060591697693,
+      "rewards/check_gptzero_func": 0.5535714030265808,
+      "rewards/check_perplexity_diff_func": 0.6071428656578064,
+      "rewards/check_winston_local_func": 0.5085356831550598,
+      "step": 111
+    },
+    {
+      "clip_ratio": 0.004626331850886345,
+      "epoch": 0.15664335664335666,
+      "grad_norm": 0.14920406573741435,
+      "kl": 0.7578125,
+      "learning_rate": 3.916083916083916e-05,
+      "loss": -0.0064,
+      "step": 112
+    },
+    {
+      "clip_ratio": 0.002694516209885478,
+      "completion_length": 92.17857360839844,
+      "epoch": 0.15804195804195803,
+      "grad_norm": 0.161427063551978,
+      "kl": 0.361328125,
+      "learning_rate": 3.9510489510489516e-05,
+      "loss": 0.0058,
+      "num_tokens": 1123466.0,
+      "reward": 1.353637456893921,
+      "reward_std": 0.5352396965026855,
+      "rewards/check_gptzero_func": 0.4821428656578064,
+      "rewards/check_perplexity_diff_func": 0.3571428656578064,
+      "rewards/check_winston_local_func": 0.5143517255783081,
+      "step": 113
+    },
+    {
+      "clip_ratio": 0.00391918933019042,
+      "epoch": 0.15944055944055943,
+      "grad_norm": 0.14561571783883442,
+      "kl": 0.39453125,
+      "learning_rate": 3.986013986013986e-05,
+      "loss": 0.0036,
+      "step": 114
+    },
+    {
+      "clip_ratio": 0.001278667594306171,
+      "completion_length": 107.12500762939453,
+      "epoch": 0.16083916083916083,
+      "grad_norm": 0.1805624176716803,
+      "kl": 0.3046875,
+      "learning_rate": 4.020979020979021e-05,
+      "loss": -0.001,
+      "num_tokens": 1144619.0,
+      "reward": 1.6282455921173096,
+      "reward_std": 0.7325619459152222,
+      "rewards/check_gptzero_func": 0.4464285671710968,
+      "rewards/check_perplexity_diff_func": 0.5357142686843872,
+      "rewards/check_winston_local_func": 0.6461027264595032,
+      "step": 115
+    },
+    {
+      "clip_ratio": 0.0017050534952431917,
+      "epoch": 0.16223776223776223,
+      "grad_norm": 0.16942268791902212,
+      "kl": 0.337890625,
+      "learning_rate": 4.055944055944056e-05,
+      "loss": -0.0027,
+      "step": 116
+    },
+    {
+      "clip_ratio": 0.0015462420415133238,
+      "completion_length": 114.8214340209961,
+      "epoch": 0.16363636363636364,
+      "grad_norm": 0.1777133538888581,
+      "kl": 0.322265625,
+      "learning_rate": 4.0909090909090915e-05,
+      "loss": -0.0028,
+      "num_tokens": 1166185.0,
+      "reward": 1.566156268119812,
+      "reward_std": 0.38663557171821594,
+      "rewards/check_gptzero_func": 0.4642857015132904,
+      "rewards/check_perplexity_diff_func": 0.5,
+      "rewards/check_winston_local_func": 0.6018705368041992,
+      "step": 117
+    },
+    {
+      "clip_ratio": 0.0030793240293860435,
+      "epoch": 0.16503496503496504,
+      "grad_norm": 0.15084856022622706,
+      "kl": 0.2890625,
+      "learning_rate": 4.125874125874126e-05,
+      "loss": -0.0048,
+      "step": 118
+    },
+    {
+      "clip_ratio": 0.0018360918620601296,
+      "completion_length": 100.25000762939453,
+      "epoch": 0.16643356643356644,
+      "grad_norm": 0.10943003413749244,
+      "kl": 0.83984375,
+      "learning_rate": 4.1608391608391614e-05,
+      "loss": 0.0043,
+      "num_tokens": 1186317.0,
+      "reward": 1.7977957725524902,
+      "reward_std": 0.5172301530838013,
+      "rewards/check_gptzero_func": 0.5714285969734192,
+      "rewards/check_perplexity_diff_func": 0.7142857313156128,
+      "rewards/check_winston_local_func": 0.5120813250541687,
+      "step": 119
+    },
+    {
+      "clip_ratio": 0.0025589358992874622,
+      "epoch": 0.16783216783216784,
+      "grad_norm": 0.10431032316313237,
+      "kl": 0.68359375,
+      "learning_rate": 4.195804195804196e-05,
+      "loss": 0.0029,
+      "step": 120
+    },
+    {
+      "clip_ratio": 0.002800893737003207,
+      "completion_length": 120.91072082519531,
+      "epoch": 0.16923076923076924,
+      "grad_norm": 0.1567960287886374,
+      "kl": 0.37890625,
+      "learning_rate": 4.230769230769231e-05,
+      "loss": -0.0016,
+      "num_tokens": 1208530.0,
+      "reward": 1.5369055271148682,
+      "reward_std": 0.7073518633842468,
+      "rewards/check_gptzero_func": 0.3214285671710968,
+      "rewards/check_perplexity_diff_func": 0.6785714030265808,
+      "rewards/check_winston_local_func": 0.5369054079055786,
+      "step": 121
+    },
+    {
+      "clip_ratio": 0.0026409339625388384,
+      "epoch": 0.17062937062937064,
+      "grad_norm": 0.15013150065673506,
+      "kl": 0.37890625,
+      "learning_rate": 4.265734265734266e-05,
+      "loss": -0.0035,
+      "step": 122
+    },
+    {
+      "clip_ratio": 0.002211854327470064,
+      "completion_length": 110.66072082519531,
+      "epoch": 0.17202797202797201,
+      "grad_norm": 0.17274409069851862,
+      "kl": 0.380859375,
+      "learning_rate": 4.300699300699301e-05,
+      "loss": -0.0062,
+      "num_tokens": 1229171.0,
+      "reward": 1.6140996217727661,
+      "reward_std": 0.7221139669418335,
+      "rewards/check_gptzero_func": 0.4642857015132904,
+      "rewards/check_perplexity_diff_func": 0.5357142686843872,
+      "rewards/check_winston_local_func": 0.6140995025634766,
+      "step": 123
+    },
+    {
+      "clip_ratio": 0.005159804597496986,
+      "epoch": 0.17342657342657342,
+      "grad_norm": 0.13306053466541726,
+      "kl": 0.443359375,
+      "learning_rate": 4.335664335664335e-05,
+      "loss": -0.009,
+      "step": 124
+    },
+    {
+      "clip_ratio": 0.0031610180158168077,
+      "completion_length": 104.30357360839844,
+      "epoch": 0.17482517482517482,
+      "grad_norm": 0.19261275892706695,
+      "kl": 0.30078125,
+      "learning_rate": 4.370629370629371e-05,
+      "loss": -0.0129,
+      "num_tokens": 1249350.0,
+      "reward": 1.9047484397888184,
+      "reward_std": 0.6767317652702332,
+      "rewards/check_gptzero_func": 0.6071428656578064,
+      "rewards/check_perplexity_diff_func": 0.6071428656578064,
+      "rewards/check_winston_local_func": 0.6904626488685608,
+      "step": 125
+    },
+    {
+      "clip_ratio": 0.004231306724250317,
+      "epoch": 0.17622377622377622,
+      "grad_norm": 0.1752737652506695,
+      "kl": 0.33203125,
+      "learning_rate": 4.405594405594406e-05,
+      "loss": -0.0162,
+      "step": 126
+    },
+    {
+      "clip_ratio": 0.003981932066380978,
+      "completion_length": 109.53572082519531,
+      "epoch": 0.17762237762237762,
+      "grad_norm": 0.1423250496935692,
+      "kl": 0.48828125,
+      "learning_rate": 4.4405594405594406e-05,
+      "loss": -0.0102,
+      "num_tokens": 1269848.0,
+      "reward": 1.7591207027435303,
+      "reward_std": 0.5321380496025085,
+      "rewards/check_gptzero_func": 0.5,
+      "rewards/check_perplexity_diff_func": 0.6785714030265808,
+      "rewards/check_winston_local_func": 0.5805493593215942,
+      "step": 127
+    },
+    {
+      "clip_ratio": 0.005287667270749807,
+      "epoch": 0.17902097902097902,
+      "grad_norm": 0.13255848380623775,
+      "kl": 0.498046875,
+      "learning_rate": 4.475524475524476e-05,
+      "loss": -0.0121,
+      "step": 128
+    },
+    {
+      "clip_ratio": 0.0025025398936122656,
+      "completion_length": 105.30357360839844,
+      "epoch": 0.18041958041958042,
+      "grad_norm": 0.1177341259986552,
+      "kl": 0.283203125,
+      "learning_rate": 4.5104895104895105e-05,
+      "loss": -0.0016,
+      "num_tokens": 1290033.0,
+      "reward": 1.7174798250198364,
+      "reward_std": 0.5096268057823181,
+      "rewards/check_gptzero_func": 0.5357142686843872,
+      "rewards/check_perplexity_diff_func": 0.5357142686843872,
+      "rewards/check_winston_local_func": 0.6460510492324829,
+      "step": 129
+    },
+    {
+      "clip_ratio": 0.0033291254658252,
+      "epoch": 0.18181818181818182,
+      "grad_norm": 0.11141937591016414,
+      "kl": 0.294921875,
+      "learning_rate": 4.545454545454546e-05,
+      "loss": -0.0032,
+      "step": 130
+    },
+    {
+      "clip_ratio": 0.0024425899609923363,
+      "completion_length": 96.30357360839844,
+      "epoch": 0.18321678321678322,
+      "grad_norm": 0.23593816935965856,
+      "kl": 0.341796875,
+      "learning_rate": 4.5804195804195805e-05,
+      "loss": -0.006,
+      "num_tokens": 1309356.0,
+      "reward": 1.7568891048431396,
+      "reward_std": 0.7255779504776001,
+      "rewards/check_gptzero_func": 0.5892857313156128,
+      "rewards/check_perplexity_diff_func": 0.5357142686843872,
+      "rewards/check_winston_local_func": 0.6318890452384949,
+      "step": 131
+    },
+    {
+      "clip_ratio": 0.004331678152084351,
+      "epoch": 0.18461538461538463,
+      "grad_norm": 0.2135627059179365,
+      "kl": 0.349609375,
+      "learning_rate": 4.615384615384616e-05,
+      "loss": -0.0089,
+      "step": 132
+    },
+    {
+      "clip_ratio": 0.0028827630449086428,
+      "completion_length": 111.6964340209961,
+      "epoch": 0.18601398601398603,
+      "grad_norm": 0.1622976058137894,
+      "kl": 0.369140625,
+      "learning_rate": 4.6503496503496505e-05,
+      "loss": 0.014,
+      "num_tokens": 1330017.0,
+      "reward": 1.6692737340927124,
+      "reward_std": 0.628279983997345,
+      "rewards/check_gptzero_func": 0.4464285671710968,
+      "rewards/check_perplexity_diff_func": 0.6428571343421936,
+      "rewards/check_winston_local_func": 0.5799878835678101,
+      "step": 133
+    },
+    {
+      "clip_ratio": 0.004608546383678913,
+      "epoch": 0.1874125874125874,
+      "grad_norm": 0.16503328062657166,
+      "kl": 0.4140625,
+      "learning_rate": 4.685314685314686e-05,
+      "loss": 0.0121,
+      "step": 134
+    },
+    {
+      "clip_ratio": 0.0027426625601947308,
+      "completion_length": 107.58928680419922,
+      "epoch": 0.1888111888111888,
+      "grad_norm": 0.27926954484170907,
+      "kl": 0.5859375,
+      "learning_rate": 4.7202797202797204e-05,
+      "loss": 0.0139,
+      "num_tokens": 1350164.0,
+      "reward": 1.9428951740264893,
+      "reward_std": 0.6565932035446167,
+      "rewards/check_gptzero_func": 0.6785714030265808,
+      "rewards/check_perplexity_diff_func": 0.5714285969734192,
+      "rewards/check_winston_local_func": 0.6928950548171997,
+      "step": 135
+    },
+    {
+      "clip_ratio": 0.00439854059368372,
+      "epoch": 0.1902097902097902,
+      "grad_norm": 0.1925514269864402,
+      "kl": 0.5546875,
+      "learning_rate": 4.755244755244756e-05,
+      "loss": 0.0074,
+      "step": 136
+    },
+    {
+      "clip_ratio": 0.0018738384824246168,
+      "completion_length": 116.4464340209961,
+      "epoch": 0.1916083916083916,
+      "grad_norm": 0.14918417137816656,
+      "kl": 0.58984375,
+      "learning_rate": 4.7902097902097904e-05,
+      "loss": 0.0037,
+      "num_tokens": 1371507.0,
+      "reward": 2.0137577056884766,
+      "reward_std": 0.5378711819648743,
+      "rewards/check_gptzero_func": 0.6071428656578064,
+      "rewards/check_perplexity_diff_func": 0.6785714030265808,
+      "rewards/check_winston_local_func": 0.7280434370040894,
+      "step": 137
+    },
+    {
+      "clip_ratio": 0.003529219189658761,
+      "epoch": 0.193006993006993,
+      "grad_norm": 0.1301060608041501,
+      "kl": 0.6171875,
+      "learning_rate": 4.825174825174825e-05,
+      "loss": 0.0016,
+      "step": 138
+    },
+    {
+      "clip_ratio": 0.0011985624441877007,
+      "completion_length": 112.6964340209961,
+      "epoch": 0.1944055944055944,
+      "grad_norm": 0.13034934146859325,
+      "kl": 0.322265625,
+      "learning_rate": 4.86013986013986e-05,
+      "loss": -0.0024,
+      "num_tokens": 1392414.0,
+      "reward": 1.9355616569519043,
+      "reward_std": 0.4726971983909607,
+      "rewards/check_gptzero_func": 0.5892857313156128,
+      "rewards/check_perplexity_diff_func": 0.75,
+      "rewards/check_winston_local_func": 0.596275806427002,
+      "step": 139
+    },
+    {
+      "clip_ratio": 0.0032699662260711193,
+      "epoch": 0.1958041958041958,
+      "grad_norm": 0.12091032490165694,
+      "kl": 0.318359375,
+      "learning_rate": 4.8951048951048956e-05,
+      "loss": -0.0039,
+      "step": 140
+    },
+    {
+      "clip_ratio": 0.0013384540798142552,
+      "completion_length": 132.05357360839844,
+      "epoch": 0.1972027972027972,
+      "grad_norm": 0.12728573210952063,
+      "kl": 0.59765625,
+      "learning_rate": 4.93006993006993e-05,
+      "loss": 0.0045,
+      "num_tokens": 1416177.0,
+      "reward": 1.7924094200134277,
+      "reward_std": 0.7202263474464417,
+      "rewards/check_gptzero_func": 0.5714285969734192,
+      "rewards/check_perplexity_diff_func": 0.7142857313156128,
+      "rewards/check_winston_local_func": 0.5066950917243958,
+      "step": 141
+    },
+    {
+      "clip_ratio": 0.0017050639726221561,
+      "epoch": 0.1986013986013986,
+      "grad_norm": 0.12231361158637834,
+      "kl": 0.65234375,
+      "learning_rate": 4.9650349650349656e-05,
+      "loss": 0.0025,
+      "step": 142
+    },
+    {
+      "clip_ratio": 0.0015860958956182003,
+      "completion_length": 99.73214721679688,
+      "epoch": 0.2,
+      "grad_norm": 0.1853946177277763,
+      "kl": 0.58984375,
+      "learning_rate": 5e-05,
+      "loss": 0.0021,
+      "num_tokens": 1435732.0,
+      "reward": 2.0557591915130615,
+      "reward_std": 0.503829836845398,
+      "rewards/check_gptzero_func": 0.6964285969734192,
+      "rewards/check_perplexity_diff_func": 0.75,
+      "rewards/check_winston_local_func": 0.6093305945396423,
+      "step": 143
+    },
+    {
+      "clip_ratio": 0.002123563550412655,
+      "epoch": 0.2013986013986014,
+      "grad_norm": 0.13470126319919157,
+      "kl": 0.423828125,
+      "learning_rate": 5.0349650349650356e-05,
+      "loss": -0.0009,
+      "step": 144
+    },
+    {
+      "clip_ratio": 0.0018757216166704893,
+      "completion_length": 95.73214721679688,
+      "epoch": 0.20279720279720279,
+      "grad_norm": 0.14930190515465938,
+      "kl": 0.388671875,
+      "learning_rate": 5.06993006993007e-05,
+      "loss": -0.0026,
+      "num_tokens": 1454705.0,
+      "reward": 2.0560998916625977,
+      "reward_std": 0.5541732311248779,
+      "rewards/check_gptzero_func": 0.6964285969734192,
+      "rewards/check_perplexity_diff_func": 0.5714285969734192,
+      "rewards/check_winston_local_func": 0.788242518901825,
+      "step": 145
+    },
+    {
+      "clip_ratio": 0.003540371311828494,
+      "epoch": 0.2041958041958042,
+      "grad_norm": 0.1304251509197979,
+      "kl": 0.41796875,
+      "learning_rate": 5.1048951048951055e-05,
+      "loss": -0.0046,
+      "step": 146
+    },
+    {
+      "clip_ratio": 0.0031265008728951216,
+      "completion_length": 80.9464340209961,
+      "epoch": 0.2055944055944056,
+      "grad_norm": 0.2087929657840581,
+      "kl": 1.1328125,
+      "learning_rate": 5.1398601398601395e-05,
+      "loss": 0.0026,
+      "num_tokens": 1472182.0,
+      "reward": 1.9076076745986938,
+      "reward_std": 0.46633273363113403,
+      "rewards/check_gptzero_func": 0.6428571343421936,
+      "rewards/check_perplexity_diff_func": 0.5,
+      "rewards/check_winston_local_func": 0.7647504210472107,
+      "step": 147
+    },
+    {
+      "clip_ratio": 0.0029583375435322523,
+      "epoch": 0.206993006993007,
+      "grad_norm": 0.178151823438605,
+      "kl": 0.66015625,
+      "learning_rate": 5.1748251748251755e-05,
+      "loss": -0.0014,
+      "step": 148
+    },
+    {
+      "clip_ratio": 0.0020277961157262325,
+      "completion_length": 126.58929443359375,
+      "epoch": 0.2083916083916084,
+      "grad_norm": 0.13736153026044381,
+      "kl": 0.373046875,
+      "learning_rate": 5.2097902097902094e-05,
+      "loss": 0.0127,
+      "num_tokens": 1494581.0,
+      "reward": 1.870171308517456,
+      "reward_std": 0.4412032663822174,
+      "rewards/check_gptzero_func": 0.4464285671710968,
+      "rewards/check_perplexity_diff_func": 0.8214285969734192,
+      "rewards/check_winston_local_func": 0.6023141741752625,
+      "step": 149
+    },
+    {
+      "clip_ratio": 0.0022643795236945152,
+      "epoch": 0.2097902097902098,
+      "grad_norm": 0.12942677801199462,
+      "kl": 0.40625,
+      "learning_rate": 5.244755244755245e-05,
+      "loss": 0.0106,
+      "step": 150
+    },
+    {
+      "clip_ratio": 0.001886948710307479,
+      "completion_length": 95.14286041259766,
+      "epoch": 0.2111888111888112,
+      "grad_norm": 0.15555312890089615,
+      "kl": 0.62890625,
+      "learning_rate": 5.279720279720281e-05,
+      "loss": 0.0013,
+      "num_tokens": 1513403.0,
+      "reward": 1.9735006093978882,
+      "reward_std": 0.5218394994735718,
+      "rewards/check_gptzero_func": 0.5535714030265808,
+      "rewards/check_perplexity_diff_func": 0.6071428656578064,
+      "rewards/check_winston_local_func": 0.8127861022949219,
+      "step": 151
+    },
+    {
+      "clip_ratio": 0.003476199461147189,
+      "epoch": 0.2125874125874126,
+      "grad_norm": 0.13833602394582134,
+      "kl": 0.5390625,
+      "learning_rate": 5.314685314685315e-05,
+      "loss": -0.0015,
+      "step": 152
+    },
+    {
+      "clip_ratio": 0.0029247915372252464,
+      "completion_length": 104.1964340209961,
+      "epoch": 0.213986013986014,
+      "grad_norm": 0.18332277821774023,
+      "kl": 0.451171875,
+      "learning_rate": 5.34965034965035e-05,
+      "loss": -0.0047,
+      "num_tokens": 1533154.0,
+      "reward": 1.977178931236267,
+      "reward_std": 0.5758928656578064,
+      "rewards/check_gptzero_func": 0.6785714030265808,
+      "rewards/check_perplexity_diff_func": 0.6785714030265808,
+      "rewards/check_winston_local_func": 0.6200361251831055,
+      "step": 153
+    },
+    {
+      "clip_ratio": 0.002894646255299449,
+      "epoch": 0.2153846153846154,
+      "grad_norm": 0.12417162413759981,
+      "kl": 0.48828125,
+      "learning_rate": 5.384615384615385e-05,
+      "loss": -0.0082,
+      "step": 154
+    },
+    {
+      "clip_ratio": 0.002533347113057971,
+      "completion_length": 106.6964340209961,
+      "epoch": 0.21678321678321677,
+      "grad_norm": 0.2165608004438418,
+      "kl": 0.4140625,
+      "learning_rate": 5.41958041958042e-05,
+      "loss": 0.0367,
+      "num_tokens": 1554005.0,
+      "reward": 2.254620313644409,
+      "reward_std": 0.6860859394073486,
+      "rewards/check_gptzero_func": 0.7857142686843872,
+      "rewards/check_perplexity_diff_func": 0.7142857313156128,
+      "rewards/check_winston_local_func": 0.7546200752258301,
+      "step": 155
+    },
+    {
+      "clip_ratio": 0.005156368017196655,
+      "epoch": 0.21818181818181817,
+      "grad_norm": 0.16185524517286934,
+      "kl": 0.4609375,
+      "learning_rate": 5.4545454545454546e-05,
+      "loss": 0.0322,
+      "step": 156
+    },
+    {
+      "clip_ratio": 0.002192203886806965,
+      "completion_length": 129.55357360839844,
+      "epoch": 0.21958041958041957,
+      "grad_norm": 0.13686573639431518,
+      "kl": 0.421875,
+      "learning_rate": 5.48951048951049e-05,
+      "loss": 0.0095,
+      "num_tokens": 1576988.0,
+      "reward": 2.202296495437622,
+      "reward_std": 0.4840867817401886,
+      "rewards/check_gptzero_func": 0.6428571343421936,
+      "rewards/check_perplexity_diff_func": 0.8928571343421936,
+      "rewards/check_winston_local_func": 0.6665821075439453,
+      "step": 157
+    },
+    {
+      "clip_ratio": 0.0037054666317999363,
+      "epoch": 0.22097902097902097,
+      "grad_norm": 0.12040805378810834,
+      "kl": 0.45703125,
+      "learning_rate": 5.524475524475524e-05,
+      "loss": 0.0077,
+      "step": 158
+    },
+    {
+      "clip_ratio": 0.002324200002476573,
+      "completion_length": 141.10714721679688,
+      "epoch": 0.22237762237762237,
+      "grad_norm": 0.1722210877266639,
+      "kl": 0.9765625,
+      "learning_rate": 5.55944055944056e-05,
+      "loss": -0.0049,
+      "num_tokens": 1600826.0,
+      "reward": 2.0245378017425537,
+      "reward_std": 0.45877185463905334,
+      "rewards/check_gptzero_func": 0.5892857313156128,
+      "rewards/check_perplexity_diff_func": 0.75,
+      "rewards/check_winston_local_func": 0.6852518916130066,
+      "step": 159
+    },
+    {
+      "clip_ratio": 0.003470373572781682,
+      "epoch": 0.22377622377622378,
+      "grad_norm": 0.13356363193434526,
+      "kl": 0.65234375,
+      "learning_rate": 5.594405594405595e-05,
+      "loss": -0.0085,
+      "step": 160
+    },
+    {
+      "clip_ratio": 0.0026535126380622387,
+      "completion_length": 109.76786041259766,
+      "epoch": 0.22517482517482518,
+      "grad_norm": 0.409404915148595,
+      "kl": 0.515625,
+      "learning_rate": 5.629370629370629e-05,
+      "loss": -0.0051,
+      "num_tokens": 1621103.0,
+      "reward": 2.0040969848632812,
+      "reward_std": 0.5942196249961853,
+      "rewards/check_gptzero_func": 0.5178571343421936,
+      "rewards/check_perplexity_diff_func": 0.7857142686843872,
+      "rewards/check_winston_local_func": 0.7005256414413452,
+      "step": 161
+    },
+    {
+      "clip_ratio": 0.0058495416305959225,
+      "epoch": 0.22657342657342658,
+      "grad_norm": 0.16376328147020022,
+      "kl": 0.5078125,
+      "learning_rate": 5.664335664335665e-05,
+      "loss": -0.0129,
+      "step": 162
+    },
+    {
+      "clip_ratio": 0.002924378262832761,
+      "completion_length": 114.3214340209961,
+      "epoch": 0.22797202797202798,
+      "grad_norm": 0.17698045766822795,
+      "kl": 0.62109375,
+      "learning_rate": 5.699300699300699e-05,
+      "loss": -0.0053,
+      "num_tokens": 1642629.0,
+      "reward": 2.0913121700286865,
+      "reward_std": 0.533393144607544,
+      "rewards/check_gptzero_func": 0.6785714030265808,
+      "rewards/check_perplexity_diff_func": 0.6785714030265808,
+      "rewards/check_winston_local_func": 0.7341693639755249,
+      "step": 163
+    },
+    {
+      "clip_ratio": 0.006533453240990639,
+      "epoch": 0.22937062937062938,
+      "grad_norm": 0.15566114152311913,
+      "kl": 0.66796875,
+      "learning_rate": 5.7342657342657345e-05,
+      "loss": -0.0085,
+      "step": 164
+    },
+    {
+      "clip_ratio": 0.002033352619037032,
+      "completion_length": 110.48214721679688,
+      "epoch": 0.23076923076923078,
+      "grad_norm": 0.19340316318251602,
+      "kl": 0.4453125,
+      "learning_rate": 5.769230769230769e-05,
+      "loss": 0.0054,
+      "num_tokens": 1663000.0,
+      "reward": 1.809409499168396,
+      "reward_std": 0.5209853053092957,
+      "rewards/check_gptzero_func": 0.6428571343421936,
+      "rewards/check_perplexity_diff_func": 0.3571428656578064,
+      "rewards/check_winston_local_func": 0.809409499168396,
+      "step": 165
+    },
+    {
+      "clip_ratio": 0.005329853855073452,
+      "epoch": 0.23216783216783216,
+      "grad_norm": 0.15299329983895626,
+      "kl": 0.55078125,
+      "learning_rate": 5.8041958041958044e-05,
+      "loss": 0.0018,
+      "step": 166
+    },
+    {
+      "clip_ratio": 0.0025733087677508593,
+      "completion_length": 95.73214721679688,
+      "epoch": 0.23356643356643356,
+      "grad_norm": 0.16102285972903455,
+      "kl": 0.458984375,
+      "learning_rate": 5.83916083916084e-05,
+      "loss": -0.0008,
+      "num_tokens": 1682123.0,
+      "reward": 1.9705681800842285,
+      "reward_std": 0.5465096235275269,
+      "rewards/check_gptzero_func": 0.6428571343421936,
+      "rewards/check_perplexity_diff_func": 0.6428571343421936,
+      "rewards/check_winston_local_func": 0.6848538517951965,
+      "step": 167
+    },
+    {
+      "clip_ratio": 0.0034487086813896894,
+      "epoch": 0.23496503496503496,
+      "grad_norm": 0.12896214559568192,
+      "kl": 0.474609375,
+      "learning_rate": 5.8741258741258744e-05,
+      "loss": -0.0039,
+      "step": 168
+    },
+    {
+      "clip_ratio": 0.0012721805833280087,
+      "completion_length": 72.125,
+      "epoch": 0.23636363636363636,
+      "grad_norm": 0.19976827264055144,
+      "kl": 0.5703125,
+      "learning_rate": 5.90909090909091e-05,
+      "loss": -0.0046,
+      "num_tokens": 1698078.0,
+      "reward": 1.8698246479034424,
+      "reward_std": 0.4430284798145294,
+      "rewards/check_gptzero_func": 0.7142857313156128,
+      "rewards/check_perplexity_diff_func": 0.3571428656578064,
+      "rewards/check_winston_local_func": 0.7983959913253784,
+      "step": 169
+    },
+    {
+      "clip_ratio": 0.005269515328109264,
+      "epoch": 0.23776223776223776,
+      "grad_norm": 0.14855282442633314,
+      "kl": 0.61328125,
+      "learning_rate": 5.944055944055944e-05,
+      "loss": -0.0079,
+      "step": 170
+    },
+    {
+      "clip_ratio": 0.0027731256559491158,
+      "completion_length": 117.83929443359375,
+      "epoch": 0.23916083916083916,
+      "grad_norm": 0.23058249563979205,
+      "kl": 0.412109375,
+      "learning_rate": 5.9790209790209796e-05,
+      "loss": 0.0055,
+      "num_tokens": 1719673.0,
+      "reward": 2.2663450241088867,
+      "reward_std": 0.46630793809890747,
+      "rewards/check_gptzero_func": 0.6785714030265808,
+      "rewards/check_perplexity_diff_func": 0.7142857313156128,
+      "rewards/check_winston_local_func": 0.8734878301620483,
+      "step": 171
+    },
+    {
+      "clip_ratio": 0.0034511894918978214,
+      "epoch": 0.24055944055944056,
+      "grad_norm": 0.17368207545699044,
+      "kl": 0.373046875,
+      "learning_rate": 6.0139860139860136e-05,
+      "loss": -0.0007,
+      "step": 172
+    },
+    {
+      "clip_ratio": 0.0032431341242045164,
+      "completion_length": 126.85714721679688,
+      "epoch": 0.24195804195804196,
+      "grad_norm": 0.17995690811350934,
+      "kl": 0.353515625,
+      "learning_rate": 6.048951048951049e-05,
+      "loss": -0.0011,
+      "num_tokens": 1742569.0,
+      "reward": 2.258990526199341,
+      "reward_std": 0.6008436679840088,
+      "rewards/check_gptzero_func": 0.75,
+      "rewards/check_perplexity_diff_func": 0.75,
+      "rewards/check_winston_local_func": 0.7589904069900513,
+      "step": 173
+    },
+    {
+      "clip_ratio": 0.006950656417757273,
+      "epoch": 0.24335664335664337,
+      "grad_norm": 0.13892848528506246,
+      "kl": 0.359375,
+      "learning_rate": 6.083916083916085e-05,
+      "loss": -0.0033,
+      "step": 174
+    },
+    {
+      "clip_ratio": 0.0027662119828164577,
+      "completion_length": 130.8928680419922,
+      "epoch": 0.24475524475524477,
+      "grad_norm": 0.10040261340554028,
+      "kl": 0.345703125,
+      "learning_rate": 6.118881118881119e-05,
+      "loss": -0.0032,
+      "num_tokens": 1765643.0,
+      "reward": 2.074061632156372,
+      "reward_std": 0.3745954632759094,
+      "rewards/check_gptzero_func": 0.6607142686843872,
+      "rewards/check_perplexity_diff_func": 0.6785714030265808,
+      "rewards/check_winston_local_func": 0.7347758412361145,
+      "step": 175
+    },
+    {
+      "clip_ratio": 0.0030742601957172155,
+      "epoch": 0.24615384615384617,
+      "grad_norm": 0.0921134439221548,
+      "kl": 0.3359375,
+      "learning_rate": 6.153846153846155e-05,
+      "loss": -0.0044,
+      "step": 176
+    },
+    {
+      "clip_ratio": 0.002027927665039897,
+      "completion_length": 89.73214721679688,
+      "epoch": 0.24755244755244754,
+      "grad_norm": 0.1709976566266231,
+      "kl": 0.50390625,
+      "learning_rate": 6.188811188811188e-05,
+      "loss": 0.0051,
+      "num_tokens": 1783674.0,
+      "reward": 1.7810758352279663,
+      "reward_std": 0.4943030774593353,
+      "rewards/check_gptzero_func": 0.6428571343421936,
+      "rewards/check_perplexity_diff_func": 0.4285714328289032,
+      "rewards/check_winston_local_func": 0.7096471786499023,
+      "step": 177
+    },
+    {
+      "clip_ratio": 0.0020612890366464853,
+      "epoch": 0.24895104895104894,
+      "grad_norm": 0.1477275348028994,
+      "kl": 0.50390625,
+      "learning_rate": 6.223776223776224e-05,
+      "loss": 0.0011,
+      "step": 178
+    },
+    {
+      "clip_ratio": 0.0021104796323925257,
+      "completion_length": 132.10714721679688,
+      "epoch": 0.25034965034965034,
+      "grad_norm": 0.08842848405697774,
+      "kl": 0.39453125,
+      "learning_rate": 6.258741258741259e-05,
+      "loss": 0.0005,
+      "num_tokens": 1806754.0,
+      "reward": 2.107767105102539,
+      "reward_std": 0.269815057516098,
+      "rewards/check_gptzero_func": 0.75,
+      "rewards/check_perplexity_diff_func": 0.5357142686843872,
+      "rewards/check_winston_local_func": 0.8220529556274414,
+      "step": 179
+    },
+    {
+      "clip_ratio": 0.004407648928463459,
+      "epoch": 0.2517482517482518,
+      "grad_norm": 0.07542595290297384,
+      "kl": 0.416015625,
+      "learning_rate": 6.293706293706293e-05,
+      "loss": -0.0006,
+      "step": 180
+    },
+    {
+      "clip_ratio": 0.0018615310546010733,
+      "completion_length": 89.30357360839844,
+      "epoch": 0.25314685314685315,
+      "grad_norm": 0.2895549842864223,
+      "kl": 0.498046875,
+      "learning_rate": 6.32867132867133e-05,
+      "loss": -0.0082,
+      "num_tokens": 1824795.0,
+      "reward": 1.8801069259643555,
+      "reward_std": 0.5523244738578796,
+      "rewards/check_gptzero_func": 0.6071428656578064,
+      "rewards/check_perplexity_diff_func": 0.5,
+      "rewards/check_winston_local_func": 0.7729640603065491,
+      "step": 181
+    },
+    {
+      "clip_ratio": 0.007555535528808832,
+      "epoch": 0.2545454545454545,
+      "grad_norm": 0.20874610098022578,
+      "kl": 0.59375,
+      "learning_rate": 6.363636363636364e-05,
+      "loss": -0.016,
+      "step": 182
+    },
+    {
+      "clip_ratio": 0.0016618422232568264,
+      "completion_length": 84.46428680419922,
+      "epoch": 0.25594405594405595,
+      "grad_norm": 0.20653753743630154,
+      "kl": 0.5625,
+      "learning_rate": 6.398601398601399e-05,
+      "loss": 0.0058,
+      "num_tokens": 1842573.0,
+      "reward": 2.1319751739501953,
+      "reward_std": 0.5030335187911987,
+      "rewards/check_gptzero_func": 0.7321428656578064,
+      "rewards/check_perplexity_diff_func": 0.5714285969734192,
+      "rewards/check_winston_local_func": 0.8284037709236145,
+      "step": 183
+    },
+    {
+      "clip_ratio": 0.00579726742580533,
+      "epoch": 0.2573426573426573,
+      "grad_norm": 0.15782001907156346,
+      "kl": 0.578125,
+      "learning_rate": 6.433566433566433e-05,
+      "loss": 0.0001,
+      "step": 184
+    },
+    {
+      "clip_ratio": 0.0016176491044461727,
+      "completion_length": 125.39286041259766,
+      "epoch": 0.25874125874125875,
+      "grad_norm": 0.12385736188741701,
+      "kl": 0.4921875,
+      "learning_rate": 6.46853146853147e-05,
+      "loss": -0.0027,
+      "num_tokens": 1864297.0,
+      "reward": 1.7732529640197754,
+      "reward_std": 0.3774341642856598,
+      "rewards/check_gptzero_func": 0.625,
+      "rewards/check_perplexity_diff_func": 0.4285714328289032,
+      "rewards/check_winston_local_func": 0.7196813821792603,
+      "step": 185
+    },
+    {
+      "clip_ratio": 0.00623862212523818,
+      "epoch": 0.2601398601398601,
+      "grad_norm": 0.10877531810926387,
+      "kl": 0.53125,
+      "learning_rate": 6.503496503496504e-05,
+      "loss": -0.0048,
+      "step": 186
+    },
+    {
+      "clip_ratio": 0.0031279984395951033,
+      "completion_length": 98.00000762939453,
+      "epoch": 0.26153846153846155,
+      "grad_norm": 0.3661429387824046,
+      "kl": 0.6328125,
+      "learning_rate": 6.538461538461539e-05,
+      "loss": -0.0063,
+      "num_tokens": 1882931.0,
+      "reward": 2.0694050788879395,
+      "reward_std": 0.6090536117553711,
+      "rewards/check_gptzero_func": 0.6785714030265808,
+      "rewards/check_perplexity_diff_func": 0.5714285969734192,
+      "rewards/check_winston_local_func": 0.819405198097229,
+      "step": 187
+    },
+    {
+      "clip_ratio": 0.010017934255301952,
+      "epoch": 0.2629370629370629,
+      "grad_norm": 0.34977841113727764,
+      "kl": 0.65234375,
+      "learning_rate": 6.573426573426573e-05,
+      "loss": -0.0185,
+      "step": 188
+    },
+    {
+      "clip_ratio": 0.00222155568189919,
+      "completion_length": 133.44644165039062,
+      "epoch": 0.26433566433566436,
+      "grad_norm": 0.18715243199492715,
+      "kl": 0.46484375,
+      "learning_rate": 6.608391608391609e-05,
+      "loss": 0.0062,
+      "num_tokens": 1906084.0,
+      "reward": 2.364468812942505,
+      "reward_std": 0.5593475699424744,
+      "rewards/check_gptzero_func": 0.6964285969734192,
+      "rewards/check_perplexity_diff_func": 0.8571428656578064,
+      "rewards/check_winston_local_func": 0.8108974099159241,
+      "step": 189
+    },
+    {
+      "clip_ratio": 0.008103223517537117,
+      "epoch": 0.26573426573426573,
+      "grad_norm": 0.14058389303467,
+      "kl": 0.45703125,
+      "learning_rate": 6.643356643356644e-05,
+      "loss": 0.0025,
+      "step": 190
+    },
+    {
+      "clip_ratio": 0.0019668787717819214,
+      "completion_length": 142.07144165039062,
+      "epoch": 0.26713286713286716,
+      "grad_norm": 0.14399812178023874,
+      "kl": 0.43359375,
+      "learning_rate": 6.678321678321679e-05,
+      "loss": -0.0031,
+      "num_tokens": 1930224.0,
+      "reward": 2.1098926067352295,
+      "reward_std": 0.5390018820762634,
+      "rewards/check_gptzero_func": 0.8214285969734192,
+      "rewards/check_perplexity_diff_func": 0.4642857015132904,
+      "rewards/check_winston_local_func": 0.8241782784461975,
+      "step": 191
+    },
+    {
+      "clip_ratio": 0.005082913674414158,
+      "epoch": 0.26853146853146853,
+      "grad_norm": 0.12024703070308183,
+      "kl": 0.447265625,
+      "learning_rate": 6.713286713286715e-05,
+      "loss": -0.0055,
+      "step": 192
+    },
+    {
+      "clip_ratio": 0.0032251765951514244,
+      "completion_length": 84.98214721679688,
+      "epoch": 0.2699300699300699,
+      "grad_norm": 0.32527529059153654,
+      "kl": 0.6640625,
+      "learning_rate": 6.748251748251748e-05,
+      "loss": 0.0186,
+      "num_tokens": 1947399.0,
+      "reward": 1.9443087577819824,
+      "reward_std": 0.6417390704154968,
+      "rewards/check_gptzero_func": 0.7678571343421936,
+      "rewards/check_perplexity_diff_func": 0.2857142984867096,
+      "rewards/check_winston_local_func": 0.8907372355461121,
+      "step": 193
+    },
+    {
+      "clip_ratio": 0.008085422217845917,
+      "epoch": 0.27132867132867133,
+      "grad_norm": 0.21806984416977268,
+      "kl": 0.69140625,
+      "learning_rate": 6.783216783216784e-05,
+      "loss": 0.0068,
+      "step": 194
+    },
+    {
+      "clip_ratio": 0.0025381618179380894,
+      "completion_length": 91.66072082519531,
+      "epoch": 0.2727272727272727,
+      "grad_norm": 0.1831840025015104,
+      "kl": 0.53125,
+      "learning_rate": 6.818181818181818e-05,
+      "loss": 0.0014,
+      "num_tokens": 1965402.0,
+      "reward": 2.3979477882385254,
+      "reward_std": 0.45264866948127747,
+      "rewards/check_gptzero_func": 0.7678571343421936,
+      "rewards/check_perplexity_diff_func": 0.7857142686843872,
+      "rewards/check_winston_local_func": 0.8443759083747864,
+      "step": 195
+    },
+    {
+      "clip_ratio": 0.005082620773464441,
+      "epoch": 0.27412587412587414,
+      "grad_norm": 0.14399845631168656,
+      "kl": 0.515625,
+      "learning_rate": 6.853146853146853e-05,
+      "loss": -0.0023,
+      "step": 196
+    },
+    {
+      "clip_ratio": 0.0025359569117426872,
+      "completion_length": 118.80357360839844,
+      "epoch": 0.2755244755244755,
+      "grad_norm": 0.1617404225919446,
+      "kl": 0.5625,
+      "learning_rate": 6.888111888111889e-05,
+      "loss": 0.0146,
+      "num_tokens": 1986731.0,
+      "reward": 2.078895330429077,
+      "reward_std": 0.31289052963256836,
+      "rewards/check_gptzero_func": 0.8035714030265808,
+      "rewards/check_perplexity_diff_func": 0.3928571343421936,
+      "rewards/check_winston_local_func": 0.882466733455658,
+      "step": 197
+    },
+    {
+      "clip_ratio": 0.0062335156835615635,
+      "epoch": 0.27692307692307694,
+      "grad_norm": 0.12006362566688389,
+      "kl": 0.69921875,
+      "learning_rate": 6.923076923076924e-05,
+      "loss": 0.0125,
+      "step": 198
+    },
+    {
+      "clip_ratio": 0.0023628328926861286,
+      "completion_length": 113.76786041259766,
+      "epoch": 0.2783216783216783,
+      "grad_norm": 0.19235514876641358,
+      "kl": 0.58984375,
+      "learning_rate": 6.958041958041958e-05,
+      "loss": -0.0069,
+      "num_tokens": 2007822.0,
+      "reward": 2.0453529357910156,
+      "reward_std": 0.576248288154602,
+      "rewards/check_gptzero_func": 0.7321428656578064,
+      "rewards/check_perplexity_diff_func": 0.5714285969734192,
+      "rewards/check_winston_local_func": 0.7417814135551453,
+      "step": 199
+    },
+    {
+      "clip_ratio": 0.0030245708767324686,
+      "epoch": 0.27972027972027974,
+      "grad_norm": 0.15320753759426498,
+      "kl": 0.578125,
+      "learning_rate": 6.993006993006993e-05,
+      "loss": -0.0115,
+      "step": 200
+    },
+    {
+      "clip_ratio": 0.0028001146856695414,
+      "completion_length": 121.51786041259766,
+      "epoch": 0.2811188811188811,
+      "grad_norm": 0.2770839119590159,
+      "kl": 0.5859375,
+      "learning_rate": 7.027972027972029e-05,
+      "loss": -0.0321,
+      "num_tokens": 2030153.0,
+      "reward": 2.223505735397339,
+      "reward_std": 0.580276370048523,
+      "rewards/check_gptzero_func": 0.8392857313156128,
+      "rewards/check_perplexity_diff_func": 0.5,
+      "rewards/check_winston_local_func": 0.8842198252677917,
+      "step": 201
+    },
+    {
+      "clip_ratio": 0.004435424692928791,
+      "epoch": 0.28251748251748254,
+      "grad_norm": 0.1752123363810394,
+      "kl": 0.61328125,
+      "learning_rate": 7.062937062937062e-05,
+      "loss": -0.0398,
+      "step": 202
+    },
+    {
+      "clip_ratio": 0.003775001736357808,
+      "completion_length": 132.96429443359375,
+      "epoch": 0.2839160839160839,
+      "grad_norm": 0.20528144857768493,
+      "kl": 0.5390625,
+      "learning_rate": 7.097902097902098e-05,
+      "loss": -0.0054,
+      "num_tokens": 2053611.0,
+      "reward": 2.4398131370544434,
+      "reward_std": 0.6120996475219727,
+      "rewards/check_gptzero_func": 0.8214285969734192,
+      "rewards/check_perplexity_diff_func": 0.8571428656578064,
+      "rewards/check_winston_local_func": 0.761241614818573,
+      "step": 203
+    },
+    {
+      "clip_ratio": 0.0037133130244910717,
+      "epoch": 0.2853146853146853,
+      "grad_norm": 0.14916355666745199,
+      "kl": 0.55859375,
+      "learning_rate": 7.132867132867134e-05,
+      "loss": -0.0112,
+      "step": 204
+    },
+    {
+      "clip_ratio": 0.002573356730863452,
+      "completion_length": 109.51786041259766,
+      "epoch": 0.2867132867132867,
+      "grad_norm": 0.15672915743695726,
+      "kl": 1.4921875,
+      "learning_rate": 7.167832167832168e-05,
+      "loss": 0.0032,
+      "num_tokens": 2073846.0,
+      "reward": 2.221467971801758,
+      "reward_std": 0.492183119058609,
+      "rewards/check_gptzero_func": 0.6964285969734192,
+      "rewards/check_perplexity_diff_func": 0.6428571343421936,
+      "rewards/check_winston_local_func": 0.8821821212768555,
+      "step": 205
+    },
+    {
+      "clip_ratio": 0.0036338225472718477,
+      "epoch": 0.2881118881118881,
+      "grad_norm": 0.27785877634425765,
+      "kl": 0.74609375,
+      "learning_rate": 7.202797202797204e-05,
+      "loss": 0.0013,
+      "step": 206
+    },
+    {
+      "clip_ratio": 0.004580673761665821,
+      "completion_length": 95.05357360839844,
+      "epoch": 0.2895104895104895,
+      "grad_norm": 0.1746155777114636,
+      "kl": 0.671875,
+      "learning_rate": 7.237762237762238e-05,
+      "loss": -0.0068,
+      "num_tokens": 2092355.0,
+      "reward": 2.0429205894470215,
+      "reward_std": 0.36572587490081787,
+      "rewards/check_gptzero_func": 0.6964285969734192,
+      "rewards/check_perplexity_diff_func": 0.4642857015132904,
+      "rewards/check_winston_local_func": 0.8822061419487,
+      "step": 207
+    },
+    {
+      "clip_ratio": 0.007289381232112646,
+      "epoch": 0.2909090909090909,
+      "grad_norm": 0.12704093759252294,
+      "kl": 0.75390625,
+      "learning_rate": 7.272727272727273e-05,
+      "loss": -0.0111,
+      "step": 208
+    },
+    {
+      "clip_ratio": 0.0038105440326035023,
+      "completion_length": 113.4464340209961,
+      "epoch": 0.2923076923076923,
+      "grad_norm": 0.20958397036994925,
+      "kl": 0.6015625,
+      "learning_rate": 7.307692307692307e-05,
+      "loss": 0.0082,
+      "num_tokens": 2113542.0,
+      "reward": 2.162785530090332,
+      "reward_std": 0.49807849526405334,
+      "rewards/check_gptzero_func": 0.75,
+      "rewards/check_perplexity_diff_func": 0.5714285969734192,
+      "rewards/check_winston_local_func": 0.8413568139076233,
+      "step": 209
+    },
+    {
+      "clip_ratio": 0.007606918923556805,
+      "epoch": 0.2937062937062937,
+      "grad_norm": 0.1766466418371376,
+      "kl": 0.63671875,
+      "learning_rate": 7.342657342657343e-05,
+      "loss": 0.0025,
+      "step": 210
+    },
+    {
+      "clip_ratio": 0.0019986790139228106,
+      "completion_length": 92.73214721679688,
+      "epoch": 0.2951048951048951,
+      "grad_norm": 0.2716166901527731,
+      "kl": 0.6171875,
+      "learning_rate": 7.377622377622378e-05,
+      "loss": -0.0005,
+      "num_tokens": 2131873.0,
+      "reward": 2.099912405014038,
+      "reward_std": 0.5661742687225342,
+      "rewards/check_gptzero_func": 0.8214285969734192,
+      "rewards/check_perplexity_diff_func": 0.4285714328289032,
+      "rewards/check_winston_local_func": 0.8499122858047485,
+      "step": 211
+    },
+    {
+      "clip_ratio": 0.005917501635849476,
+      "epoch": 0.2965034965034965,
+      "grad_norm": 0.19270405158731238,
+      "kl": 0.62890625,
+      "learning_rate": 7.412587412587413e-05,
+      "loss": -0.0079,
+      "step": 212
+    },
+    {
+      "clip_ratio": 0.002530448604375124,
+      "completion_length": 96.51786041259766,
+      "epoch": 0.29790209790209793,
+      "grad_norm": 0.22723256705117592,
+      "kl": 0.69921875,
+      "learning_rate": 7.447552447552449e-05,
+      "loss": -0.009,
+      "num_tokens": 2150462.0,
+      "reward": 2.1355700492858887,
+      "reward_std": 0.3738899827003479,
+      "rewards/check_gptzero_func": 0.8571428656578064,
+      "rewards/check_perplexity_diff_func": 0.3571428656578064,
+      "rewards/check_winston_local_func": 0.9212842583656311,
+      "step": 213
+    },
+    {
+      "clip_ratio": 0.011397747322916985,
+      "epoch": 0.2993006993006993,
+      "grad_norm": 0.18189402035580596,
+      "kl": 0.7265625,
+      "learning_rate": 7.482517482517482e-05,
+      "loss": -0.0139,
+      "step": 214
+    },
+    {
+      "clip_ratio": 0.004513947293162346,
+      "completion_length": 94.66072082519531,
+      "epoch": 0.3006993006993007,
+      "grad_norm": 0.2675295151019955,
+      "kl": 0.67578125,
+      "learning_rate": 7.517482517482518e-05,
+      "loss": 0.0014,
+      "num_tokens": 2169243.0,
+      "reward": 2.2817916870117188,
+      "reward_std": 0.4118366539478302,
+      "rewards/check_gptzero_func": 0.8571428656578064,
+      "rewards/check_perplexity_diff_func": 0.4642857015132904,
+      "rewards/check_winston_local_func": 0.9603630900382996,
+      "step": 215
+    },
+    {
+      "clip_ratio": 0.010964194312691689,
+      "epoch": 0.3020979020979021,
+      "grad_norm": 0.17031623307626814,
+      "kl": 0.875,
+      "learning_rate": 7.552447552447553e-05,
+      "loss": -0.0057,
+      "step": 216
+    },
+    {
+      "clip_ratio": 0.003371638245880604,
+      "completion_length": 118.17857360839844,
+      "epoch": 0.3034965034965035,
+      "grad_norm": 0.1847364030092435,
+      "kl": 0.60546875,
+      "learning_rate": 7.587412587412587e-05,
+      "loss": -0.0022,
+      "num_tokens": 2190571.0,
+      "reward": 2.1225900650024414,
+      "reward_std": 0.5403507947921753,
+      "rewards/check_gptzero_func": 0.75,
+      "rewards/check_perplexity_diff_func": 0.4642857015132904,
+      "rewards/check_winston_local_func": 0.9083043336868286,
+      "step": 217
+    },
+    {
+      "clip_ratio": 0.01002263929694891,
+      "epoch": 0.3048951048951049,
+      "grad_norm": 0.1449002041728082,
+      "kl": 0.61328125,
+      "learning_rate": 7.622377622377622e-05,
+      "loss": -0.0067,
+      "step": 218
+    },
+    {
+      "clip_ratio": 0.0013253266224637628,
+      "completion_length": 129.33929443359375,
+      "epoch": 0.3062937062937063,
+      "grad_norm": 0.1504069021149342,
+      "kl": 0.578125,
+      "learning_rate": 7.657342657342658e-05,
+      "loss": -0.0075,
+      "num_tokens": 2213408.0,
+      "reward": 2.1974618434906006,
+      "reward_std": 0.4602964520454407,
+      "rewards/check_gptzero_func": 0.7857142686843872,
+      "rewards/check_perplexity_diff_func": 0.5,
+      "rewards/check_winston_local_func": 0.911747395992279,
+      "step": 219
+    },
+    {
+      "clip_ratio": 0.006876418832689524,
+      "epoch": 0.3076923076923077,
+      "grad_norm": 0.11992971275284438,
+      "kl": 0.57421875,
+      "learning_rate": 7.692307692307693e-05,
+      "loss": -0.0101,
+      "step": 220
+    },
+    {
+      "clip_ratio": 0.0024474726524204016,
+      "completion_length": 89.67857360839844,
+      "epoch": 0.3090909090909091,
+      "grad_norm": 0.46915411380362765,
+      "kl": 0.73046875,
+      "learning_rate": 7.727272727272727e-05,
+      "loss": 0.0333,
+      "num_tokens": 2231248.0,
+      "reward": 2.161350727081299,
+      "reward_std": 0.47136664390563965,
+      "rewards/check_gptzero_func": 0.8214285969734192,
+      "rewards/check_perplexity_diff_func": 0.4642857015132904,
+      "rewards/check_winston_local_func": 0.8756363987922668,
+      "step": 221
+    },
+    {
+      "clip_ratio": 0.02464432455599308,
+      "epoch": 0.3104895104895105,
+      "grad_norm": 0.30204889343007607,
+      "kl": 0.8046875,
+      "learning_rate": 7.762237762237763e-05,
+      "loss": 0.0211,
+      "step": 222
+    },
+    {
+      "clip_ratio": 0.0017527465242892504,
+      "completion_length": 134.17857360839844,
+      "epoch": 0.3118881118881119,
+      "grad_norm": 0.18948229022832336,
+      "kl": 0.6328125,
+      "learning_rate": 7.797202797202798e-05,
+      "loss": -0.0049,
+      "num_tokens": 2255162.0,
+      "reward": 2.4446465969085693,
+      "reward_std": 0.5171672105789185,
+      "rewards/check_gptzero_func": 0.7678571343421936,
+      "rewards/check_perplexity_diff_func": 0.7857142686843872,
+      "rewards/check_winston_local_func": 0.8910752534866333,
+      "step": 223
+    },
+    {
+      "clip_ratio": 0.006593752186745405,
+      "epoch": 0.3132867132867133,
+      "grad_norm": 0.15013602627394823,
+      "kl": 0.640625,
+      "learning_rate": 7.832167832167832e-05,
+      "loss": -0.0097,
+      "step": 224
+    },
+    {
+      "clip_ratio": 0.0019926591776311398,
+      "completion_length": 127.00000762939453,
+      "epoch": 0.3146853146853147,
+      "grad_norm": 0.22875481051435598,
+      "kl": 0.6875,
+      "learning_rate": 7.867132867132867e-05,
+      "loss": 0.0021,
+      "num_tokens": 2277320.0,
+      "reward": 2.2040371894836426,
+      "reward_std": 0.5528277158737183,
+      "rewards/check_gptzero_func": 0.7321428656578064,
+      "rewards/check_perplexity_diff_func": 0.5357142686843872,
+      "rewards/check_winston_local_func": 0.936180055141449,
+      "step": 225
+    },
+    {
+      "clip_ratio": 0.005839359946548939,
+      "epoch": 0.31608391608391606,
+      "grad_norm": 0.17829200080123941,
+      "kl": 0.68359375,
+      "learning_rate": 7.902097902097903e-05,
+      "loss": -0.0053,
+      "step": 226
+    },
+    {
+      "clip_ratio": 0.0026252593379467726,
+      "completion_length": 130.0357208251953,
+      "epoch": 0.3174825174825175,
+      "grad_norm": 0.16076101957107508,
+      "kl": 0.66015625,
+      "learning_rate": 7.937062937062938e-05,
+      "loss": -0.0033,
+      "num_tokens": 2299818.0,
+      "reward": 2.3278117179870605,
+      "reward_std": 0.48680880665779114,
+      "rewards/check_gptzero_func": 0.8035714030265808,
+      "rewards/check_perplexity_diff_func": 0.6428571343421936,
+      "rewards/check_winston_local_func": 0.8813830614089966,
+      "step": 227
+    },
+    {
+      "clip_ratio": 0.004954828415066004,
+      "epoch": 0.31888111888111886,
+      "grad_norm": 0.12734264227867195,
+      "kl": 0.6875,
+      "learning_rate": 7.972027972027972e-05,
+      "loss": -0.0071,
+      "step": 228
+    },
+    {
+      "clip_ratio": 0.0034341278951615095,
+      "completion_length": 133.7857208251953,
+      "epoch": 0.3202797202797203,
+      "grad_norm": 0.24065024754984402,
+      "kl": 0.6875,
+      "learning_rate": 8.006993006993007e-05,
+      "loss": -0.0052,
+      "num_tokens": 2322948.0,
+      "reward": 2.5632615089416504,
+      "reward_std": 0.6206622123718262,
+      "rewards/check_gptzero_func": 0.9642857313156128,
+      "rewards/check_perplexity_diff_func": 0.6785714030265808,
+      "rewards/check_winston_local_func": 0.9204041361808777,
+      "step": 229
+    },
+    {
+      "clip_ratio": 0.006521092262119055,
+      "epoch": 0.32167832167832167,
+      "grad_norm": 0.18602361781154253,
+      "kl": 0.6796875,
+      "learning_rate": 8.041958041958042e-05,
+      "loss": -0.013,
+      "step": 230
+    },
+    {
+      "clip_ratio": 0.004207184072583914,
+      "completion_length": 116.53572082519531,
+      "epoch": 0.3230769230769231,
+      "grad_norm": 0.36704569865090053,
+      "kl": 0.69921875,
+      "learning_rate": 8.076923076923078e-05,
+      "loss": 0.0055,
+      "num_tokens": 2344106.0,
+      "reward": 2.4917171001434326,
+      "reward_std": 0.6073458790779114,
+      "rewards/check_gptzero_func": 0.8392857313156128,
+      "rewards/check_perplexity_diff_func": 0.75,
+      "rewards/check_winston_local_func": 0.902431309223175,
+      "step": 231
+    },
+    {
+      "clip_ratio": 0.013317321427166462,
+      "epoch": 0.32447552447552447,
+      "grad_norm": 0.13844634287775134,
+      "kl": 0.73046875,
+      "learning_rate": 8.111888111888112e-05,
+      "loss": -0.0039,
+      "step": 232
+    },
+    {
+      "clip_ratio": 0.004024635534733534,
+      "completion_length": 134.92857360839844,
+      "epoch": 0.3258741258741259,
+      "grad_norm": 0.34168858248313216,
+      "kl": 0.64453125,
+      "learning_rate": 8.146853146853147e-05,
+      "loss": -0.0057,
+      "num_tokens": 2367724.0,
+      "reward": 2.20210599899292,
+      "reward_std": 0.5888614654541016,
+      "rewards/check_gptzero_func": 0.7321428656578064,
+      "rewards/check_perplexity_diff_func": 0.6071428656578064,
+      "rewards/check_winston_local_func": 0.8628200888633728,
+      "step": 233
+    },
+    {
+      "clip_ratio": 0.009708845987915993,
+      "epoch": 0.32727272727272727,
+      "grad_norm": 0.19135959991877635,
+      "kl": 0.67578125,
+      "learning_rate": 8.181818181818183e-05,
+      "loss": -0.0146,
+      "step": 234
+    },
+    {
+      "clip_ratio": 0.005571329966187477,
+      "completion_length": 98.75000762939453,
+      "epoch": 0.32867132867132864,
+      "grad_norm": 0.32328761096657005,
+      "kl": 0.77734375,
+      "learning_rate": 8.216783216783218e-05,
+      "loss": 0.0128,
+      "num_tokens": 2386636.0,
+      "reward": 2.2358410358428955,
+      "reward_std": 0.599101185798645,
+      "rewards/check_gptzero_func": 0.8214285969734192,
+      "rewards/check_perplexity_diff_func": 0.5357142686843872,
+      "rewards/check_winston_local_func": 0.8786982297897339,
+      "step": 235
+    },
+    {
+      "clip_ratio": 0.009436835534870625,
+      "epoch": 0.3300699300699301,
+      "grad_norm": 0.19557085227388898,
+      "kl": 0.8515625,
+      "learning_rate": 8.251748251748252e-05,
+      "loss": 0.0022,
+      "step": 236
+    },
+    {
+      "clip_ratio": 0.006164635997265577,
+      "completion_length": 79.75,
+      "epoch": 0.33146853146853145,
+      "grad_norm": 0.3250433111655094,
+      "kl": 0.98828125,
+      "learning_rate": 8.286713286713287e-05,
+      "loss": -0.0126,
+      "num_tokens": 2403400.0,
+      "reward": 2.1074206829071045,
+      "reward_std": 0.41642776131629944,
+      "rewards/check_gptzero_func": 0.9107142686843872,
+      "rewards/check_perplexity_diff_func": 0.25,
+      "rewards/check_winston_local_func": 0.946706235408783,
+      "step": 237
+    },
+    {
+      "clip_ratio": 0.0066314926370978355,
+      "epoch": 0.3328671328671329,
+      "grad_norm": 0.1612667131666411,
+      "kl": 0.98828125,
+      "learning_rate": 8.321678321678323e-05,
+      "loss": -0.0225,
+      "step": 238
+    },
+    {
+      "clip_ratio": 0.003079179208725691,
+      "completion_length": 117.10714721679688,
+      "epoch": 0.33426573426573425,
+      "grad_norm": 0.2505534626607256,
+      "kl": 0.6796875,
+      "learning_rate": 8.356643356643356e-05,
+      "loss": -0.0103,
+      "num_tokens": 2425354.0,
+      "reward": 2.3166799545288086,
+      "reward_std": 0.5270359516143799,
+      "rewards/check_gptzero_func": 0.8214285969734192,
+      "rewards/check_perplexity_diff_func": 0.5714285969734192,
+      "rewards/check_winston_local_func": 0.9238227605819702,
+      "step": 239
+    },
+    {
+      "clip_ratio": 0.006801524665206671,
+      "epoch": 0.3356643356643357,
+      "grad_norm": 0.16826419365505382,
+      "kl": 0.671875,
+      "learning_rate": 8.391608391608392e-05,
+      "loss": -0.0182,
+      "step": 240
+    },
+    {
+      "clip_ratio": 0.004480496048927307,
+      "completion_length": 125.05357360839844,
+      "epoch": 0.33706293706293705,
+      "grad_norm": 0.26807661638826913,
+      "kl": 0.60546875,
+      "learning_rate": 8.426573426573428e-05,
+      "loss": -0.0086,
+      "num_tokens": 2447333.0,
+      "reward": 2.2939677238464355,
+      "reward_std": 0.39558398723602295,
+      "rewards/check_gptzero_func": 0.9107142686843872,
+      "rewards/check_perplexity_diff_func": 0.5357142686843872,
+      "rewards/check_winston_local_func": 0.8475390672683716,
+      "step": 241
+    },
+    {
+      "clip_ratio": 0.007587091531604528,
+      "epoch": 0.3384615384615385,
+      "grad_norm": 0.13602673124742348,
+      "kl": 0.57421875,
+      "learning_rate": 8.461538461538461e-05,
+      "loss": -0.0152,
+      "step": 242
+    },
+    {
+      "clip_ratio": 0.0034923183266073465,
+      "completion_length": 98.71428680419922,
+      "epoch": 0.33986013986013985,
+      "grad_norm": 0.7773422467135167,
+      "kl": 6.5,
+      "learning_rate": 8.496503496503497e-05,
+      "loss": 0.0071,
+      "num_tokens": 2466435.0,
+      "reward": 2.4026196002960205,
+      "reward_std": 0.4264836013317108,
+      "rewards/check_gptzero_func": 0.8571428656578064,
+      "rewards/check_perplexity_diff_func": 0.6785714030265808,
+      "rewards/check_winston_local_func": 0.8669052124023438,
+      "step": 243
+    },
+    {
+      "clip_ratio": 0.026189187541604042,
+      "epoch": 0.3412587412587413,
+      "grad_norm": 130.9971047023429,
+      "kl": 0.70703125,
+      "learning_rate": 8.531468531468532e-05,
+      "loss": 1.8998,
+      "step": 244
+    },
+    {
+      "clip_ratio": 0.004780410788953304,
+      "completion_length": 135.0,
+      "epoch": 0.34265734265734266,
+      "grad_norm": 0.2329479243774761,
+      "kl": 0.87109375,
+      "learning_rate": 8.566433566433567e-05,
+      "loss": 0.0011,
+      "num_tokens": 2489651.0,
+      "reward": 2.2166569232940674,
+      "reward_std": 0.5797281265258789,
+      "rewards/check_gptzero_func": 0.9107142686843872,
+      "rewards/check_perplexity_diff_func": 0.4285714328289032,
+      "rewards/check_winston_local_func": 0.8773713111877441,
+      "step": 245
+    },
+    {
+      "clip_ratio": 0.021273447200655937,
+      "epoch": 0.34405594405594403,
+      "grad_norm": 0.22672502397660207,
+      "kl": 0.9609375,
+      "learning_rate": 8.601398601398601e-05,
+      "loss": -0.0045,
+      "step": 246
+    },
+    {
+      "clip_ratio": 0.002856873208656907,
+      "completion_length": 122.66072082519531,
+      "epoch": 0.34545454545454546,
+      "grad_norm": 0.19772681964626543,
+      "kl": 0.984375,
+      "learning_rate": 8.636363636363637e-05,
+      "loss": 0.0028,
+      "num_tokens": 2510506.0,
+      "reward": 2.323245048522949,
+      "reward_std": 0.46726977825164795,
+      "rewards/check_gptzero_func": 0.9107142686843872,
+      "rewards/check_perplexity_diff_func": 0.5,
+      "rewards/check_winston_local_func": 0.9125306010246277,
+      "step": 247
+    },
+    {
+      "clip_ratio": 0.02299562282860279,
+      "epoch": 0.34685314685314683,
+      "grad_norm": 0.1812005085005865,
+      "kl": 0.87890625,
+      "learning_rate": 8.67132867132867e-05,
+      "loss": -0.001,
+      "step": 248
+    },
+    {
+      "clip_ratio": 0.0026335662696510553,
+      "completion_length": 116.85714721679688,
+      "epoch": 0.34825174825174826,
+      "grad_norm": 0.28731547808119623,
+      "kl": 0.76953125,
+      "learning_rate": 8.706293706293707e-05,
+      "loss": 0.024,
+      "num_tokens": 2531254.0,
+      "reward": 2.5312695503234863,
+      "reward_std": 0.5851892828941345,
+      "rewards/check_gptzero_func": 0.8571428656578064,
+      "rewards/check_perplexity_diff_func": 0.8214285969734192,
+      "rewards/check_winston_local_func": 0.8526979088783264,
+      "step": 249
+    },
+    {
+      "clip_ratio": 0.012148472480475903,
+      "epoch": 0.34965034965034963,
+      "grad_norm": 0.21250398409595742,
+      "kl": 0.734375,
+      "learning_rate": 8.741258741258743e-05,
+      "loss": 0.0156,
+      "step": 250
+    },
+    {
+      "clip_ratio": 0.0036881309933960438,
+      "completion_length": 109.12500762939453,
+      "epoch": 0.35104895104895106,
+      "grad_norm": 0.278787459589393,
+      "kl": 1.3671875,
+      "learning_rate": 8.776223776223776e-05,
+      "loss": -0.0127,
+      "num_tokens": 2551265.0,
+      "reward": 2.24423885345459,
+      "reward_std": 0.6221095323562622,
+      "rewards/check_gptzero_func": 0.875,
+      "rewards/check_perplexity_diff_func": 0.5,
+      "rewards/check_winston_local_func": 0.8692389130592346,
+      "step": 251
+    },
+    {
+      "clip_ratio": 0.011438491754233837,
+      "epoch": 0.35244755244755244,
+      "grad_norm": 0.21324998809887244,
+      "kl": 1.0859375,
+      "learning_rate": 8.811188811188812e-05,
+      "loss": -0.023,
+      "step": 252
+    },
+    {
+      "clip_ratio": 0.0039367591962218285,
+      "completion_length": 112.14286041259766,
+      "epoch": 0.35384615384615387,
+      "grad_norm": 0.31069817037713615,
+      "kl": 0.91015625,
+      "learning_rate": 8.846153846153847e-05,
+      "loss": 0.0167,
+      "num_tokens": 2572119.0,
+      "reward": 2.492161989212036,
+      "reward_std": 0.6407585144042969,
+      "rewards/check_gptzero_func": 0.875,
+      "rewards/check_perplexity_diff_func": 0.75,
+      "rewards/check_winston_local_func": 0.8671619296073914,
+      "step": 253
+    },
+    {
+      "clip_ratio": 0.005880096461623907,
+      "epoch": 0.35524475524475524,
+      "grad_norm": 0.19008007229653096,
+      "kl": 0.90234375,
+      "learning_rate": 8.881118881118881e-05,
+      "loss": 0.0057,
+      "step": 254
+    },
+    {
+      "clip_ratio": 0.003151817014440894,
+      "completion_length": 120.78572082519531,
+      "epoch": 0.35664335664335667,
+      "grad_norm": 0.4113945542101958,
+      "kl": 0.890625,
+      "learning_rate": 8.916083916083916e-05,
+      "loss": 0.0079,
+      "num_tokens": 2593081.0,
+      "reward": 2.3228745460510254,
+      "reward_std": 0.5329480171203613,
+      "rewards/check_gptzero_func": 0.875,
+      "rewards/check_perplexity_diff_func": 0.6428571343421936,
+      "rewards/check_winston_local_func": 0.8050170540809631,
+      "step": 255
+    },
+    {
+      "clip_ratio": 0.01224872563034296,
+      "epoch": 0.35804195804195804,
+      "grad_norm": 0.24677625747594412,
+      "kl": 1.0078125,
+      "learning_rate": 8.951048951048952e-05,
+      "loss": 0.0006,
+      "step": 256
+    },
+    {
+      "clip_ratio": 0.003666934324428439,
+      "completion_length": 146.5178680419922,
+      "epoch": 0.3594405594405594,
+      "grad_norm": 0.21537926236506497,
+      "kl": 0.484375,
+      "learning_rate": 8.986013986013986e-05,
+      "loss": 0.0068,
+      "num_tokens": 2618758.0,
+      "reward": 2.198406934738159,
+      "reward_std": 0.3875991106033325,
+      "rewards/check_gptzero_func": 0.9285714030265808,
+      "rewards/check_perplexity_diff_func": 0.4285714328289032,
+      "rewards/check_winston_local_func": 0.8412641882896423,
+      "step": 257
+    },
+    {
+      "clip_ratio": 0.012731654569506645,
+      "epoch": 0.36083916083916084,
+      "grad_norm": 0.1543455831571161,
+      "kl": 0.478515625,
+      "learning_rate": 9.020979020979021e-05,
+      "loss": 0.0025,
+      "step": 258
+    },
+    {
+      "clip_ratio": 0.0012050624936819077,
+      "completion_length": 129.375,
+      "epoch": 0.3622377622377622,
+      "grad_norm": 0.16107873823673136,
+      "kl": 1.6640625,
+      "learning_rate": 9.055944055944057e-05,
+      "loss": 0.0054,
+      "num_tokens": 2641815.0,
+      "reward": 2.4805288314819336,
+      "reward_std": 0.3510296940803528,
+      "rewards/check_gptzero_func": 0.8392857313156128,
+      "rewards/check_perplexity_diff_func": 0.8214285969734192,
+      "rewards/check_winston_local_func": 0.8198142647743225,
+      "step": 259
+    },
+    {
+      "clip_ratio": 0.006208478473126888,
+      "epoch": 0.36363636363636365,
+      "grad_norm": 0.10546346547579268,
+      "kl": 0.8046875,
+      "learning_rate": 9.090909090909092e-05,
+      "loss": 0.0019,
+      "step": 260
+    },
+    {
+      "clip_ratio": 0.0025593352038413286,
+      "completion_length": 128.57144165039062,
+      "epoch": 0.365034965034965,
+      "grad_norm": 0.1986264685447053,
+      "kl": 0.6171875,
+      "learning_rate": 9.125874125874126e-05,
+      "loss": -0.0025,
+      "num_tokens": 2664343.0,
+      "reward": 2.261420249938965,
+      "reward_std": 0.5704914927482605,
+      "rewards/check_gptzero_func": 0.7857142686843872,
+      "rewards/check_perplexity_diff_func": 0.6428571343421936,
+      "rewards/check_winston_local_func": 0.8328486680984497,
+      "step": 261
+    },
+    {
+      "clip_ratio": 0.002767725382000208,
+      "epoch": 0.36643356643356645,
+      "grad_norm": 0.14267481856078115,
+      "kl": 0.66015625,
+      "learning_rate": 9.160839160839161e-05,
+      "loss": -0.0071,
+      "step": 262
+    },
+    {
+      "clip_ratio": 0.0036019759718328714,
+      "completion_length": 108.30357360839844,
+      "epoch": 0.3678321678321678,
+      "grad_norm": 0.25429899311591847,
+      "kl": 0.61328125,
+      "learning_rate": 9.195804195804196e-05,
+      "loss": 0.0114,
+      "num_tokens": 2684460.0,
+      "reward": 2.1853349208831787,
+      "reward_std": 0.5545978546142578,
+      "rewards/check_gptzero_func": 0.7142857313156128,
+      "rewards/check_perplexity_diff_func": 0.6428571343421936,
+      "rewards/check_winston_local_func": 0.828191876411438,
+      "step": 263
+    },
+    {
+      "clip_ratio": 0.002995865885168314,
+      "epoch": 0.36923076923076925,
+      "grad_norm": 0.17454926732818832,
+      "kl": 0.62109375,
+      "learning_rate": 9.230769230769232e-05,
+      "loss": 0.0028,
+      "step": 264
+    },
+    {
+      "clip_ratio": 0.002558046253398061,
+      "completion_length": 102.80357360839844,
+      "epoch": 0.3706293706293706,
+      "grad_norm": 0.18701691067416304,
+      "kl": 0.734375,
+      "learning_rate": 9.265734265734266e-05,
+      "loss": 0.0401,
+      "num_tokens": 2704225.0,
+      "reward": 2.1770143508911133,
+      "reward_std": 0.573693037033081,
+      "rewards/check_gptzero_func": 0.8035714030265808,
+      "rewards/check_perplexity_diff_func": 0.5357142686843872,
+      "rewards/check_winston_local_func": 0.8377286195755005,
+      "step": 265
+    },
+    {
+      "clip_ratio": 0.006838452070951462,
+      "epoch": 0.37202797202797205,
+      "grad_norm": 0.13942096443359253,
+      "kl": 0.71875,
+      "learning_rate": 9.300699300699301e-05,
+      "loss": 0.0354,
+      "step": 266
+    },
+    {
+      "clip_ratio": 0.00375689216889441,
+      "completion_length": 106.37500762939453,
+      "epoch": 0.3734265734265734,
+      "grad_norm": 0.25266992310982356,
+      "kl": 0.7421875,
+      "learning_rate": 9.335664335664336e-05,
+      "loss": -0.0156,
+      "num_tokens": 2724530.0,
+      "reward": 2.11887526512146,
+      "reward_std": 0.5366143584251404,
+      "rewards/check_gptzero_func": 0.875,
+      "rewards/check_perplexity_diff_func": 0.4642857015132904,
+      "rewards/check_winston_local_func": 0.7795897126197815,
+      "step": 267
+    },
+    {
+      "clip_ratio": 0.006593361962586641,
+      "epoch": 0.3748251748251748,
+      "grad_norm": 0.18497903526945206,
+      "kl": 0.7265625,
+      "learning_rate": 9.370629370629372e-05,
+      "loss": -0.0243,
+      "step": 268
+    },
+    {
+      "clip_ratio": 0.004569682292640209,
+      "completion_length": 124.39286041259766,
+      "epoch": 0.37622377622377623,
+      "grad_norm": 0.21331332062489702,
+      "kl": 0.953125,
+      "learning_rate": 9.405594405594406e-05,
+      "loss": 0.0346,
+      "num_tokens": 2746246.0,
+      "reward": 2.403602123260498,
+      "reward_std": 0.5621734857559204,
+      "rewards/check_gptzero_func": 0.7857142686843872,
+      "rewards/check_perplexity_diff_func": 0.75,
+      "rewards/check_winston_local_func": 0.8678879141807556,
+      "step": 269
+    },
+    {
+      "clip_ratio": 0.006006812676787376,
+      "epoch": 0.3776223776223776,
+      "grad_norm": 0.15420322253654764,
+      "kl": 0.8671875,
+      "learning_rate": 9.440559440559441e-05,
+      "loss": 0.0283,
+      "step": 270
+    },
+    {
+      "clip_ratio": 0.002397142816334963,
+      "completion_length": 102.41072082519531,
+      "epoch": 0.37902097902097903,
+      "grad_norm": 0.22567114655173642,
+      "kl": 0.609375,
+      "learning_rate": 9.475524475524477e-05,
+      "loss": 0.0231,
+      "num_tokens": 2765203.0,
+      "reward": 2.649076223373413,
+      "reward_std": 0.47521141171455383,
+      "rewards/check_gptzero_func": 0.875,
+      "rewards/check_perplexity_diff_func": 1.0,
+      "rewards/check_winston_local_func": 0.7740762829780579,
+      "step": 271
+    },
+    {
+      "clip_ratio": 0.007639365270733833,
+      "epoch": 0.3804195804195804,
+      "grad_norm": 0.18662930190408594,
+      "kl": 0.578125,
+      "learning_rate": 9.510489510489511e-05,
+      "loss": 0.0165,
+      "step": 272
+    },
+    {
+      "clip_ratio": 0.0010056205792352557,
+      "completion_length": 115.28572082519531,
+      "epoch": 0.38181818181818183,
+      "grad_norm": 0.14954423121719349,
+      "kl": 0.625,
+      "learning_rate": 9.545454545454546e-05,
+      "loss": 0.0237,
+      "num_tokens": 2785985.0,
+      "reward": 2.3541975021362305,
+      "reward_std": 0.5445494651794434,
+      "rewards/check_gptzero_func": 0.8392857313156128,
+      "rewards/check_perplexity_diff_func": 0.75,
+      "rewards/check_winston_local_func": 0.7649118304252625,
+      "step": 273
+    },
+    {
+      "clip_ratio": 0.003297280054539442,
+      "epoch": 0.3832167832167832,
+      "grad_norm": 0.11713806873969078,
+      "kl": 0.640625,
+      "learning_rate": 9.580419580419581e-05,
+      "loss": 0.0206,
+      "step": 274
+    },
+    {
+      "clip_ratio": 0.0016172031173482537,
+      "completion_length": 128.73214721679688,
+      "epoch": 0.38461538461538464,
+      "grad_norm": 0.155009547701029,
+      "kl": 0.51171875,
+      "learning_rate": 9.615384615384617e-05,
+      "loss": 0.0191,
+      "num_tokens": 2808414.0,
+      "reward": 2.376694679260254,
+      "reward_std": 0.33215656876564026,
+      "rewards/check_gptzero_func": 0.8392857313156128,
+      "rewards/check_perplexity_diff_func": 0.8571428656578064,
+      "rewards/check_winston_local_func": 0.6802659630775452,
+      "step": 275
+    },
+    {
+      "clip_ratio": 0.004294655751436949,
+      "epoch": 0.386013986013986,
+      "grad_norm": 0.11409067239204185,
+      "kl": 0.515625,
+      "learning_rate": 9.65034965034965e-05,
+      "loss": 0.0155,
+      "step": 276
+    },
+    {
+      "clip_ratio": 0.002925862092524767,
+      "completion_length": 116.41072082519531,
+      "epoch": 0.38741258741258744,
+      "grad_norm": 0.2764503095198425,
+      "kl": 0.578125,
+      "learning_rate": 9.685314685314686e-05,
+      "loss": 0.01,
+      "num_tokens": 2829285.0,
+      "reward": 2.3352210521698,
+      "reward_std": 0.7207943201065063,
+      "rewards/check_gptzero_func": 0.6607142686843872,
+      "rewards/check_perplexity_diff_func": 0.9642857313156128,
+      "rewards/check_winston_local_func": 0.7102211713790894,
+      "step": 277
+    },
+    {
+      "clip_ratio": 0.005557883996516466,
+      "epoch": 0.3888111888111888,
+      "grad_norm": 0.2140511595370052,
+      "kl": 0.5859375,
+      "learning_rate": 9.72027972027972e-05,
+      "loss": -0.0004,
+      "step": 278
+    },
+    {
+      "clip_ratio": 0.0029403052758425474,
+      "completion_length": 114.25000762939453,
+      "epoch": 0.3902097902097902,
+      "grad_norm": 0.21363596109825803,
+      "kl": 0.63671875,
+      "learning_rate": 9.755244755244755e-05,
+      "loss": 0.0259,
+      "num_tokens": 2849989.0,
+      "reward": 2.6570937633514404,
+      "reward_std": 0.5007120370864868,
+      "rewards/check_gptzero_func": 0.75,
+      "rewards/check_perplexity_diff_func": 1.2142857313156128,
+      "rewards/check_winston_local_func": 0.6928080320358276,
+      "step": 279
+    },
+    {
+      "clip_ratio": 0.00384224159643054,
+      "epoch": 0.3916083916083916,
+      "grad_norm": 0.15102706915256958,
+      "kl": 0.6875,
+      "learning_rate": 9.790209790209791e-05,
+      "loss": 0.0185,
+      "step": 280
+    },
+    {
+      "clip_ratio": 0.0027283646631985903,
+      "completion_length": 90.25000762939453,
+      "epoch": 0.393006993006993,
+      "grad_norm": 0.3585134615134873,
+      "kl": 0.73046875,
+      "learning_rate": 9.825174825174826e-05,
+      "loss": -0.006,
+      "num_tokens": 2867567.0,
+      "reward": 2.4528698921203613,
+      "reward_std": 0.6383817791938782,
+      "rewards/check_gptzero_func": 0.6071428656578064,
+      "rewards/check_perplexity_diff_func": 1.1428571939468384,
+      "rewards/check_winston_local_func": 0.7028695940971375,
+      "step": 281
+    },
+    {
+      "clip_ratio": 0.016295742243528366,
+      "epoch": 0.3944055944055944,
+      "grad_norm": 0.2358766583499307,
+      "kl": 0.73046875,
+      "learning_rate": 9.86013986013986e-05,
+      "loss": -0.0151,
+      "step": 282
+    },
+    {
+      "clip_ratio": 0.0025735762901604176,
+      "completion_length": 132.125,
+      "epoch": 0.3958041958041958,
+      "grad_norm": 0.12427534341566092,
+      "kl": 0.515625,
+      "learning_rate": 9.895104895104895e-05,
+      "loss": 0.006,
+      "num_tokens": 2890954.0,
+      "reward": 2.172079086303711,
+      "reward_std": 0.31105130910873413,
+      "rewards/check_gptzero_func": 0.6607142686843872,
+      "rewards/check_perplexity_diff_func": 0.7857142686843872,
+      "rewards/check_winston_local_func": 0.7256504893302917,
+      "step": 283
+    },
+    {
+      "clip_ratio": 0.008568070828914642,
+      "epoch": 0.3972027972027972,
+      "grad_norm": 0.10751149742872082,
+      "kl": 0.5234375,
+      "learning_rate": 9.930069930069931e-05,
+      "loss": 0.0038,
+      "step": 284
+    },
+    {
+      "clip_ratio": 0.0013303400482982397,
+      "completion_length": 115.50000762939453,
+      "epoch": 0.3986013986013986,
+      "grad_norm": 0.11155145645226237,
+      "kl": 0.427734375,
+      "learning_rate": 9.965034965034964e-05,
+      "loss": 0.0136,
+      "num_tokens": 2912320.0,
+      "reward": 2.499699354171753,
+      "reward_std": 0.22870054841041565,
+      "rewards/check_gptzero_func": 0.6071428656578064,
+      "rewards/check_perplexity_diff_func": 1.3214285373687744,
+      "rewards/check_winston_local_func": 0.5711276531219482,
+      "step": 285
+    },
+    {
+      "clip_ratio": 0.004690885543823242,
+      "epoch": 0.4,
+      "grad_norm": 0.08246302090980155,
+      "kl": 0.443359375,
+      "learning_rate": 0.0001,
+      "loss": 0.0117,
+      "step": 286
+    },
+    {
+      "clip_ratio": 0.0013457606546580791,
+      "completion_length": 153.60714721679688,
+      "epoch": 0.4013986013986014,
+      "grad_norm": 0.0959414180998777,
+      "kl": 0.458984375,
+      "learning_rate": 9.999996275889018e-05,
+      "loss": -0.0037,
+      "num_tokens": 2937286.0,
+      "reward": 2.60290789604187,
+      "reward_std": 0.2643657624721527,
+      "rewards/check_gptzero_func": 0.625,
+      "rewards/check_perplexity_diff_func": 1.3214285373687744,
+      "rewards/check_winston_local_func": 0.6564791798591614,
+      "step": 287
+    },
+    {
+      "clip_ratio": 0.0025680752005428076,
+      "epoch": 0.4027972027972028,
+      "grad_norm": 0.08295696905412306,
+      "kl": 0.453125,
+      "learning_rate": 9.999985103561615e-05,
+      "loss": -0.0053,
+      "step": 288
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 106.39286041259766,
+      "epoch": 0.4041958041958042,
+      "grad_norm": 0.12524695486629228,
+      "kl": 0.3671875,
+      "learning_rate": 9.999966483034437e-05,
+      "loss": -0.0031,
+      "num_tokens": 2957572.0,
+      "reward": 2.607825994491577,
+      "reward_std": 0.2113000899553299,
+      "rewards/check_gptzero_func": 0.5535714030265808,
+      "rewards/check_perplexity_diff_func": 1.4642857313156128,
+      "rewards/check_winston_local_func": 0.5899689793586731,
+      "step": 289
+    },
+    {
+      "clip_ratio": 0.0015743181575089693,
+      "epoch": 0.40559440559440557,
+      "grad_norm": 0.08958540244405325,
+      "kl": 0.36328125,
+      "learning_rate": 9.999940414335222e-05,
+      "loss": -0.0046,
+      "step": 290
+    },
+    {
+      "clip_ratio": 0.0017556955572217703,
+      "completion_length": 96.0714340209961,
+      "epoch": 0.406993006993007,
+      "grad_norm": 0.19185238369967877,
+      "kl": 2.109375,
+      "learning_rate": 9.999906897502803e-05,
+      "loss": 0.008,
+      "num_tokens": 2976076.0,
+      "reward": 2.6148557662963867,
+      "reward_std": 0.24822747707366943,
+      "rewards/check_gptzero_func": 0.5178571343421936,
+      "rewards/check_perplexity_diff_func": 1.4285714626312256,
+      "rewards/check_winston_local_func": 0.6684269309043884,
+      "step": 291
+    },
+    {
+      "clip_ratio": 0.00337741756811738,
+      "epoch": 0.4083916083916084,
+      "grad_norm": 0.08769186584743054,
+      "kl": 0.671875,
+      "learning_rate": 9.999865932587107e-05,
+      "loss": 0.0046,
+      "step": 292
+    },
+    {
+      "clip_ratio": 0.0012849332997575402,
+      "completion_length": 91.87500762939453,
+      "epoch": 0.4097902097902098,
+      "grad_norm": 0.13036265265723118,
+      "kl": 0.4453125,
+      "learning_rate": 9.999817519649158e-05,
+      "loss": 0.0127,
+      "num_tokens": 2994773.0,
+      "reward": 2.552300214767456,
+      "reward_std": 0.14500018954277039,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 1.5,
+      "rewards/check_winston_local_func": 0.6237286329269409,
+      "step": 293
+    },
+    {
+      "clip_ratio": 0.0027297178748995066,
+      "epoch": 0.4111888111888112,
+      "grad_norm": 0.0976002240930887,
+      "kl": 0.439453125,
+      "learning_rate": 9.99976165876107e-05,
+      "loss": 0.0101,
+      "step": 294
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 113.35714721679688,
+      "epoch": 0.4125874125874126,
+      "grad_norm": 0.02609480806956132,
+      "kl": 0.1591796875,
+      "learning_rate": 9.999698350006063e-05,
+      "loss": 0.0012,
+      "num_tokens": 3015993.0,
+      "reward": 2.874926805496216,
+      "reward_std": 0.0874238833785057,
+      "rewards/check_gptzero_func": 0.4821428656578064,
+      "rewards/check_perplexity_diff_func": 1.8928571939468384,
+      "rewards/check_winston_local_func": 0.4999266266822815,
+      "step": 295
+    },
+    {
+      "clip_ratio": 9.294544724980369e-05,
+      "epoch": 0.413986013986014,
+      "grad_norm": 0.02050573548257188,
+      "kl": 0.16015625,
+      "learning_rate": 9.999627593478442e-05,
+      "loss": 0.0011,
+      "step": 296
+    },
+    {
+      "clip_ratio": 0.00012664640962611884,
+      "completion_length": 98.5714340209961,
+      "epoch": 0.4153846153846154,
+      "grad_norm": 0.01680753751461254,
+      "kl": 0.201171875,
+      "learning_rate": 9.999549389283606e-05,
+      "loss": -0.0003,
+      "num_tokens": 3035155.0,
+      "reward": 3.0224545001983643,
+      "reward_std": 0.016296973451972008,
+      "rewards/check_gptzero_func": 0.4107142984867096,
+      "rewards/check_perplexity_diff_func": 1.9642857313156128,
+      "rewards/check_winston_local_func": 0.647454559803009,
+      "step": 297
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.4167832167832168,
+      "grad_norm": 0.012884367061070623,
+      "kl": 0.203125,
+      "learning_rate": 9.999463737538053e-05,
+      "loss": -0.0003,
+      "step": 298
+    },
+    {
+      "clip_ratio": 0.0003541912592481822,
+      "completion_length": 95.76786041259766,
+      "epoch": 0.41818181818181815,
+      "grad_norm": 0.04526732622569909,
+      "kl": 0.291015625,
+      "learning_rate": 9.999370638369377e-05,
+      "loss": 0.0002,
+      "num_tokens": 3054174.0,
+      "reward": 2.3303959369659424,
+      "reward_std": 0.08971218019723892,
+      "rewards/check_gptzero_func": 0.3035714328289032,
+      "rewards/check_perplexity_diff_func": 1.6071428060531616,
+      "rewards/check_winston_local_func": 0.4196813404560089,
+      "step": 299
+    },
+    {
+      "clip_ratio": 0.00011806376278400421,
+      "epoch": 0.4195804195804196,
+      "grad_norm": 0.03862986955837794,
+      "kl": 0.296875,
+      "learning_rate": 9.999270091916257e-05,
+      "loss": -0.0002,
+      "step": 300
+    },
+    {
+      "clip_ratio": 0.0012040403671562672,
+      "completion_length": 101.73214721679688,
+      "epoch": 0.42097902097902096,
+      "grad_norm": 0.12616717395678975,
+      "kl": 0.35546875,
+      "learning_rate": 9.999162098328474e-05,
+      "loss": 0.0033,
+      "num_tokens": 3073401.0,
+      "reward": 2.8153316974639893,
+      "reward_std": 0.18918544054031372,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 1.7142857313156128,
+      "rewards/check_winston_local_func": 0.6724745631217957,
+      "step": 301
+    },
+    {
+      "clip_ratio": 0.0013847867958247662,
+      "epoch": 0.4223776223776224,
+      "grad_norm": 0.06929991057149441,
+      "kl": 0.412109375,
+      "learning_rate": 9.999046657766903e-05,
+      "loss": 0.0012,
+      "step": 302
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 81.3214340209961,
+      "epoch": 0.42377622377622376,
+      "grad_norm": 0.06730614630696258,
+      "kl": 0.2099609375,
+      "learning_rate": 9.998923770403505e-05,
+      "loss": -0.0019,
+      "num_tokens": 3090625.0,
+      "reward": 2.802220106124878,
+      "reward_std": 0.03781326860189438,
+      "rewards/check_gptzero_func": 0.375,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.42722010612487793,
+      "step": 303
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.4251748251748252,
+      "grad_norm": 0.022175315902565728,
+      "kl": 0.2109375,
+      "learning_rate": 9.99879343642134e-05,
+      "loss": -0.0024,
+      "step": 304
+    },
+    {
+      "clip_ratio": 0.00013683634460903704,
+      "completion_length": 105.53572082519531,
+      "epoch": 0.42657342657342656,
+      "grad_norm": 0.04867704636152166,
+      "kl": 0.173828125,
+      "learning_rate": 9.998655656014561e-05,
+      "loss": 0.0117,
+      "num_tokens": 3110877.0,
+      "reward": 2.6083197593688965,
+      "reward_std": 0.13502858579158783,
+      "rewards/check_gptzero_func": 0.3571428656578064,
+      "rewards/check_perplexity_diff_func": 1.8928571939468384,
+      "rewards/check_winston_local_func": 0.35831958055496216,
+      "step": 305
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.427972027972028,
+      "grad_norm": 0.043987709695075865,
+      "kl": 0.1728515625,
+      "learning_rate": 9.99851042938841e-05,
+      "loss": 0.0113,
+      "step": 306
+    },
+    {
+      "clip_ratio": 0.0004155792994424701,
+      "completion_length": 124.03572082519531,
+      "epoch": 0.42937062937062936,
+      "grad_norm": 0.03684760048760383,
+      "kl": 0.2080078125,
+      "learning_rate": 9.998357756759222e-05,
+      "loss": -0.0258,
+      "num_tokens": 3134097.0,
+      "reward": 2.7439539432525635,
+      "reward_std": 0.11075940728187561,
+      "rewards/check_gptzero_func": 0.4642857015132904,
+      "rewards/check_perplexity_diff_func": 1.7857142686843872,
+      "rewards/check_winston_local_func": 0.4939536452293396,
+      "step": 307
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.4307692307692308,
+      "grad_norm": 0.013648247572273129,
+      "kl": 0.20703125,
+      "learning_rate": 9.998197638354428e-05,
+      "loss": -0.026,
+      "step": 308
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 89.12500762939453,
+      "epoch": 0.43216783216783217,
+      "grad_norm": 0.02465210474674629,
+      "kl": 0.443359375,
+      "learning_rate": 9.998030074412545e-05,
+      "loss": 0.0014,
+      "num_tokens": 3152358.0,
+      "reward": 2.420307159423828,
+      "reward_std": 0.11704181134700775,
+      "rewards/check_gptzero_func": 0.2321428507566452,
+      "rewards/check_perplexity_diff_func": 1.75,
+      "rewards/check_winston_local_func": 0.4381641745567322,
+      "step": 309
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.43356643356643354,
+      "grad_norm": 0.021184461362474653,
+      "kl": 0.48828125,
+      "learning_rate": 9.997855065183184e-05,
+      "loss": 0.0014,
+      "step": 310
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 115.73214721679688,
+      "epoch": 0.43496503496503497,
+      "grad_norm": 0.0049134080725686995,
+      "kl": 0.162109375,
+      "learning_rate": 9.997672610927047e-05,
+      "loss": -0.0047,
+      "num_tokens": 3173997.0,
+      "reward": 2.61786150932312,
+      "reward_std": 0.013112460263073444,
+      "rewards/check_gptzero_func": 0.375,
+      "rewards/check_perplexity_diff_func": 1.8214285373687744,
+      "rewards/check_winston_local_func": 0.4214327931404114,
+      "step": 311
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.43636363636363634,
+      "grad_norm": 0.00595537475288129,
+      "kl": 0.173828125,
+      "learning_rate": 9.997482711915927e-05,
+      "loss": -0.0047,
+      "step": 312
+    },
+    {
+      "clip_ratio": 0.0001360544265480712,
+      "completion_length": 90.42857360839844,
+      "epoch": 0.43776223776223777,
+      "grad_norm": 0.007409687643468917,
+      "kl": 0.2275390625,
+      "learning_rate": 9.997285368432703e-05,
+      "loss": 0.0006,
+      "num_tokens": 3192417.0,
+      "reward": 2.682650566101074,
+      "reward_std": 0.02417343109846115,
+      "rewards/check_gptzero_func": 0.375,
+      "rewards/check_perplexity_diff_func": 1.8214285373687744,
+      "rewards/check_winston_local_func": 0.4862218499183655,
+      "step": 313
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.43916083916083914,
+      "grad_norm": 0.007129921917131849,
+      "kl": 0.2275390625,
+      "learning_rate": 9.997080580771349e-05,
+      "loss": 0.0006,
+      "step": 314
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 83.14286041259766,
+      "epoch": 0.4405594405594406,
+      "grad_norm": 1.3606081670938062e-05,
+      "kl": 0.2255859375,
+      "learning_rate": 9.996868349236927e-05,
+      "loss": 0.0002,
+      "num_tokens": 3209875.0,
+      "reward": 2.671295404434204,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3571428656578064,
+      "rewards/check_perplexity_diff_func": 1.8571428060531616,
+      "rewards/check_winston_local_func": 0.45700937509536743,
+      "step": 315
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.44195804195804195,
+      "grad_norm": 1.3537787639105168e-05,
+      "kl": 0.2255859375,
+      "learning_rate": 9.996648674145583e-05,
+      "loss": 0.0002,
+      "step": 316
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 126.21429443359375,
+      "epoch": 0.4433566433566434,
+      "grad_norm": 0.00991208286904361,
+      "kl": 0.15625,
+      "learning_rate": 9.996421555824558e-05,
+      "loss": 0.0103,
+      "num_tokens": 3232963.0,
+      "reward": 2.6654393672943115,
+      "reward_std": 0.0725974440574646,
+      "rewards/check_gptzero_func": 0.5357142686843872,
+      "rewards/check_perplexity_diff_func": 1.6071428060531616,
+      "rewards/check_winston_local_func": 0.5225821733474731,
+      "step": 317
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.44475524475524475,
+      "grad_norm": 0.010677817125212348,
+      "kl": 0.154296875,
+      "learning_rate": 9.996186994612176e-05,
+      "loss": 0.0102,
+      "step": 318
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 103.4464340209961,
+      "epoch": 0.4461538461538462,
+      "grad_norm": 5.275581365835224e-05,
+      "kl": 0.16796875,
+      "learning_rate": 9.995944990857849e-05,
+      "loss": 0.0002,
+      "num_tokens": 3252898.0,
+      "reward": 3.0420854091644287,
+      "reward_std": 0.00039901022682897747,
+      "rewards/check_gptzero_func": 0.5714285969734192,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.5420854687690735,
+      "step": 319
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.44755244755244755,
+      "grad_norm": 6.10762506510971e-05,
+      "kl": 0.16796875,
+      "learning_rate": 9.995695544922077e-05,
+      "loss": 0.0002,
+      "step": 320
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 102.28572082519531,
+      "epoch": 0.4489510489510489,
+      "grad_norm": 0.010696195320132886,
+      "kl": 0.1669921875,
+      "learning_rate": 9.995438657176448e-05,
+      "loss": -0.0177,
+      "num_tokens": 3272574.0,
+      "reward": 2.7592146396636963,
+      "reward_std": 0.03917063772678375,
+      "rewards/check_gptzero_func": 0.3571428656578064,
+      "rewards/check_perplexity_diff_func": 1.9642857313156128,
+      "rewards/check_winston_local_func": 0.437785804271698,
+      "step": 321
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.45034965034965035,
+      "grad_norm": 0.01083841197384593,
+      "kl": 0.1669921875,
+      "learning_rate": 9.995174328003631e-05,
+      "loss": -0.0177,
+      "step": 322
+    },
+    {
+      "clip_ratio": 8.473140769638121e-05,
+      "completion_length": 117.41072082519531,
+      "epoch": 0.45174825174825173,
+      "grad_norm": 0.004522641361347185,
+      "kl": 0.1640625,
+      "learning_rate": 9.994902557797382e-05,
+      "loss": -0.0002,
+      "num_tokens": 3294025.0,
+      "reward": 2.92221736907959,
+      "reward_std": 0.025547320023179054,
+      "rewards/check_gptzero_func": 0.5535714030265808,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.36864587664604187,
+      "step": 323
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.45314685314685316,
+      "grad_norm": 0.004737684573095642,
+      "kl": 0.1650390625,
+      "learning_rate": 9.994623346962544e-05,
+      "loss": -0.0002,
+      "step": 324
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 95.26786041259766,
+      "epoch": 0.45454545454545453,
+      "grad_norm": 3.372088632491828e-06,
+      "kl": 0.1796875,
+      "learning_rate": 9.99433669591504e-05,
+      "loss": 0.0002,
+      "num_tokens": 3312902.0,
+      "reward": 2.595499038696289,
+      "reward_std": 4.388691013446078e-05,
+      "rewards/check_gptzero_func": 0.25,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.34549909830093384,
+      "step": 325
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.45594405594405596,
+      "grad_norm": 3.349405120445447e-06,
+      "kl": 0.1796875,
+      "learning_rate": 9.994042605081879e-05,
+      "loss": 0.0002,
+      "step": 326
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 97.80357360839844,
+      "epoch": 0.45734265734265733,
+      "grad_norm": 0.010175912326793437,
+      "kl": 0.177734375,
+      "learning_rate": 9.993741074901153e-05,
+      "loss": 0.0003,
+      "num_tokens": 3331815.0,
+      "reward": 2.496483325958252,
+      "reward_std": 0.04960284009575844,
+      "rewards/check_gptzero_func": 0.2142857164144516,
+      "rewards/check_perplexity_diff_func": 1.9642857313156128,
+      "rewards/check_winston_local_func": 0.31791168451309204,
+      "step": 327
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.45874125874125876,
+      "grad_norm": 0.01017493586702021,
+      "kl": 0.177734375,
+      "learning_rate": 9.993432105822034e-05,
+      "loss": 0.0003,
+      "step": 328
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 96.30357360839844,
+      "epoch": 0.46013986013986014,
+      "grad_norm": 0.0165654292436505,
+      "kl": 0.1572265625,
+      "learning_rate": 9.993115698304774e-05,
+      "loss": 0.0003,
+      "num_tokens": 3351210.0,
+      "reward": 2.8074374198913574,
+      "reward_std": 0.07505974173545837,
+      "rewards/check_gptzero_func": 0.5178571343421936,
+      "rewards/check_perplexity_diff_func": 1.8571428060531616,
+      "rewards/check_winston_local_func": 0.4324372708797455,
+      "step": 329
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.46153846153846156,
+      "grad_norm": 0.014211382809754432,
+      "kl": 0.16015625,
+      "learning_rate": 9.992791852820709e-05,
+      "loss": 0.0003,
+      "step": 330
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 86.1964340209961,
+      "epoch": 0.46293706293706294,
+      "grad_norm": 0.0580909978469744,
+      "kl": 0.2119140625,
+      "learning_rate": 9.992460569852256e-05,
+      "loss": 0.0013,
+      "num_tokens": 3369025.0,
+      "reward": 2.986760377883911,
+      "reward_std": 0.10425443947315216,
+      "rewards/check_gptzero_func": 0.4642857015132904,
+      "rewards/check_perplexity_diff_func": 1.8571428060531616,
+      "rewards/check_winston_local_func": 0.6653318405151367,
+      "step": 331
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.4643356643356643,
+      "grad_norm": 0.0012843504186470114,
+      "kl": 0.2109375,
+      "learning_rate": 9.992121849892904e-05,
+      "loss": 0.0012,
+      "step": 332
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 113.42857360839844,
+      "epoch": 0.46573426573426574,
+      "grad_norm": 0.03054414776989089,
+      "kl": 0.1494140625,
+      "learning_rate": 9.99177569344723e-05,
+      "loss": -0.0,
+      "num_tokens": 3390093.0,
+      "reward": 2.5890934467315674,
+      "reward_std": 0.1023683100938797,
+      "rewards/check_gptzero_func": 0.25,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.41052165627479553,
+      "step": 333
+    },
+    {
+      "clip_ratio": 0.00022784233442507684,
+      "epoch": 0.4671328671328671,
+      "grad_norm": 0.006141110067038759,
+      "kl": 0.150390625,
+      "learning_rate": 9.99142210103088e-05,
+      "loss": -0.0001,
+      "step": 334
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 109.00000762939453,
+      "epoch": 0.46853146853146854,
+      "grad_norm": 0.011702000814979486,
+      "kl": 0.158203125,
+      "learning_rate": 9.991061073170585e-05,
+      "loss": 0.0001,
+      "num_tokens": 3410511.0,
+      "reward": 2.601181745529175,
+      "reward_std": 0.04859553650021553,
+      "rewards/check_gptzero_func": 0.2857142984867096,
+      "rewards/check_perplexity_diff_func": 1.8928571939468384,
+      "rewards/check_winston_local_func": 0.42261001467704773,
+      "step": 335
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.4699300699300699,
+      "grad_norm": 0.004380253375033637,
+      "kl": 0.158203125,
+      "learning_rate": 9.990692610404145e-05,
+      "loss": 0.0001,
+      "step": 336
+    },
+    {
+      "clip_ratio": 0.00015908369095996022,
+      "completion_length": 124.37500762939453,
+      "epoch": 0.47132867132867134,
+      "grad_norm": 0.011635533606345248,
+      "kl": 0.1318359375,
+      "learning_rate": 9.99031671328044e-05,
+      "loss": 0.0002,
+      "num_tokens": 3433250.0,
+      "reward": 2.719287395477295,
+      "reward_std": 0.05448899790644646,
+      "rewards/check_gptzero_func": 0.5,
+      "rewards/check_perplexity_diff_func": 1.75,
+      "rewards/check_winston_local_func": 0.4692873954772949,
+      "step": 337
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.4727272727272727,
+      "grad_norm": 0.009393665715625162,
+      "kl": 0.1318359375,
+      "learning_rate": 9.989933382359422e-05,
+      "loss": 0.0001,
+      "step": 338
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 110.8214340209961,
+      "epoch": 0.47412587412587415,
+      "grad_norm": 6.954019260271828e-07,
+      "kl": 0.15625,
+      "learning_rate": 9.98954261821212e-05,
+      "loss": 0.0002,
+      "num_tokens": 3454202.0,
+      "reward": 2.5842461585998535,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 1.7857142686843872,
+      "rewards/check_winston_local_func": 0.40567442774772644,
+      "step": 339
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.4755244755244755,
+      "grad_norm": 6.417345751201852e-07,
+      "kl": 0.15625,
+      "learning_rate": 9.98914442142063e-05,
+      "loss": 0.0002,
+      "step": 340
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 76.55357360839844,
+      "epoch": 0.47692307692307695,
+      "grad_norm": 0.008322229771779063,
+      "kl": 0.263671875,
+      "learning_rate": 9.988738792578126e-05,
+      "loss": 0.0005,
+      "num_tokens": 3471041.0,
+      "reward": 2.916919231414795,
+      "reward_std": 0.05185602605342865,
+      "rewards/check_gptzero_func": 0.3214285671710968,
+      "rewards/check_perplexity_diff_func": 1.9642857313156128,
+      "rewards/check_winston_local_func": 0.631205141544342,
+      "step": 341
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.4783216783216783,
+      "grad_norm": 0.010116392524351476,
+      "kl": 0.263671875,
+      "learning_rate": 9.988325732288851e-05,
+      "loss": 0.0004,
+      "step": 342
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 122.21429443359375,
+      "epoch": 0.4797202797202797,
+      "grad_norm": 1.4137707959903356e-05,
+      "kl": 0.134765625,
+      "learning_rate": 9.987905241168117e-05,
+      "loss": 0.0001,
+      "num_tokens": 3493483.0,
+      "reward": 2.7157416343688965,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 1.8571428060531616,
+      "rewards/check_winston_local_func": 0.46574172377586365,
+      "step": 343
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.4811188811188811,
+      "grad_norm": 1.3729789968238485e-05,
+      "kl": 0.134765625,
+      "learning_rate": 9.987477319842307e-05,
+      "loss": 0.0001,
+      "step": 344
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 93.53572082519531,
+      "epoch": 0.4825174825174825,
+      "grad_norm": 7.468121696306164e-07,
+      "kl": 0.1845703125,
+      "learning_rate": 9.987041968948869e-05,
+      "loss": 0.0002,
+      "num_tokens": 3512009.0,
+      "reward": 2.5505218505859375,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.1785714328289032,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.44337886571884155,
+      "step": 345
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.48391608391608393,
+      "grad_norm": 7.137709602226676e-07,
+      "kl": 0.1845703125,
+      "learning_rate": 9.986599189136325e-05,
+      "loss": 0.0002,
+      "step": 346
+    },
+    {
+      "clip_ratio": 0.00031959093757905066,
+      "completion_length": 102.58928680419922,
+      "epoch": 0.4853146853146853,
+      "grad_norm": 0.04707770778424934,
+      "kl": 0.1982421875,
+      "learning_rate": 9.986148981064258e-05,
+      "loss": 0.0004,
+      "num_tokens": 3531574.0,
+      "reward": 2.7341363430023193,
+      "reward_std": 0.0231277234852314,
+      "rewards/check_gptzero_func": 0.4107142984867096,
+      "rewards/check_perplexity_diff_func": 1.8928571939468384,
+      "rewards/check_winston_local_func": 0.430564820766449,
+      "step": 347
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.48671328671328673,
+      "grad_norm": 0.020919799034125892,
+      "kl": 0.2060546875,
+      "learning_rate": 9.985691345403316e-05,
+      "loss": 0.0002,
+      "step": 348
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 127.67857360839844,
+      "epoch": 0.4881118881118881,
+      "grad_norm": 6.517786151476589e-05,
+      "kl": 0.1328125,
+      "learning_rate": 9.985226282835216e-05,
+      "loss": 0.0001,
+      "num_tokens": 3554440.0,
+      "reward": 2.926541566848755,
+      "reward_std": 0.0005768488626927137,
+      "rewards/check_gptzero_func": 0.5357142686843872,
+      "rewards/check_perplexity_diff_func": 1.8571428060531616,
+      "rewards/check_winston_local_func": 0.5336844325065613,
+      "step": 349
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.48951048951048953,
+      "grad_norm": 6.370254129427888e-05,
+      "kl": 0.1328125,
+      "learning_rate": 9.984753794052735e-05,
+      "loss": 0.0001,
+      "step": 350
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 98.75000762939453,
+      "epoch": 0.4909090909090909,
+      "grad_norm": 2.085540464425244e-05,
+      "kl": 0.205078125,
+      "learning_rate": 9.984273879759713e-05,
+      "loss": 0.0002,
+      "num_tokens": 3573976.0,
+      "reward": 2.8204286098480225,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.4989997446537018,
+      "step": 351
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.49230769230769234,
+      "grad_norm": 2.009578048518353e-05,
+      "kl": 0.205078125,
+      "learning_rate": 9.983786540671051e-05,
+      "loss": 0.0002,
+      "step": 352
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 85.21428680419922,
+      "epoch": 0.4937062937062937,
+      "grad_norm": 2.1499404490472988e-06,
+      "kl": 0.1826171875,
+      "learning_rate": 9.983291777512711e-05,
+      "loss": 0.0002,
+      "num_tokens": 3591644.0,
+      "reward": 2.8764142990112305,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.44784265756607056,
+      "step": 353
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.4951048951048951,
+      "grad_norm": 2.160124410750654e-06,
+      "kl": 0.1826171875,
+      "learning_rate": 9.982789591021715e-05,
+      "loss": 0.0002,
+      "step": 354
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 112.00000762939453,
+      "epoch": 0.4965034965034965,
+      "grad_norm": 0.00014814082463704463,
+      "kl": 0.1669921875,
+      "learning_rate": 9.982279981946143e-05,
+      "loss": 0.0002,
+      "num_tokens": 3612596.0,
+      "reward": 2.8398191928863525,
+      "reward_std": 0.0006443771999329329,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.5183902382850647,
+      "step": 355
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.4979020979020979,
+      "grad_norm": 0.00014606763289714295,
+      "kl": 0.1669921875,
+      "learning_rate": 9.98176295104513e-05,
+      "loss": 0.0002,
+      "step": 356
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 88.50000762939453,
+      "epoch": 0.4993006993006993,
+      "grad_norm": 3.33550016470454e-07,
+      "kl": 0.1669921875,
+      "learning_rate": 9.98123849908887e-05,
+      "loss": 0.0002,
+      "num_tokens": 3630862.0,
+      "reward": 2.655492067337036,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3214285671710968,
+      "rewards/check_perplexity_diff_func": 1.8571428060531616,
+      "rewards/check_winston_local_func": 0.47692039608955383,
+      "step": 357
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5006993006993007,
+      "grad_norm": 3.3462192763241657e-07,
+      "kl": 0.1669921875,
+      "learning_rate": 9.980706626858607e-05,
+      "loss": 0.0002,
+      "step": 358
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 126.39286041259766,
+      "epoch": 0.5020979020979021,
+      "grad_norm": 3.4860786526471938e-06,
+      "kl": 0.1416015625,
+      "learning_rate": 9.980167335146642e-05,
+      "loss": 0.0001,
+      "num_tokens": 3653738.0,
+      "reward": 2.765826463699341,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.5,
+      "rewards/check_perplexity_diff_func": 1.8571428060531616,
+      "rewards/check_winston_local_func": 0.4086834490299225,
+      "step": 359
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5034965034965035,
+      "grad_norm": 3.71619859471226e-06,
+      "kl": 0.1416015625,
+      "learning_rate": 9.979620624756329e-05,
+      "loss": 0.0001,
+      "step": 360
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 129.92857360839844,
+      "epoch": 0.5048951048951049,
+      "grad_norm": 8.195642286342138e-07,
+      "kl": 0.1376953125,
+      "learning_rate": 9.979066496502074e-05,
+      "loss": 0.0001,
+      "num_tokens": 3677274.0,
+      "reward": 2.8696835041046143,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.5357142686843872,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.4053979218006134,
+      "step": 361
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5062937062937063,
+      "grad_norm": 8.760444459307852e-07,
+      "kl": 0.1376953125,
+      "learning_rate": 9.978504951209327e-05,
+      "loss": 0.0001,
+      "step": 362
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 118.0714340209961,
+      "epoch": 0.5076923076923077,
+      "grad_norm": 1.595860947799347e-06,
+      "kl": 0.162109375,
+      "learning_rate": 9.977935989714595e-05,
+      "loss": 0.0002,
+      "num_tokens": 3698896.0,
+      "reward": 2.6664552688598633,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.2857142984867096,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.45216941833496094,
+      "step": 363
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.509090909090909,
+      "grad_norm": 1.6260194368178069e-06,
+      "kl": 0.162109375,
+      "learning_rate": 9.977359612865423e-05,
+      "loss": 0.0002,
+      "step": 364
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 85.96428680419922,
+      "epoch": 0.5104895104895105,
+      "grad_norm": 4.357301744161908e-06,
+      "kl": 0.2138671875,
+      "learning_rate": 9.976775821520412e-05,
+      "loss": 0.0002,
+      "num_tokens": 3716536.0,
+      "reward": 2.8877956867218018,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.566366970539093,
+      "step": 365
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5118881118881119,
+      "grad_norm": 4.353902612581284e-06,
+      "kl": 0.2138671875,
+      "learning_rate": 9.976184616549203e-05,
+      "loss": 0.0002,
+      "step": 366
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 116.96429443359375,
+      "epoch": 0.5132867132867133,
+      "grad_norm": 1.5148419096170459e-06,
+      "kl": 0.146484375,
+      "learning_rate": 9.97558599883248e-05,
+      "loss": 0.0001,
+      "num_tokens": 3737950.0,
+      "reward": 2.687312602996826,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.36588403582572937,
+      "step": 367
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5146853146853146,
+      "grad_norm": 1.4470246412763422e-06,
+      "kl": 0.146484375,
+      "learning_rate": 9.97497996926197e-05,
+      "loss": 0.0001,
+      "step": 368
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 94.83928680419922,
+      "epoch": 0.5160839160839161,
+      "grad_norm": 0.0004647591690002515,
+      "kl": 0.2138671875,
+      "learning_rate": 9.974366528740441e-05,
+      "loss": 0.0002,
+      "num_tokens": 3756935.0,
+      "reward": 2.634523391723633,
+      "reward_std": 0.0005499019753187895,
+      "rewards/check_gptzero_func": 0.2857142984867096,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.4202377200126648,
+      "step": 369
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5174825174825175,
+      "grad_norm": 0.0004885464229234562,
+      "kl": 0.2138671875,
+      "learning_rate": 9.973745678181705e-05,
+      "loss": 0.0002,
+      "step": 370
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 100.17857360839844,
+      "epoch": 0.5188811188811189,
+      "grad_norm": 3.0418176986251483e-06,
+      "kl": 0.1640625,
+      "learning_rate": 9.973117418510605e-05,
+      "loss": 0.0002,
+      "num_tokens": 3776087.0,
+      "reward": 2.5216457843780518,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.2857142984867096,
+      "rewards/check_perplexity_diff_func": 1.8571428060531616,
+      "rewards/check_winston_local_func": 0.3787885308265686,
+      "step": 371
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5202797202797202,
+      "grad_norm": 3.0723635480561103e-06,
+      "kl": 0.1640625,
+      "learning_rate": 9.972481750663026e-05,
+      "loss": 0.0002,
+      "step": 372
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 89.66072082519531,
+      "epoch": 0.5216783216783217,
+      "grad_norm": 0.008590656277289149,
+      "kl": 0.40234375,
+      "learning_rate": 9.971838675585888e-05,
+      "loss": 0.0007,
+      "num_tokens": 3794550.0,
+      "reward": 2.5267839431762695,
+      "reward_std": 0.05061079189181328,
+      "rewards/check_gptzero_func": 0.3214285671710968,
+      "rewards/check_perplexity_diff_func": 1.8214285373687744,
+      "rewards/check_winston_local_func": 0.383926659822464,
+      "step": 373
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5230769230769231,
+      "grad_norm": 0.003770576536622864,
+      "kl": 0.4296875,
+      "learning_rate": 9.97118819423714e-05,
+      "loss": 0.0007,
+      "step": 374
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 80.17857360839844,
+      "epoch": 0.5244755244755245,
+      "grad_norm": 3.560813902520289e-07,
+      "kl": 0.2216796875,
+      "learning_rate": 9.970530307585773e-05,
+      "loss": 0.0002,
+      "num_tokens": 3811756.0,
+      "reward": 2.883481740951538,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.5263389348983765,
+      "step": 375
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5258741258741259,
+      "grad_norm": 3.5463275242856317e-07,
+      "kl": 0.2216796875,
+      "learning_rate": 9.9698650166118e-05,
+      "loss": 0.0002,
+      "step": 376
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 86.75000762939453,
+      "epoch": 0.5272727272727272,
+      "grad_norm": 4.776682547251814e-07,
+      "kl": 0.205078125,
+      "learning_rate": 9.969192322306271e-05,
+      "loss": 0.0002,
+      "num_tokens": 3829616.0,
+      "reward": 2.9529268741607666,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4642857015132904,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.5600695610046387,
+      "step": 377
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5286713286713287,
+      "grad_norm": 4.804658823710114e-07,
+      "kl": 0.205078125,
+      "learning_rate": 9.96851222567126e-05,
+      "loss": 0.0002,
+      "step": 378
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 117.89286041259766,
+      "epoch": 0.5300699300699301,
+      "grad_norm": 9.143102926453055e-06,
+      "kl": 0.158203125,
+      "learning_rate": 9.96782472771987e-05,
+      "loss": 0.0002,
+      "num_tokens": 3851170.0,
+      "reward": 2.7617037296295166,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.40456071496009827,
+      "step": 379
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5314685314685315,
+      "grad_norm": 9.41817239216292e-06,
+      "kl": 0.158203125,
+      "learning_rate": 9.967129829476228e-05,
+      "loss": 0.0002,
+      "step": 380
+    },
+    {
+      "clip_ratio": 0.000286861730273813,
+      "completion_length": 80.8214340209961,
+      "epoch": 0.5328671328671328,
+      "grad_norm": 0.03337399461775748,
+      "kl": 0.224609375,
+      "learning_rate": 9.966427531975483e-05,
+      "loss": -0.0402,
+      "num_tokens": 3868508.0,
+      "reward": 2.820319652557373,
+      "reward_std": 0.0946396142244339,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 1.8571428060531616,
+      "rewards/check_winston_local_func": 0.5346053838729858,
+      "step": 381
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5342657342657343,
+      "grad_norm": 0.00034351439093039966,
+      "kl": 0.224609375,
+      "learning_rate": 9.965717836263812e-05,
+      "loss": -0.0403,
+      "step": 382
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 95.03572082519531,
+      "epoch": 0.5356643356643357,
+      "grad_norm": 3.3945579906577137e-07,
+      "kl": 0.216796875,
+      "learning_rate": 9.965000743398408e-05,
+      "loss": 0.0002,
+      "num_tokens": 3887044.0,
+      "reward": 2.8529021739959717,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.4243304431438446,
+      "step": 383
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5370629370629371,
+      "grad_norm": 2.8390305863820515e-07,
+      "kl": 0.216796875,
+      "learning_rate": 9.964276254447484e-05,
+      "loss": 0.0002,
+      "step": 384
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 140.1428680419922,
+      "epoch": 0.5384615384615384,
+      "grad_norm": 8.521451122880038e-07,
+      "kl": 0.134765625,
+      "learning_rate": 9.96354437049027e-05,
+      "loss": 0.0001,
+      "num_tokens": 3911264.0,
+      "reward": 2.4770619869232178,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.2857142984867096,
+      "rewards/check_perplexity_diff_func": 1.8571428060531616,
+      "rewards/check_winston_local_func": 0.33420488238334656,
+      "step": 385
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5398601398601398,
+      "grad_norm": 6.753306754213034e-07,
+      "kl": 0.134765625,
+      "learning_rate": 9.962805092617016e-05,
+      "loss": 0.0001,
+      "step": 386
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 97.64286041259766,
+      "epoch": 0.5412587412587413,
+      "grad_norm": 0.08784199050747857,
+      "kl": 2.28125,
+      "learning_rate": 9.962058421928979e-05,
+      "loss": 0.0023,
+      "num_tokens": 3930478.0,
+      "reward": 2.897810459136963,
+      "reward_std": 0.002231778111308813,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.4692386984825134,
+      "step": 387
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5426573426573427,
+      "grad_norm": 0.03620083942117924,
+      "kl": 0.98828125,
+      "learning_rate": 9.961304359538437e-05,
+      "loss": 0.001,
+      "step": 388
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 102.21428680419922,
+      "epoch": 0.544055944055944,
+      "grad_norm": 0.025447754609912356,
+      "kl": 0.1806640625,
+      "learning_rate": 9.96054290656867e-05,
+      "loss": 0.0002,
+      "num_tokens": 3950362.0,
+      "reward": 2.8110532760620117,
+      "reward_std": 0.04895726963877678,
+      "rewards/check_gptzero_func": 0.5357142686843872,
+      "rewards/check_perplexity_diff_func": 1.8214285373687744,
+      "rewards/check_winston_local_func": 0.45391014218330383,
+      "step": 389
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5454545454545454,
+      "grad_norm": 5.1964811610508515e-05,
+      "kl": 0.1806640625,
+      "learning_rate": 9.959774064153977e-05,
+      "loss": 0.0001,
+      "step": 390
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 103.5714340209961,
+      "epoch": 0.5468531468531469,
+      "grad_norm": 3.816406536402076e-07,
+      "kl": 0.173828125,
+      "learning_rate": 9.958997833439657e-05,
+      "loss": 0.0002,
+      "num_tokens": 3970164.0,
+      "reward": 2.5441431999206543,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3214285671710968,
+      "rewards/check_perplexity_diff_func": 1.8571428060531616,
+      "rewards/check_winston_local_func": 0.3655718266963959,
+      "step": 391
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5482517482517483,
+      "grad_norm": 4.199010025491445e-07,
+      "kl": 0.173828125,
+      "learning_rate": 9.958214215582018e-05,
+      "loss": 0.0002,
+      "step": 392
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 126.8214340209961,
+      "epoch": 0.5496503496503496,
+      "grad_norm": 6.309179959293235e-06,
+      "kl": 0.1533203125,
+      "learning_rate": 9.957423211748374e-05,
+      "loss": 0.0002,
+      "num_tokens": 3992994.0,
+      "reward": 2.7432734966278076,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.42184481024742126,
+      "step": 393
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.551048951048951,
+      "grad_norm": 5.547540566387994e-06,
+      "kl": 0.1533203125,
+      "learning_rate": 9.956624823117036e-05,
+      "loss": 0.0002,
+      "step": 394
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 125.62500762939453,
+      "epoch": 0.5524475524475524,
+      "grad_norm": 0.007987942179452279,
+      "kl": 0.142578125,
+      "learning_rate": 9.955819050877321e-05,
+      "loss": 0.0001,
+      "num_tokens": 4015669.0,
+      "reward": 3.091567039489746,
+      "reward_std": 0.018998777493834496,
+      "rewards/check_gptzero_func": 0.5535714030265808,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.6094242334365845,
+      "step": 395
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5538461538461539,
+      "grad_norm": 0.0034758785486018436,
+      "kl": 0.142578125,
+      "learning_rate": 9.955005896229543e-05,
+      "loss": 0.0,
+      "step": 396
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 96.75000762939453,
+      "epoch": 0.5552447552447553,
+      "grad_norm": 0.011495641299277456,
+      "kl": 0.283203125,
+      "learning_rate": 9.954185360385013e-05,
+      "loss": 0.0006,
+      "num_tokens": 4035041.0,
+      "reward": 2.950004816055298,
+      "reward_std": 0.01496803853660822,
+      "rewards/check_gptzero_func": 0.4821428656578064,
+      "rewards/check_perplexity_diff_func": 1.8928571939468384,
+      "rewards/check_winston_local_func": 0.5750047564506531,
+      "step": 397
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5566433566433566,
+      "grad_norm": 0.010082500068390739,
+      "kl": 0.283203125,
+      "learning_rate": 9.953357444566039e-05,
+      "loss": 0.0006,
+      "step": 398
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 110.8214340209961,
+      "epoch": 0.558041958041958,
+      "grad_norm": 8.489376613416e-06,
+      "kl": 0.1787109375,
+      "learning_rate": 9.952522150005919e-05,
+      "loss": 0.0002,
+      "num_tokens": 4055855.0,
+      "reward": 2.913240909576416,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4642857015132904,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.5203836560249329,
+      "step": 399
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5594405594405595,
+      "grad_norm": 8.766485288799225e-06,
+      "kl": 0.1787109375,
+      "learning_rate": 9.951679477948947e-05,
+      "loss": 0.0002,
+      "step": 400
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 105.46428680419922,
+      "epoch": 0.5608391608391609,
+      "grad_norm": 8.652849000461684e-06,
+      "kl": 0.150390625,
+      "learning_rate": 9.9508294296504e-05,
+      "loss": 0.0002,
+      "num_tokens": 4076051.0,
+      "reward": 2.871354818344116,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.44278329610824585,
+      "step": 401
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5622377622377622,
+      "grad_norm": 8.18246821524729e-06,
+      "kl": 0.150390625,
+      "learning_rate": 9.949972006376556e-05,
+      "loss": 0.0002,
+      "step": 402
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 75.25,
+      "epoch": 0.5636363636363636,
+      "grad_norm": 7.351342876079361e-07,
+      "kl": 0.1953125,
+      "learning_rate": 9.949107209404665e-05,
+      "loss": 0.0002,
+      "num_tokens": 4092527.0,
+      "reward": 2.9322519302368164,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3571428656578064,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.5751089453697205,
+      "step": 403
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5650349650349651,
+      "grad_norm": 7.136344464618312e-07,
+      "kl": 0.1953125,
+      "learning_rate": 9.948235040022966e-05,
+      "loss": 0.0002,
+      "step": 404
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 108.08928680419922,
+      "epoch": 0.5664335664335665,
+      "grad_norm": 0.007956878335048736,
+      "kl": 0.169921875,
+      "learning_rate": 9.947355499530683e-05,
+      "loss": 0.0049,
+      "num_tokens": 4113282.0,
+      "reward": 3.1798436641693115,
+      "reward_std": 0.06495096534490585,
+      "rewards/check_gptzero_func": 0.6785714030265808,
+      "rewards/check_perplexity_diff_func": 1.9642857313156128,
+      "rewards/check_winston_local_func": 0.5369864702224731,
+      "step": 405
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5678321678321678,
+      "grad_norm": 0.006640329404323574,
+      "kl": 0.1728515625,
+      "learning_rate": 9.946468589238021e-05,
+      "loss": 0.0049,
+      "step": 406
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 121.39286041259766,
+      "epoch": 0.5692307692307692,
+      "grad_norm": 1.5620797981026132e-06,
+      "kl": 0.1396484375,
+      "learning_rate": 9.945574310466159e-05,
+      "loss": 0.0001,
+      "num_tokens": 4135506.0,
+      "reward": 2.573127269744873,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.2857142984867096,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.3588416278362274,
+      "step": 407
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5706293706293706,
+      "grad_norm": 1.51593967204227e-06,
+      "kl": 0.1396484375,
+      "learning_rate": 9.944672664547252e-05,
+      "loss": 0.0001,
+      "step": 408
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 107.8214340209961,
+      "epoch": 0.5720279720279721,
+      "grad_norm": 0.0007775272826913423,
+      "kl": 0.17578125,
+      "learning_rate": 9.943763652824436e-05,
+      "loss": 0.0002,
+      "num_tokens": 4156060.0,
+      "reward": 2.55350661277771,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3571428656578064,
+      "rewards/check_perplexity_diff_func": 1.7142857313156128,
+      "rewards/check_winston_local_func": 0.48207801580429077,
+      "step": 409
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5734265734265734,
+      "grad_norm": 0.0008914625567854122,
+      "kl": 0.1767578125,
+      "learning_rate": 9.942847276651811e-05,
+      "loss": 0.0002,
+      "step": 410
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 101.62500762939453,
+      "epoch": 0.5748251748251748,
+      "grad_norm": 4.9539161428254635e-05,
+      "kl": 0.1474609375,
+      "learning_rate": 9.941923537394456e-05,
+      "loss": 0.0001,
+      "num_tokens": 4176019.0,
+      "reward": 2.8906424045562744,
+      "reward_std": 0.00012528452498372644,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.5334992408752441,
+      "step": 411
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5762237762237762,
+      "grad_norm": 5.844422779033556e-05,
+      "kl": 0.1474609375,
+      "learning_rate": 9.940992436428409e-05,
+      "loss": 0.0001,
+      "step": 412
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 105.28572082519531,
+      "epoch": 0.5776223776223777,
+      "grad_norm": 0.0006008878616346907,
+      "kl": 0.2216796875,
+      "learning_rate": 9.940053975140684e-05,
+      "loss": 0.0002,
+      "num_tokens": 4196667.0,
+      "reward": 2.943253755569458,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.5357142686843872,
+      "rewards/check_perplexity_diff_func": 1.8571428060531616,
+      "rewards/check_winston_local_func": 0.5503963828086853,
+      "step": 413
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.579020979020979,
+      "grad_norm": 0.0003365492175069851,
+      "kl": 0.220703125,
+      "learning_rate": 9.939108154929252e-05,
+      "loss": 0.0002,
+      "step": 414
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 114.60714721679688,
+      "epoch": 0.5804195804195804,
+      "grad_norm": 0.008680207230787525,
+      "kl": 0.171875,
+      "learning_rate": 9.938154977203049e-05,
+      "loss": -0.0005,
+      "num_tokens": 4217727.0,
+      "reward": 2.7773444652557373,
+      "reward_std": 0.051937274634838104,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 1.8214285373687744,
+      "rewards/check_winston_local_func": 0.5273441076278687,
+      "step": 415
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5818181818181818,
+      "grad_norm": 0.00033864647249595475,
+      "kl": 0.171875,
+      "learning_rate": 9.937194443381972e-05,
+      "loss": -0.0005,
+      "step": 416
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 100.60714721679688,
+      "epoch": 0.5832167832167832,
+      "grad_norm": 4.017405087628847e-06,
+      "kl": 0.193359375,
+      "learning_rate": 9.936226554896875e-05,
+      "loss": 0.0002,
+      "num_tokens": 4237189.0,
+      "reward": 2.7684807777404785,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.44705215096473694,
+      "step": 417
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5846153846153846,
+      "grad_norm": 3.880139616189675e-06,
+      "kl": 0.193359375,
+      "learning_rate": 9.935251313189564e-05,
+      "loss": 0.0002,
+      "step": 418
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 107.75000762939453,
+      "epoch": 0.586013986013986,
+      "grad_norm": 0.00032680437267053027,
+      "kl": 0.1826171875,
+      "learning_rate": 9.934268719712807e-05,
+      "loss": 0.0002,
+      "num_tokens": 4258213.0,
+      "reward": 2.577641487121582,
+      "reward_std": 0.0015505586052313447,
+      "rewards/check_gptzero_func": 0.2857142984867096,
+      "rewards/check_perplexity_diff_func": 1.8571428060531616,
+      "rewards/check_winston_local_func": 0.4347843825817108,
+      "step": 419
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5874125874125874,
+      "grad_norm": 0.0003238465648274268,
+      "kl": 0.1826171875,
+      "learning_rate": 9.933278775930317e-05,
+      "loss": 0.0002,
+      "step": 420
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 108.67857360839844,
+      "epoch": 0.5888111888111888,
+      "grad_norm": 0.015591092439371165,
+      "kl": 0.16015625,
+      "learning_rate": 9.932281483316758e-05,
+      "loss": 0.0004,
+      "num_tokens": 4278847.0,
+      "reward": 2.615447998046875,
+      "reward_std": 0.05428782477974892,
+      "rewards/check_gptzero_func": 0.3214285671710968,
+      "rewards/check_perplexity_diff_func": 1.8214285373687744,
+      "rewards/check_winston_local_func": 0.47259071469306946,
+      "step": 421
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5902097902097903,
+      "grad_norm": 0.00016911180587449477,
+      "kl": 0.16015625,
+      "learning_rate": 9.931276843357742e-05,
+      "loss": 0.0003,
+      "step": 422
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 111.16072082519531,
+      "epoch": 0.5916083916083916,
+      "grad_norm": 1.6847767080478394e-05,
+      "kl": 0.1630859375,
+      "learning_rate": 9.930264857549825e-05,
+      "loss": 0.0002,
+      "num_tokens": 4300096.0,
+      "reward": 2.857067823410034,
+      "reward_std": 0.00101040443405509,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 1.8571428060531616,
+      "rewards/check_winston_local_func": 0.6070676445960999,
+      "step": 423
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.593006993006993,
+      "grad_norm": 2.453025078669727e-05,
+      "kl": 0.1630859375,
+      "learning_rate": 9.929245527400503e-05,
+      "loss": 0.0002,
+      "step": 424
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 102.50000762939453,
+      "epoch": 0.5944055944055944,
+      "grad_norm": 0.0015183266942611472,
+      "kl": 0.173828125,
+      "learning_rate": 9.928218854428221e-05,
+      "loss": 0.0002,
+      "num_tokens": 4320064.0,
+      "reward": 2.6430578231811523,
+      "reward_std": 5.020291791879572e-05,
+      "rewards/check_gptzero_func": 0.3571428656578064,
+      "rewards/check_perplexity_diff_func": 1.7857142686843872,
+      "rewards/check_winston_local_func": 0.5002006888389587,
+      "step": 425
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5958041958041959,
+      "grad_norm": 0.0013596061771866538,
+      "kl": 0.1708984375,
+      "learning_rate": 9.927184840162354e-05,
+      "loss": 0.0002,
+      "step": 426
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 110.66072082519531,
+      "epoch": 0.5972027972027972,
+      "grad_norm": 0.12076121626694686,
+      "kl": 0.189453125,
+      "learning_rate": 9.926143486143214e-05,
+      "loss": -0.001,
+      "num_tokens": 4341145.0,
+      "reward": 2.6579177379608154,
+      "reward_std": 0.05154726281762123,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 1.8214285373687744,
+      "rewards/check_winston_local_func": 0.4079175889492035,
+      "step": 427
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.5986013986013986,
+      "grad_norm": 0.0008289218843548021,
+      "kl": 0.19140625,
+      "learning_rate": 9.92509479392205e-05,
+      "loss": -0.0013,
+      "step": 428
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 89.67857360839844,
+      "epoch": 0.6,
+      "grad_norm": 0.005379412774279031,
+      "kl": 0.1787109375,
+      "learning_rate": 9.924038765061042e-05,
+      "loss": 0.0001,
+      "num_tokens": 4359277.0,
+      "reward": 2.633777141571045,
+      "reward_std": 0.05039219558238983,
+      "rewards/check_gptzero_func": 0.2857142984867096,
+      "rewards/check_perplexity_diff_func": 1.9642857313156128,
+      "rewards/check_winston_local_func": 0.38377735018730164,
+      "step": 429
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6013986013986014,
+      "grad_norm": 0.010439668945298157,
+      "kl": 0.1787109375,
+      "learning_rate": 9.922975401133293e-05,
+      "loss": 0.0002,
+      "step": 430
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 102.67857360839844,
+      "epoch": 0.6027972027972028,
+      "grad_norm": 0.06317955454404899,
+      "kl": 0.201171875,
+      "learning_rate": 9.92190470372284e-05,
+      "loss": 0.0068,
+      "num_tokens": 4379109.0,
+      "reward": 2.6650938987731934,
+      "reward_std": 0.05930115655064583,
+      "rewards/check_gptzero_func": 0.25,
+      "rewards/check_perplexity_diff_func": 1.9642857313156128,
+      "rewards/check_winston_local_func": 0.4508081078529358,
+      "step": 431
+    },
+    {
+      "clip_ratio": 0.0002587991766631603,
+      "epoch": 0.6041958041958042,
+      "grad_norm": 0.05122842630547611,
+      "kl": 0.1865234375,
+      "learning_rate": 9.920826674424642e-05,
+      "loss": 0.0063,
+      "step": 432
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 80.35714721679688,
+      "epoch": 0.6055944055944056,
+      "grad_norm": 7.162658137250354e-07,
+      "kl": 0.1875,
+      "learning_rate": 9.919741314844577e-05,
+      "loss": 0.0002,
+      "num_tokens": 4396693.0,
+      "reward": 3.0637807846069336,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.5357142686843872,
+      "rewards/check_perplexity_diff_func": 1.8571428060531616,
+      "rewards/check_winston_local_func": 0.6709235310554504,
+      "step": 433
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.606993006993007,
+      "grad_norm": 8.451796450955314e-07,
+      "kl": 0.1875,
+      "learning_rate": 9.918648626599447e-05,
+      "loss": 0.0002,
+      "step": 434
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 124.98214721679688,
+      "epoch": 0.6083916083916084,
+      "grad_norm": 0.016099687268680437,
+      "kl": 0.1455078125,
+      "learning_rate": 9.91754861131697e-05,
+      "loss": -0.0001,
+      "num_tokens": 4418950.0,
+      "reward": 2.94807505607605,
+      "reward_std": 0.050821539014577866,
+      "rewards/check_gptzero_func": 0.5357142686843872,
+      "rewards/check_perplexity_diff_func": 1.8928571939468384,
+      "rewards/check_winston_local_func": 0.5195035338401794,
+      "step": 435
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6097902097902098,
+      "grad_norm": 0.014390989758817311,
+      "kl": 0.1455078125,
+      "learning_rate": 9.916441270635772e-05,
+      "loss": -0.0001,
+      "step": 436
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 106.64286041259766,
+      "epoch": 0.6111888111888112,
+      "grad_norm": 1.968332582426962e-06,
+      "kl": 0.14453125,
+      "learning_rate": 9.915326606205404e-05,
+      "loss": 0.0001,
+      "num_tokens": 4439018.0,
+      "reward": 2.77024507522583,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.4131018817424774,
+      "step": 437
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6125874125874126,
+      "grad_norm": 1.8961430119120163e-06,
+      "kl": 0.14453125,
+      "learning_rate": 9.914204619686314e-05,
+      "loss": 0.0001,
+      "step": 438
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 102.21428680419922,
+      "epoch": 0.6139860139860139,
+      "grad_norm": 0.000668351376631818,
+      "kl": 0.1728515625,
+      "learning_rate": 9.913075312749866e-05,
+      "loss": 0.0002,
+      "num_tokens": 4458806.0,
+      "reward": 2.8182172775268555,
+      "reward_std": 0.0025242711417376995,
+      "rewards/check_gptzero_func": 0.5357142686843872,
+      "rewards/check_perplexity_diff_func": 1.8571428060531616,
+      "rewards/check_winston_local_func": 0.42535990476608276,
+      "step": 439
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6153846153846154,
+      "grad_norm": 0.0006751756209535994,
+      "kl": 0.1728515625,
+      "learning_rate": 9.911938687078324e-05,
+      "loss": 0.0002,
+      "step": 440
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 81.25,
+      "epoch": 0.6167832167832168,
+      "grad_norm": 0.007904133813646444,
+      "kl": 0.240234375,
+      "learning_rate": 9.910794744364857e-05,
+      "loss": -0.0001,
+      "num_tokens": 4475982.0,
+      "reward": 2.8269970417022705,
+      "reward_std": 0.05014092102646828,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 1.8928571939468384,
+      "rewards/check_winston_local_func": 0.5412828326225281,
+      "step": 441
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6181818181818182,
+      "grad_norm": 0.00827597712943806,
+      "kl": 0.240234375,
+      "learning_rate": 9.909643486313533e-05,
+      "loss": -0.0001,
+      "step": 442
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 113.00000762939453,
+      "epoch": 0.6195804195804195,
+      "grad_norm": 6.016032313066646e-06,
+      "kl": 0.16796875,
+      "learning_rate": 9.908484914639318e-05,
+      "loss": 0.0002,
+      "num_tokens": 4497200.0,
+      "reward": 2.9243876934051514,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.5315303206443787,
+      "step": 443
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.620979020979021,
+      "grad_norm": 7.198804696892823e-06,
+      "kl": 0.16796875,
+      "learning_rate": 9.90731903106807e-05,
+      "loss": 0.0002,
+      "step": 444
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 129.375,
+      "epoch": 0.6223776223776224,
+      "grad_norm": 0.02075011817403209,
+      "kl": 0.158203125,
+      "learning_rate": 9.90614583733654e-05,
+      "loss": -0.0001,
+      "num_tokens": 4519807.0,
+      "reward": 2.866011619567871,
+      "reward_std": 0.10101933032274246,
+      "rewards/check_gptzero_func": 0.4642857015132904,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.47315436601638794,
+      "step": 445
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6237762237762238,
+      "grad_norm": 0.0034619333956659956,
+      "kl": 0.1591796875,
+      "learning_rate": 9.904965335192373e-05,
+      "loss": -0.0002,
+      "step": 446
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 98.92857360839844,
+      "epoch": 0.6251748251748251,
+      "grad_norm": 0.010057352537761423,
+      "kl": 0.212890625,
+      "learning_rate": 9.903777526394094e-05,
+      "loss": -0.0086,
+      "num_tokens": 4539039.0,
+      "reward": 2.980358362197876,
+      "reward_std": 0.05100167542695999,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 1.9642857313156128,
+      "rewards/check_winston_local_func": 0.587501049041748,
+      "step": 447
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6265734265734266,
+      "grad_norm": 0.006958390301115274,
+      "kl": 0.21875,
+      "learning_rate": 9.90258241271112e-05,
+      "loss": -0.0087,
+      "step": 448
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 111.71429443359375,
+      "epoch": 0.627972027972028,
+      "grad_norm": 1.6536770511152556e-06,
+      "kl": 0.142578125,
+      "learning_rate": 9.901379995923738e-05,
+      "loss": 0.0001,
+      "num_tokens": 4559675.0,
+      "reward": 2.681791067123413,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.2857142984867096,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.4675052762031555,
+      "step": 449
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6293706293706294,
+      "grad_norm": 1.6773170531853214e-06,
+      "kl": 0.142578125,
+      "learning_rate": 9.900170277823129e-05,
+      "loss": 0.0001,
+      "step": 450
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 107.35714721679688,
+      "epoch": 0.6307692307692307,
+      "grad_norm": 2.0958023033337787e-06,
+      "kl": 0.2109375,
+      "learning_rate": 9.898953260211338e-05,
+      "loss": 0.0002,
+      "num_tokens": 4579995.0,
+      "reward": 2.7400293350219727,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.4186006188392639,
+      "step": 451
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6321678321678321,
+      "grad_norm": 1.982971590769392e-06,
+      "kl": 0.2109375,
+      "learning_rate": 9.897728944901292e-05,
+      "loss": 0.0002,
+      "step": 452
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 118.26786041259766,
+      "epoch": 0.6335664335664336,
+      "grad_norm": 0.009450375112303724,
+      "kl": 0.1416015625,
+      "learning_rate": 9.896497333716783e-05,
+      "loss": -0.0051,
+      "num_tokens": 4602048.0,
+      "reward": 2.640944004058838,
+      "reward_std": 0.043768420815467834,
+      "rewards/check_gptzero_func": 0.5357142686843872,
+      "rewards/check_perplexity_diff_func": 1.5357142686843872,
+      "rewards/check_winston_local_func": 0.5695151090621948,
+      "step": 453
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.634965034965035,
+      "grad_norm": 0.0017713963386990425,
+      "kl": 0.1416015625,
+      "learning_rate": 9.895258428492475e-05,
+      "loss": -0.0051,
+      "step": 454
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 98.42857360839844,
+      "epoch": 0.6363636363636364,
+      "grad_norm": 0.014431904681184264,
+      "kl": 0.1943359375,
+      "learning_rate": 9.894012231073894e-05,
+      "loss": 0.0002,
+      "num_tokens": 4621654.0,
+      "reward": 2.7700040340423584,
+      "reward_std": 0.03728308901190758,
+      "rewards/check_gptzero_func": 0.4107142984867096,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.43071839213371277,
+      "step": 455
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6377622377622377,
+      "grad_norm": 0.008239966094546339,
+      "kl": 0.1943359375,
+      "learning_rate": 9.892758743317434e-05,
+      "loss": 0.0002,
+      "step": 456
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 90.50000762939453,
+      "epoch": 0.6391608391608392,
+      "grad_norm": 2.579298856684113e-06,
+      "kl": 0.1923828125,
+      "learning_rate": 9.891497967090344e-05,
+      "loss": 0.0002,
+      "num_tokens": 4640140.0,
+      "reward": 2.6383018493652344,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.25,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.4597306251525879,
+      "step": 457
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6405594405594406,
+      "grad_norm": 2.6625863420088633e-06,
+      "kl": 0.1923828125,
+      "learning_rate": 9.890229904270731e-05,
+      "loss": 0.0002,
+      "step": 458
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 113.03572082519531,
+      "epoch": 0.641958041958042,
+      "grad_norm": 1.7932858894376123e-05,
+      "kl": 0.1748046875,
+      "learning_rate": 9.888954556747563e-05,
+      "loss": 0.0002,
+      "num_tokens": 4661276.0,
+      "reward": 2.74210524559021,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3571428656578064,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.3849623203277588,
+      "step": 459
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6433566433566433,
+      "grad_norm": 1.7905922488331998e-05,
+      "kl": 0.1748046875,
+      "learning_rate": 9.887671926420648e-05,
+      "loss": 0.0002,
+      "step": 460
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 112.42857360839844,
+      "epoch": 0.6447552447552447,
+      "grad_norm": 0.014077582219597583,
+      "kl": 0.173828125,
+      "learning_rate": 9.886382015200652e-05,
+      "loss": 0.0002,
+      "num_tokens": 4682450.0,
+      "reward": 2.8680295944213867,
+      "reward_std": 0.05049533396959305,
+      "rewards/check_gptzero_func": 0.4642857015132904,
+      "rewards/check_perplexity_diff_func": 1.9642857313156128,
+      "rewards/check_winston_local_func": 0.43945807218551636,
+      "step": 461
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6461538461538462,
+      "grad_norm": 0.010796297977350395,
+      "kl": 0.173828125,
+      "learning_rate": 9.885084825009086e-05,
+      "loss": 0.0001,
+      "step": 462
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 105.14286041259766,
+      "epoch": 0.6475524475524476,
+      "grad_norm": 0.02303930816984521,
+      "kl": 0.17578125,
+      "learning_rate": 9.883780357778299e-05,
+      "loss": 0.0003,
+      "num_tokens": 4702710.0,
+      "reward": 2.885434150695801,
+      "reward_std": 0.06019994616508484,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 1.75,
+      "rewards/check_winston_local_func": 0.7068625092506409,
+      "step": 463
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6489510489510489,
+      "grad_norm": 0.0015050223064125838,
+      "kl": 0.1767578125,
+      "learning_rate": 9.882468615451484e-05,
+      "loss": 0.0002,
+      "step": 464
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 127.5714340209961,
+      "epoch": 0.6503496503496503,
+      "grad_norm": 3.0460982652821775e-06,
+      "kl": 0.1650390625,
+      "learning_rate": 9.881149599982671e-05,
+      "loss": 0.0002,
+      "num_tokens": 4725654.0,
+      "reward": 2.407341957092285,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.2857142984867096,
+      "rewards/check_perplexity_diff_func": 1.8571428060531616,
+      "rewards/check_winston_local_func": 0.2644847333431244,
+      "step": 465
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6517482517482518,
+      "grad_norm": 2.8166666383288673e-06,
+      "kl": 0.1650390625,
+      "learning_rate": 9.879823313336722e-05,
+      "loss": 0.0002,
+      "step": 466
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 122.16072082519531,
+      "epoch": 0.6531468531468532,
+      "grad_norm": 0.0001021700279542065,
+      "kl": 0.140625,
+      "learning_rate": 9.878489757489337e-05,
+      "loss": 0.0001,
+      "num_tokens": 4748375.0,
+      "reward": 2.6119117736816406,
+      "reward_std": 2.7804879209725186e-05,
+      "rewards/check_gptzero_func": 0.5357142686843872,
+      "rewards/check_perplexity_diff_func": 1.5714285373687744,
+      "rewards/check_winston_local_func": 0.5047687888145447,
+      "step": 467
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6545454545454545,
+      "grad_norm": 9.33433928584176e-05,
+      "kl": 0.140625,
+      "learning_rate": 9.877148934427037e-05,
+      "loss": 0.0001,
+      "step": 468
+    },
+    {
+      "clip_ratio": 0.0006711409660056233,
+      "completion_length": 95.64286041259766,
+      "epoch": 0.6559440559440559,
+      "grad_norm": 0.0004684591425632982,
+      "kl": 0.171875,
+      "learning_rate": 9.87580084614717e-05,
+      "loss": 0.0002,
+      "num_tokens": 4767439.0,
+      "reward": 2.3873257637023926,
+      "reward_std": 0.0017833748133853078,
+      "rewards/check_gptzero_func": 0.25,
+      "rewards/check_perplexity_diff_func": 1.7142857313156128,
+      "rewards/check_winston_local_func": 0.42303988337516785,
+      "step": 469
+    },
+    {
+      "clip_ratio": 0.0004793864209204912,
+      "epoch": 0.6573426573426573,
+      "grad_norm": 0.00047799981287926325,
+      "kl": 0.171875,
+      "learning_rate": 9.874445494657911e-05,
+      "loss": 0.0002,
+      "step": 470
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 79.39286041259766,
+      "epoch": 0.6587412587412588,
+      "grad_norm": 0.011441113949782547,
+      "kl": 0.216796875,
+      "learning_rate": 9.873082881978251e-05,
+      "loss": -0.0064,
+      "num_tokens": 4784715.0,
+      "reward": 3.037661075592041,
+      "reward_std": 0.023328183218836784,
+      "rewards/check_gptzero_func": 0.4821428656578064,
+      "rewards/check_perplexity_diff_func": 1.9642857313156128,
+      "rewards/check_winston_local_func": 0.5912323594093323,
+      "step": 471
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6601398601398601,
+      "grad_norm": 0.008433986682608557,
+      "kl": 0.216796875,
+      "learning_rate": 9.871713010137997e-05,
+      "loss": -0.0064,
+      "step": 472
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 119.42857360839844,
+      "epoch": 0.6615384615384615,
+      "grad_norm": 7.638914800437301e-06,
+      "kl": 0.1435546875,
+      "learning_rate": 9.870335881177774e-05,
+      "loss": 0.0001,
+      "num_tokens": 4806207.0,
+      "reward": 2.9706830978393555,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4642857015132904,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.5063972473144531,
+      "step": 473
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6629370629370629,
+      "grad_norm": 7.505737589947172e-06,
+      "kl": 0.1435546875,
+      "learning_rate": 9.868951497149011e-05,
+      "loss": 0.0001,
+      "step": 474
+    },
+    {
+      "clip_ratio": 0.00019073051225859672,
+      "completion_length": 89.87500762939453,
+      "epoch": 0.6643356643356644,
+      "grad_norm": 0.03662860311910791,
+      "kl": 0.2060546875,
+      "learning_rate": 9.86755986011395e-05,
+      "loss": 0.0065,
+      "num_tokens": 4824332.0,
+      "reward": 2.643022060394287,
+      "reward_std": 0.04776443541049957,
+      "rewards/check_gptzero_func": 0.3214285671710968,
+      "rewards/check_perplexity_diff_func": 1.8928571939468384,
+      "rewards/check_winston_local_func": 0.42873620986938477,
+      "step": 475
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6657342657342658,
+      "grad_norm": 0.03126231173733144,
+      "kl": 0.2060546875,
+      "learning_rate": 9.866160972145634e-05,
+      "loss": 0.0062,
+      "step": 476
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 123.10714721679688,
+      "epoch": 0.6671328671328671,
+      "grad_norm": 0.0002559996362709361,
+      "kl": 0.1357421875,
+      "learning_rate": 9.864754835327909e-05,
+      "loss": 0.0001,
+      "num_tokens": 4847094.0,
+      "reward": 2.767104387283325,
+      "reward_std": 0.0007265785825438797,
+      "rewards/check_gptzero_func": 0.5,
+      "rewards/check_perplexity_diff_func": 1.7142857313156128,
+      "rewards/check_winston_local_func": 0.5528185963630676,
+      "step": 477
+    },
+    {
+      "clip_ratio": 8.97343925316818e-05,
+      "epoch": 0.6685314685314685,
+      "grad_norm": 0.00042633622080268514,
+      "kl": 0.1357421875,
+      "learning_rate": 9.86334145175542e-05,
+      "loss": 0.0001,
+      "step": 478
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 96.67857360839844,
+      "epoch": 0.66993006993007,
+      "grad_norm": 0.0008693689003033054,
+      "kl": 0.16015625,
+      "learning_rate": 9.861920823533606e-05,
+      "loss": 0.0002,
+      "num_tokens": 4866394.0,
+      "reward": 2.5761330127716064,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3571428656578064,
+      "rewards/check_perplexity_diff_func": 1.7142857313156128,
+      "rewards/check_winston_local_func": 0.504704475402832,
+      "step": 479
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6713286713286714,
+      "grad_norm": 0.0008435837049153919,
+      "kl": 0.16015625,
+      "learning_rate": 9.860492952778696e-05,
+      "loss": 0.0002,
+      "step": 480
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 94.39286041259766,
+      "epoch": 0.6727272727272727,
+      "grad_norm": 3.349433978828943e-06,
+      "kl": 0.177734375,
+      "learning_rate": 9.859057841617709e-05,
+      "loss": 0.0002,
+      "num_tokens": 4885318.0,
+      "reward": 2.8612334728240967,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3571428656578064,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.5755191445350647,
+      "step": 481
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6741258741258741,
+      "grad_norm": 3.5095106255593373e-06,
+      "kl": 0.177734375,
+      "learning_rate": 9.857615492188452e-05,
+      "loss": 0.0002,
+      "step": 482
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 87.98214721679688,
+      "epoch": 0.6755244755244755,
+      "grad_norm": 0.00016582191324414645,
+      "kl": 0.208984375,
+      "learning_rate": 9.856165906639513e-05,
+      "loss": 0.0002,
+      "num_tokens": 4903401.0,
+      "reward": 2.797102928161621,
+      "reward_std": 0.0008326892857439816,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.4756740629673004,
+      "step": 483
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.676923076923077,
+      "grad_norm": 0.00017592290286556547,
+      "kl": 0.208984375,
+      "learning_rate": 9.85470908713026e-05,
+      "loss": 0.0002,
+      "step": 484
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 122.14286041259766,
+      "epoch": 0.6783216783216783,
+      "grad_norm": 3.840262302649335e-05,
+      "kl": 0.140625,
+      "learning_rate": 9.853245035830834e-05,
+      "loss": 0.0001,
+      "num_tokens": 4925777.0,
+      "reward": 2.80380916595459,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4642857015132904,
+      "rewards/check_perplexity_diff_func": 1.8571428060531616,
+      "rewards/check_winston_local_func": 0.4823804199695587,
+      "step": 485
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6797202797202797,
+      "grad_norm": 3.855750270963149e-05,
+      "kl": 0.140625,
+      "learning_rate": 9.851773754922152e-05,
+      "loss": 0.0001,
+      "step": 486
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 110.5714340209961,
+      "epoch": 0.6811188811188811,
+      "grad_norm": 1.0000521495087345e-06,
+      "kl": 0.1572265625,
+      "learning_rate": 9.850295246595898e-05,
+      "loss": 0.0002,
+      "num_tokens": 4946515.0,
+      "reward": 2.9591448307037354,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.5,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.5305731296539307,
+      "step": 487
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6825174825174826,
+      "grad_norm": 9.579988148024946e-07,
+      "kl": 0.1572265625,
+      "learning_rate": 9.848809513054523e-05,
+      "loss": 0.0002,
+      "step": 488
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 104.67857360839844,
+      "epoch": 0.6839160839160839,
+      "grad_norm": 4.4290499511336513e-07,
+      "kl": 0.1728515625,
+      "learning_rate": 9.847316556511245e-05,
+      "loss": 0.0002,
+      "num_tokens": 4966537.0,
+      "reward": 3.0352413654327393,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4642857015132904,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.5709553956985474,
+      "step": 489
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6853146853146853,
+      "grad_norm": 4.228282290104698e-07,
+      "kl": 0.1728515625,
+      "learning_rate": 9.845816379190036e-05,
+      "loss": 0.0002,
+      "step": 490
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 106.92857360839844,
+      "epoch": 0.6867132867132867,
+      "grad_norm": 7.161264654013381e-07,
+      "kl": 0.1865234375,
+      "learning_rate": 9.844308983325625e-05,
+      "loss": 0.0002,
+      "num_tokens": 4986837.0,
+      "reward": 2.8011527061462402,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4642857015132904,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.40829533338546753,
+      "step": 491
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6881118881118881,
+      "grad_norm": 6.606409864369399e-07,
+      "kl": 0.1865234375,
+      "learning_rate": 9.842794371163501e-05,
+      "loss": 0.0002,
+      "step": 492
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 127.3214340209961,
+      "epoch": 0.6895104895104895,
+      "grad_norm": 3.5056088482557826e-05,
+      "kl": 0.11767578125,
+      "learning_rate": 9.841272544959892e-05,
+      "loss": 0.0001,
+      "num_tokens": 5009783.0,
+      "reward": 2.6788315773010254,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.5,
+      "rewards/check_perplexity_diff_func": 1.7857142686843872,
+      "rewards/check_winston_local_func": 0.393117219209671,
+      "step": 493
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6909090909090909,
+      "grad_norm": 3.5363245519599165e-05,
+      "kl": 0.11767578125,
+      "learning_rate": 9.839743506981782e-05,
+      "loss": 0.0001,
+      "step": 494
+    },
+    {
+      "clip_ratio": 0.0004093328316230327,
+      "completion_length": 99.08928680419922,
+      "epoch": 0.6923076923076923,
+      "grad_norm": 0.0028759704706919223,
+      "kl": 0.1650390625,
+      "learning_rate": 9.838207259506891e-05,
+      "loss": 0.0002,
+      "num_tokens": 5029238.0,
+      "reward": 2.5216493606567383,
+      "reward_std": 0.0011310166446492076,
+      "rewards/check_gptzero_func": 0.3571428656578064,
+      "rewards/check_perplexity_diff_func": 1.7142857313156128,
+      "rewards/check_winston_local_func": 0.45022064447402954,
+      "step": 495
+    },
+    {
+      "clip_ratio": 0.0004093328316230327,
+      "epoch": 0.6937062937062937,
+      "grad_norm": 0.0035050811312865673,
+      "kl": 0.1650390625,
+      "learning_rate": 9.836663804823683e-05,
+      "loss": 0.0002,
+      "step": 496
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 124.85714721679688,
+      "epoch": 0.6951048951048951,
+      "grad_norm": 3.8191875557546916e-05,
+      "kl": 0.158203125,
+      "learning_rate": 9.835113145231356e-05,
+      "loss": 0.0002,
+      "num_tokens": 5051408.0,
+      "reward": 2.8089945316314697,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4642857015132904,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.4161372184753418,
+      "step": 497
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6965034965034965,
+      "grad_norm": 3.1854181850026e-05,
+      "kl": 0.158203125,
+      "learning_rate": 9.833555283039842e-05,
+      "loss": 0.0002,
+      "step": 498
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 115.17857360839844,
+      "epoch": 0.6979020979020979,
+      "grad_norm": 3.735234931957505e-07,
+      "kl": 0.15234375,
+      "learning_rate": 9.831990220569801e-05,
+      "loss": 0.0002,
+      "num_tokens": 5072486.0,
+      "reward": 2.7535111904144287,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3571428656578064,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.46779707074165344,
+      "step": 499
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.6993006993006993,
+      "grad_norm": 3.749561048042854e-07,
+      "kl": 0.15234375,
+      "learning_rate": 9.83041796015262e-05,
+      "loss": 0.0002,
+      "step": 500
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 137.85714721679688,
+      "epoch": 0.7006993006993008,
+      "grad_norm": 0.004581651252064326,
+      "kl": 0.1357421875,
+      "learning_rate": 9.828838504130406e-05,
+      "loss": 0.0001,
+      "num_tokens": 24128.0,
+      "reward": 2.5751242637634277,
+      "reward_std": 0.050465863198041916,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 1.8214285373687744,
+      "rewards/check_winston_local_func": 0.36083847284317017,
+      "step": 501
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7020979020979021,
+      "grad_norm": 0.004623783518082634,
+      "kl": 0.1357421875,
+      "learning_rate": 9.827251854855991e-05,
+      "loss": 0.0001,
+      "step": 502
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 84.0714340209961,
+      "epoch": 0.7034965034965035,
+      "grad_norm": 5.4224474596128455e-06,
+      "kl": 0.1962890625,
+      "learning_rate": 9.825658014692914e-05,
+      "loss": 0.0002,
+      "num_tokens": 41660.0,
+      "reward": 2.881941556930542,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.48908427357673645,
+      "step": 503
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7048951048951049,
+      "grad_norm": 5.774418529053615e-06,
+      "kl": 0.1962890625,
+      "learning_rate": 9.824056986015433e-05,
+      "loss": 0.0002,
+      "step": 504
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 125.76786041259766,
+      "epoch": 0.7062937062937062,
+      "grad_norm": 0.012190639094338367,
+      "kl": 0.158203125,
+      "learning_rate": 9.82244877120851e-05,
+      "loss": -0.0,
+      "num_tokens": 63809.0,
+      "reward": 2.9153220653533936,
+      "reward_std": 0.05151599273085594,
+      "rewards/check_gptzero_func": 0.5,
+      "rewards/check_perplexity_diff_func": 1.9642857313156128,
+      "rewards/check_winston_local_func": 0.45103612542152405,
+      "step": 505
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7076923076923077,
+      "grad_norm": 0.0008706312902749691,
+      "kl": 0.1572265625,
+      "learning_rate": 9.820833372667812e-05,
+      "loss": -0.0001,
+      "step": 506
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 106.67857360839844,
+      "epoch": 0.7090909090909091,
+      "grad_norm": 7.4245877265908374e-06,
+      "kl": 0.171875,
+      "learning_rate": 9.819210792799712e-05,
+      "loss": 0.0002,
+      "num_tokens": 83877.0,
+      "reward": 2.742607831954956,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3571428656578064,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.3854646682739258,
+      "step": 507
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7104895104895105,
+      "grad_norm": 5.9863315776137925e-06,
+      "kl": 0.171875,
+      "learning_rate": 9.817581034021272e-05,
+      "loss": 0.0002,
+      "step": 508
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 112.21429443359375,
+      "epoch": 0.7118881118881119,
+      "grad_norm": 1.5219106062929897e-05,
+      "kl": 0.154296875,
+      "learning_rate": 9.815944098760257e-05,
+      "loss": 0.0002,
+      "num_tokens": 105225.0,
+      "reward": 2.886364698410034,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.5649359226226807,
+      "step": 509
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7132867132867133,
+      "grad_norm": 3.826854189322478e-05,
+      "kl": 0.154296875,
+      "learning_rate": 9.814299989455117e-05,
+      "loss": 0.0002,
+      "step": 510
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 93.50000762939453,
+      "epoch": 0.7146853146853147,
+      "grad_norm": 3.6634275067756628e-06,
+      "kl": 0.18359375,
+      "learning_rate": 9.81264870855499e-05,
+      "loss": 0.0002,
+      "num_tokens": 124237.0,
+      "reward": 2.9750092029571533,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.5,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.5464377403259277,
+      "step": 511
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7160839160839161,
+      "grad_norm": 3.3120330925968346e-06,
+      "kl": 0.18359375,
+      "learning_rate": 9.810990258519699e-05,
+      "loss": 0.0002,
+      "step": 512
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 114.5714340209961,
+      "epoch": 0.7174825174825175,
+      "grad_norm": 1.5887563588226216e-07,
+      "kl": 0.1474609375,
+      "learning_rate": 9.809324641819741e-05,
+      "loss": 0.0001,
+      "num_tokens": 145557.0,
+      "reward": 2.767850637435913,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3571428656578064,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.4107076823711395,
+      "step": 513
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7188811188811188,
+      "grad_norm": 1.5359302478120885e-07,
+      "kl": 0.1474609375,
+      "learning_rate": 9.807651860936297e-05,
+      "loss": 0.0001,
+      "step": 514
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 110.3214340209961,
+      "epoch": 0.7202797202797203,
+      "grad_norm": 1.620447571811333e-07,
+      "kl": 0.1630859375,
+      "learning_rate": 9.805971918361214e-05,
+      "loss": 0.0002,
+      "num_tokens": 165997.0,
+      "reward": 2.9055941104888916,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.5127367377281189,
+      "step": 515
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7216783216783217,
+      "grad_norm": 1.472040658632873e-07,
+      "kl": 0.1630859375,
+      "learning_rate": 9.804284816597008e-05,
+      "loss": 0.0002,
+      "step": 516
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 91.67857360839844,
+      "epoch": 0.7230769230769231,
+      "grad_norm": 2.4623250489986615e-06,
+      "kl": 0.142578125,
+      "learning_rate": 9.802590558156862e-05,
+      "loss": 0.0001,
+      "num_tokens": 184617.0,
+      "reward": 2.853912830352783,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.46105554699897766,
+      "step": 517
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7244755244755244,
+      "grad_norm": 2.0559549189734124e-06,
+      "kl": 0.142578125,
+      "learning_rate": 9.800889145564617e-05,
+      "loss": 0.0001,
+      "step": 518
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 127.21429443359375,
+      "epoch": 0.7258741258741259,
+      "grad_norm": 2.3530034340292295e-07,
+      "kl": 0.1552734375,
+      "learning_rate": 9.799180581354774e-05,
+      "loss": 0.0002,
+      "num_tokens": 207223.0,
+      "reward": 2.7364511489868164,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3571428656578064,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.45073673129081726,
+      "step": 519
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7272727272727273,
+      "grad_norm": 2.2322343800833895e-07,
+      "kl": 0.1552734375,
+      "learning_rate": 9.797464868072488e-05,
+      "loss": 0.0002,
+      "step": 520
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 104.5714340209961,
+      "epoch": 0.7286713286713287,
+      "grad_norm": 2.968346809067501e-07,
+      "kl": 0.177734375,
+      "learning_rate": 9.795742008273558e-05,
+      "loss": 0.0002,
+      "num_tokens": 227059.0,
+      "reward": 2.6932995319366455,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3571428656578064,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.3361565172672272,
+      "step": 521
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.73006993006993,
+      "grad_norm": 2.852980195095417e-07,
+      "kl": 0.177734375,
+      "learning_rate": 9.794012004524434e-05,
+      "loss": 0.0002,
+      "step": 522
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 105.42857360839844,
+      "epoch": 0.7314685314685314,
+      "grad_norm": 3.3796985539550523e-06,
+      "kl": 0.1640625,
+      "learning_rate": 9.792274859402205e-05,
+      "loss": 0.0002,
+      "num_tokens": 246949.0,
+      "reward": 2.6483211517333984,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3214285671710968,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.3983212113380432,
+      "step": 523
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7328671328671329,
+      "grad_norm": 3.1476330793325015e-06,
+      "kl": 0.1640625,
+      "learning_rate": 9.790530575494603e-05,
+      "loss": 0.0002,
+      "step": 524
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 125.03572082519531,
+      "epoch": 0.7342657342657343,
+      "grad_norm": 0.0060337949589422705,
+      "kl": 0.16796875,
+      "learning_rate": 9.788779155399987e-05,
+      "loss": -0.0001,
+      "num_tokens": 269441.0,
+      "reward": 2.768662214279175,
+      "reward_std": 0.049707408994436264,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 1.9642857313156128,
+      "rewards/check_winston_local_func": 0.4115191400051117,
+      "step": 525
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7356643356643356,
+      "grad_norm": 0.0068588182961818464,
+      "kl": 0.16796875,
+      "learning_rate": 9.787020601727352e-05,
+      "loss": -0.0,
+      "step": 526
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 78.39286041259766,
+      "epoch": 0.737062937062937,
+      "grad_norm": 1.3153821115529096e-05,
+      "kl": 0.2353515625,
+      "learning_rate": 9.785254917096318e-05,
+      "loss": 0.0002,
+      "num_tokens": 286345.0,
+      "reward": 3.074557304382324,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.5,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.5745573043823242,
+      "step": 527
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7384615384615385,
+      "grad_norm": 1.256427796379324e-05,
+      "kl": 0.2353515625,
+      "learning_rate": 9.783482104137127e-05,
+      "loss": 0.0002,
+      "step": 528
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 124.39286041259766,
+      "epoch": 0.7398601398601399,
+      "grad_norm": 1.3517164643713064e-05,
+      "kl": 0.12353515625,
+      "learning_rate": 9.781702165490639e-05,
+      "loss": 0.0001,
+      "num_tokens": 309195.0,
+      "reward": 2.4804210662841797,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 1.6428571939468384,
+      "rewards/check_winston_local_func": 0.44470664858818054,
+      "step": 529
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7412587412587412,
+      "grad_norm": 1.3131407312669822e-05,
+      "kl": 0.12353515625,
+      "learning_rate": 9.779915103808328e-05,
+      "loss": 0.0001,
+      "step": 530
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 111.53572082519531,
+      "epoch": 0.7426573426573426,
+      "grad_norm": 7.637416736333039e-07,
+      "kl": 0.15625,
+      "learning_rate": 9.778120921752285e-05,
+      "loss": 0.0002,
+      "num_tokens": 330505.0,
+      "reward": 3.070021390914917,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.5357142686843872,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.5343068838119507,
+      "step": 531
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7440559440559441,
+      "grad_norm": 7.316834780823122e-07,
+      "kl": 0.15625,
+      "learning_rate": 9.776319621995201e-05,
+      "loss": 0.0002,
+      "step": 532
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 106.35714721679688,
+      "epoch": 0.7454545454545455,
+      "grad_norm": 3.663080215251399e-07,
+      "kl": 0.193359375,
+      "learning_rate": 9.77451120722037e-05,
+      "loss": 0.0002,
+      "num_tokens": 350813.0,
+      "reward": 2.63775634765625,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3214285671710968,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.3877563774585724,
+      "step": 533
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7468531468531469,
+      "grad_norm": 3.8678973306591443e-07,
+      "kl": 0.193359375,
+      "learning_rate": 9.77269568012169e-05,
+      "loss": 0.0002,
+      "step": 534
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 106.67857360839844,
+      "epoch": 0.7482517482517482,
+      "grad_norm": 6.123173544504598e-06,
+      "kl": 0.2216796875,
+      "learning_rate": 9.770873043403648e-05,
+      "loss": 0.0002,
+      "num_tokens": 371585.0,
+      "reward": 2.874614953994751,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.5174719095230103,
+      "step": 535
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7496503496503496,
+      "grad_norm": 6.141198966103098e-06,
+      "kl": 0.2216796875,
+      "learning_rate": 9.769043299781327e-05,
+      "loss": 0.0002,
+      "step": 536
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 104.71428680419922,
+      "epoch": 0.7510489510489511,
+      "grad_norm": 6.024782598667822e-07,
+      "kl": 0.212890625,
+      "learning_rate": 9.767206451980394e-05,
+      "loss": 0.0002,
+      "num_tokens": 391201.0,
+      "reward": 2.666369676589966,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.25,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.4163695275783539,
+      "step": 537
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7524475524475525,
+      "grad_norm": 5.651815000751003e-07,
+      "kl": 0.212890625,
+      "learning_rate": 9.765362502737097e-05,
+      "loss": 0.0002,
+      "step": 538
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 107.21428680419922,
+      "epoch": 0.7538461538461538,
+      "grad_norm": 4.5359076717402366e-07,
+      "kl": 0.19140625,
+      "learning_rate": 9.763511454798268e-05,
+      "loss": 0.0002,
+      "num_tokens": 412087.0,
+      "reward": 2.660142660140991,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.25,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.4815710484981537,
+      "step": 539
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7552447552447552,
+      "grad_norm": 7.72545386297213e-07,
+      "kl": 0.19140625,
+      "learning_rate": 9.761653310921307e-05,
+      "loss": 0.0002,
+      "step": 540
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 133.85714721679688,
+      "epoch": 0.7566433566433567,
+      "grad_norm": 8.430058973039463e-08,
+      "kl": 0.146484375,
+      "learning_rate": 9.759788073874189e-05,
+      "loss": 0.0001,
+      "num_tokens": 435535.0,
+      "reward": 2.720244884490967,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.3631021976470947,
+      "step": 541
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7580419580419581,
+      "grad_norm": 8.327469845321351e-08,
+      "kl": 0.146484375,
+      "learning_rate": 9.757915746435453e-05,
+      "loss": 0.0001,
+      "step": 542
+    },
+    {
+      "clip_ratio": 0.0002040816325461492,
+      "completion_length": 102.89286041259766,
+      "epoch": 0.7594405594405594,
+      "grad_norm": 0.000356312443076848,
+      "kl": 0.203125,
+      "learning_rate": 9.756036331394202e-05,
+      "loss": 0.0002,
+      "num_tokens": 455137.0,
+      "reward": 2.6731011867523193,
+      "reward_std": 1.646135569899343e-05,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.351672500371933,
+      "step": 543
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7608391608391608,
+      "grad_norm": 0.00031516713281171804,
+      "kl": 0.203125,
+      "learning_rate": 9.754149831550098e-05,
+      "loss": 0.0002,
+      "step": 544
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 89.35714721679688,
+      "epoch": 0.7622377622377622,
+      "grad_norm": 2.70311309473727e-07,
+      "kl": 0.2080078125,
+      "learning_rate": 9.752256249713351e-05,
+      "loss": 0.0002,
+      "num_tokens": 473297.0,
+      "reward": 2.869070291519165,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3571428656578064,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.5119272470474243,
+      "step": 545
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7636363636363637,
+      "grad_norm": 2.830814999627113e-07,
+      "kl": 0.2080078125,
+      "learning_rate": 9.750355588704727e-05,
+      "loss": 0.0002,
+      "step": 546
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 132.44644165039062,
+      "epoch": 0.765034965034965,
+      "grad_norm": 0.0009688545197921594,
+      "kl": 0.1806640625,
+      "learning_rate": 9.748447851355535e-05,
+      "loss": 0.0002,
+      "num_tokens": 496926.0,
+      "reward": 2.7388832569122314,
+      "reward_std": 0.0023702967446297407,
+      "rewards/check_gptzero_func": 0.5357142686843872,
+      "rewards/check_perplexity_diff_func": 1.7142857313156128,
+      "rewards/check_winston_local_func": 0.4888834059238434,
+      "step": 547
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7664335664335664,
+      "grad_norm": 0.0016734864168510978,
+      "kl": 0.1806640625,
+      "learning_rate": 9.746533040507624e-05,
+      "loss": 0.0002,
+      "step": 548
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 97.03572082519531,
+      "epoch": 0.7678321678321678,
+      "grad_norm": 1.9163803223436476e-06,
+      "kl": 0.177734375,
+      "learning_rate": 9.744611159013381e-05,
+      "loss": 0.0002,
+      "num_tokens": 516080.0,
+      "reward": 2.9889118671417236,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.5,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.48891177773475647,
+      "step": 549
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7692307692307693,
+      "grad_norm": 1.2761616262245509e-06,
+      "kl": 0.177734375,
+      "learning_rate": 9.742682209735727e-05,
+      "loss": 0.0002,
+      "step": 550
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 92.39286041259766,
+      "epoch": 0.7706293706293706,
+      "grad_norm": 1.896094459987994e-07,
+      "kl": 0.1923828125,
+      "learning_rate": 9.740746195548112e-05,
+      "loss": 0.0002,
+      "num_tokens": 534554.0,
+      "reward": 2.943678855895996,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.5151072144508362,
+      "step": 551
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.772027972027972,
+      "grad_norm": 2.1137044966091938e-07,
+      "kl": 0.1923828125,
+      "learning_rate": 9.738803119334506e-05,
+      "loss": 0.0002,
+      "step": 552
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 103.21428680419922,
+      "epoch": 0.7734265734265734,
+      "grad_norm": 4.705318347406964e-07,
+      "kl": 0.1826171875,
+      "learning_rate": 9.736852983989404e-05,
+      "loss": 0.0002,
+      "num_tokens": 554220.0,
+      "reward": 2.9958415031433105,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.5,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.4958415627479553,
+      "step": 553
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7748251748251749,
+      "grad_norm": 5.24917119837944e-07,
+      "kl": 0.1826171875,
+      "learning_rate": 9.734895792417811e-05,
+      "loss": 0.0002,
+      "step": 554
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 102.53572082519531,
+      "epoch": 0.7762237762237763,
+      "grad_norm": 0.0005595837692618245,
+      "kl": 0.1611328125,
+      "learning_rate": 9.73293154753525e-05,
+      "loss": -0.0001,
+      "num_tokens": 574520.0,
+      "reward": 2.660358190536499,
+      "reward_std": 0.05061452463269234,
+      "rewards/check_gptzero_func": 0.2857142984867096,
+      "rewards/check_perplexity_diff_func": 1.75,
+      "rewards/check_winston_local_func": 0.6246438026428223,
+      "step": 555
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7776223776223776,
+      "grad_norm": 0.0005476655329061088,
+      "kl": 0.1611328125,
+      "learning_rate": 9.730960252267743e-05,
+      "loss": -0.0001,
+      "step": 556
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 110.28572082519531,
+      "epoch": 0.779020979020979,
+      "grad_norm": 1.9994241055330475e-07,
+      "kl": 0.162109375,
+      "learning_rate": 9.728981909551824e-05,
+      "loss": 0.0002,
+      "num_tokens": 595056.0,
+      "reward": 2.7659287452697754,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3571428656578064,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.40878555178642273,
+      "step": 557
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7804195804195804,
+      "grad_norm": 2.102720872162732e-07,
+      "kl": 0.162109375,
+      "learning_rate": 9.726996522334516e-05,
+      "loss": 0.0002,
+      "step": 558
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 116.3214340209961,
+      "epoch": 0.7818181818181819,
+      "grad_norm": 1.4418577967265551e-06,
+      "kl": 0.197265625,
+      "learning_rate": 9.725004093573342e-05,
+      "loss": 0.0002,
+      "num_tokens": 616218.0,
+      "reward": 2.90523099899292,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.4766596257686615,
+      "step": 559
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7832167832167832,
+      "grad_norm": 1.4761501812795077e-06,
+      "kl": 0.197265625,
+      "learning_rate": 9.723004626236314e-05,
+      "loss": 0.0002,
+      "step": 560
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 92.28572082519531,
+      "epoch": 0.7846153846153846,
+      "grad_norm": 6.640556191873379e-07,
+      "kl": 0.1904296875,
+      "learning_rate": 9.720998123301923e-05,
+      "loss": 0.0002,
+      "num_tokens": 634450.0,
+      "reward": 2.910466432571411,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4642857015132904,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.4461804926395416,
+      "step": 561
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.786013986013986,
+      "grad_norm": 6.785626739636823e-07,
+      "kl": 0.1904296875,
+      "learning_rate": 9.718984587759148e-05,
+      "loss": 0.0002,
+      "step": 562
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 97.67857360839844,
+      "epoch": 0.7874125874125875,
+      "grad_norm": 0.0002364011375932314,
+      "kl": 0.1953125,
+      "learning_rate": 9.71696402260744e-05,
+      "loss": 0.0002,
+      "num_tokens": 653730.0,
+      "reward": 3.09153151512146,
+      "reward_std": 4.5663102355320007e-05,
+      "rewards/check_gptzero_func": 0.5714285969734192,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.5201030373573303,
+      "step": 563
+    },
+    {
+      "clip_ratio": 0.00014228800137061626,
+      "epoch": 0.7888111888111888,
+      "grad_norm": 0.00020388090420570156,
+      "kl": 0.1953125,
+      "learning_rate": 9.714936430856723e-05,
+      "loss": 0.0002,
+      "step": 564
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 120.66072082519531,
+      "epoch": 0.7902097902097902,
+      "grad_norm": 8.089432221384178e-05,
+      "kl": 0.1357421875,
+      "learning_rate": 9.712901815527386e-05,
+      "loss": 0.0001,
+      "num_tokens": 675657.0,
+      "reward": 2.674098253250122,
+      "reward_std": 0.0001108883589040488,
+      "rewards/check_gptzero_func": 0.3571428656578064,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.38838380575180054,
+      "step": 565
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7916083916083916,
+      "grad_norm": 8.889658368646956e-05,
+      "kl": 0.1357421875,
+      "learning_rate": 9.710860179650287e-05,
+      "loss": 0.0001,
+      "step": 566
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 107.3214340209961,
+      "epoch": 0.793006993006993,
+      "grad_norm": 0.0001037228042730343,
+      "kl": 0.17578125,
+      "learning_rate": 9.70881152626673e-05,
+      "loss": 0.0002,
+      "num_tokens": 695807.0,
+      "reward": 2.615325927734375,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.25,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.4367544949054718,
+      "step": 567
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7944055944055944,
+      "grad_norm": 9.62410190276379e-05,
+      "kl": 0.17578125,
+      "learning_rate": 9.706755858428486e-05,
+      "loss": 0.0002,
+      "step": 568
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 128.3928680419922,
+      "epoch": 0.7958041958041958,
+      "grad_norm": 3.2052372287941215e-05,
+      "kl": 0.12890625,
+      "learning_rate": 9.704693179197767e-05,
+      "loss": 0.0001,
+      "num_tokens": 719201.0,
+      "reward": 2.616921901702881,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 1.7857142686843872,
+      "rewards/check_winston_local_func": 0.40263620018959045,
+      "step": 569
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.7972027972027972,
+      "grad_norm": 3.1589569913845654e-05,
+      "kl": 0.12890625,
+      "learning_rate": 9.702623491647233e-05,
+      "loss": 0.0001,
+      "step": 570
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 97.3214340209961,
+      "epoch": 0.7986013986013986,
+      "grad_norm": 2.966365002312992e-07,
+      "kl": 0.1572265625,
+      "learning_rate": 9.70054679885998e-05,
+      "loss": 0.0002,
+      "num_tokens": 738947.0,
+      "reward": 2.835333824157715,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.47819074988365173,
+      "step": 571
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8,
+      "grad_norm": 4.1573650424054196e-07,
+      "kl": 0.1572265625,
+      "learning_rate": 9.698463103929542e-05,
+      "loss": 0.0002,
+      "step": 572
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 105.78572082519531,
+      "epoch": 0.8013986013986014,
+      "grad_norm": 0.000588551803293507,
+      "kl": 0.1611328125,
+      "learning_rate": 9.696372409959886e-05,
+      "loss": 0.0002,
+      "num_tokens": 759359.0,
+      "reward": 2.967468738555908,
+      "reward_std": 0.0020603849552571774,
+      "rewards/check_gptzero_func": 0.5,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.5388973355293274,
+      "step": 573
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8027972027972028,
+      "grad_norm": 0.0006215594211925717,
+      "kl": 0.1611328125,
+      "learning_rate": 9.694274720065399e-05,
+      "loss": 0.0002,
+      "step": 574
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 97.14286041259766,
+      "epoch": 0.8041958041958042,
+      "grad_norm": 3.3967306751168834e-06,
+      "kl": 0.17578125,
+      "learning_rate": 9.692170037370898e-05,
+      "loss": 0.0002,
+      "num_tokens": 778637.0,
+      "reward": 2.9381115436553955,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.509539783000946,
+      "step": 575
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8055944055944056,
+      "grad_norm": 3.2036118908588705e-06,
+      "kl": 0.17578125,
+      "learning_rate": 9.690058365011607e-05,
+      "loss": 0.0002,
+      "step": 576
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 102.64286041259766,
+      "epoch": 0.806993006993007,
+      "grad_norm": 8.125129903430204e-06,
+      "kl": 0.1630859375,
+      "learning_rate": 9.687939706133168e-05,
+      "loss": 0.0002,
+      "num_tokens": 798473.0,
+      "reward": 2.781018018722534,
+      "reward_std": 0.025253813713788986,
+      "rewards/check_gptzero_func": 0.4107142984867096,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.4417320787906647,
+      "step": 577
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8083916083916084,
+      "grad_norm": 7.98774787404577e-06,
+      "kl": 0.1630859375,
+      "learning_rate": 9.685814063891631e-05,
+      "loss": 0.0002,
+      "step": 578
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 126.85714721679688,
+      "epoch": 0.8097902097902098,
+      "grad_norm": 3.679599498892351e-07,
+      "kl": 0.138671875,
+      "learning_rate": 9.683681441453445e-05,
+      "loss": 0.0001,
+      "num_tokens": 821081.0,
+      "reward": 2.8629186153411865,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.43434715270996094,
+      "step": 579
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8111888111888111,
+      "grad_norm": 3.93663211553172e-07,
+      "kl": 0.138671875,
+      "learning_rate": 9.681541841995461e-05,
+      "loss": 0.0001,
+      "step": 580
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 106.00000762939453,
+      "epoch": 0.8125874125874126,
+      "grad_norm": 8.022029708518222e-07,
+      "kl": 0.162109375,
+      "learning_rate": 9.67939526870492e-05,
+      "loss": 0.0002,
+      "num_tokens": 841529.0,
+      "reward": 2.8752760887145996,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.55384761095047,
+      "step": 581
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.813986013986014,
+      "grad_norm": 8.501682817147847e-07,
+      "kl": 0.162109375,
+      "learning_rate": 9.677241724779453e-05,
+      "loss": 0.0002,
+      "step": 582
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 109.85714721679688,
+      "epoch": 0.8153846153846154,
+      "grad_norm": 1.7930012533014129e-06,
+      "kl": 0.1689453125,
+      "learning_rate": 9.675081213427076e-05,
+      "loss": 0.0002,
+      "num_tokens": 862213.0,
+      "reward": 2.9139628410339355,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.48539119958877563,
+      "step": 583
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8167832167832167,
+      "grad_norm": 1.6519465345892093e-06,
+      "kl": 0.1689453125,
+      "learning_rate": 9.672913737866179e-05,
+      "loss": 0.0002,
+      "step": 584
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 98.66072082519531,
+      "epoch": 0.8181818181818182,
+      "grad_norm": 0.002787236207669909,
+      "kl": 0.25,
+      "learning_rate": 9.670739301325534e-05,
+      "loss": 0.0002,
+      "num_tokens": 881842.0,
+      "reward": 2.6230576038360596,
+      "reward_std": 0.006216124631464481,
+      "rewards/check_gptzero_func": 0.3571428656578064,
+      "rewards/check_perplexity_diff_func": 1.7857142686843872,
+      "rewards/check_winston_local_func": 0.48020049929618835,
+      "step": 585
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8195804195804196,
+      "grad_norm": 0.002885109231591189,
+      "kl": 0.2490234375,
+      "learning_rate": 9.668557907044276e-05,
+      "loss": 0.0002,
+      "step": 586
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 79.14286041259766,
+      "epoch": 0.820979020979021,
+      "grad_norm": 4.462250819797621e-06,
+      "kl": 0.1923828125,
+      "learning_rate": 9.666369558271909e-05,
+      "loss": 0.0002,
+      "num_tokens": 898716.0,
+      "reward": 2.900649309158325,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3928571343421936,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.5077921152114868,
+      "step": 587
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8223776223776224,
+      "grad_norm": 4.488075066816524e-06,
+      "kl": 0.1923828125,
+      "learning_rate": 9.66417425826829e-05,
+      "loss": 0.0002,
+      "step": 588
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 112.03572082519531,
+      "epoch": 0.8237762237762237,
+      "grad_norm": 1.3821264354883407e-07,
+      "kl": 0.14453125,
+      "learning_rate": 9.661972010303641e-05,
+      "loss": 0.0001,
+      "num_tokens": 919636.0,
+      "reward": 2.7668612003326416,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4642857015132904,
+      "rewards/check_perplexity_diff_func": 1.8571428060531616,
+      "rewards/check_winston_local_func": 0.4454323649406433,
+      "step": 589
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8251748251748252,
+      "grad_norm": 1.4174011116816794e-07,
+      "kl": 0.14453125,
+      "learning_rate": 9.659762817658524e-05,
+      "loss": 0.0001,
+      "step": 590
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 107.92857360839844,
+      "epoch": 0.8265734265734266,
+      "grad_norm": 8.773734381605369e-07,
+      "kl": 0.1572265625,
+      "learning_rate": 9.65754668362385e-05,
+      "loss": 0.0002,
+      "num_tokens": 940096.0,
+      "reward": 2.8289589881896973,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4642857015132904,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.4361015856266022,
+      "step": 591
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.827972027972028,
+      "grad_norm": 1.0023469469060766e-06,
+      "kl": 0.1572265625,
+      "learning_rate": 9.655323611500875e-05,
+      "loss": 0.0002,
+      "step": 592
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 113.85714721679688,
+      "epoch": 0.8293706293706293,
+      "grad_norm": 6.1541718025779055e-06,
+      "kl": 0.166015625,
+      "learning_rate": 9.653093604601183e-05,
+      "loss": 0.0002,
+      "num_tokens": 961340.0,
+      "reward": 2.9606730937957764,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.5357142686843872,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.4249587059020996,
+      "step": 593
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8307692307692308,
+      "grad_norm": 6.512813773208274e-06,
+      "kl": 0.166015625,
+      "learning_rate": 9.650856666246693e-05,
+      "loss": 0.0002,
+      "step": 594
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 91.60714721679688,
+      "epoch": 0.8321678321678322,
+      "grad_norm": 3.2190340652283214e-06,
+      "kl": 0.1767578125,
+      "learning_rate": 9.648612799769644e-05,
+      "loss": 0.0002,
+      "num_tokens": 979916.0,
+      "reward": 3.0096070766448975,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 2.0,
+      "rewards/check_winston_local_func": 0.5810357332229614,
+      "step": 595
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8335664335664336,
+      "grad_norm": 3.1137752081162388e-06,
+      "kl": 0.1767578125,
+      "learning_rate": 9.646362008512602e-05,
+      "loss": 0.0002,
+      "step": 596
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 106.53572082519531,
+      "epoch": 0.8349650349650349,
+      "grad_norm": 6.812051267845749e-06,
+      "kl": 0.185546875,
+      "learning_rate": 9.644104295828447e-05,
+      "loss": 0.0002,
+      "num_tokens": 1000300.0,
+      "reward": 2.7739078998565674,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 1.9285714626312256,
+      "rewards/check_winston_local_func": 0.4167649447917938,
+      "step": 597
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8363636363636363,
+      "grad_norm": 2.398683493432653e-06,
+      "kl": 0.185546875,
+      "learning_rate": 9.641839665080363e-05,
+      "loss": 0.0002,
+      "step": 598
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 107.41072082519531,
+      "epoch": 0.8377622377622378,
+      "grad_norm": 0.0008785473557125962,
+      "kl": 0.1796875,
+      "learning_rate": 9.63956811964185e-05,
+      "loss": 0.0002,
+      "num_tokens": 1021105.0,
+      "reward": 2.91171932220459,
+      "reward_std": 0.00400555832311511,
+      "rewards/check_gptzero_func": 0.5357142686843872,
+      "rewards/check_perplexity_diff_func": 1.8571428060531616,
+      "rewards/check_winston_local_func": 0.518862247467041,
+      "step": 599
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8391608391608392,
+      "grad_norm": 0.000892784560673435,
+      "kl": 0.1796875,
+      "learning_rate": 9.6372896628967e-05,
+      "loss": 0.0002,
+      "step": 600
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 78.8214340209961,
+      "epoch": 0.8405594405594405,
+      "grad_norm": 3.255738764439449e-05,
+      "kl": 0.2236328125,
+      "learning_rate": 9.635004298239004e-05,
+      "loss": 0.0002,
+      "num_tokens": 16776.0,
+      "reward": 2.546457290649414,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 1.0714285373687744,
+      "rewards/check_perplexity_diff_func": 0.9285714030265808,
+      "rewards/check_winston_local_func": 0.5464571118354797,
+      "step": 601
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8419580419580419,
+      "grad_norm": 2.996409313034994e-05,
+      "kl": 0.2236328125,
+      "learning_rate": 9.632712029073141e-05,
+      "loss": 0.0002,
+      "step": 602
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 98.50000762939453,
+      "epoch": 0.8433566433566434,
+      "grad_norm": 6.836714560556308e-07,
+      "kl": 0.1796875,
+      "learning_rate": 9.63041285881378e-05,
+      "loss": 0.0002,
+      "num_tokens": 36316.0,
+      "reward": 2.2952983379364014,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.8571428656578064,
+      "rewards/check_perplexity_diff_func": 0.9642857313156128,
+      "rewards/check_winston_local_func": 0.4738696217536926,
+      "step": 603
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8447552447552448,
+      "grad_norm": 6.51699821327856e-07,
+      "kl": 0.1796875,
+      "learning_rate": 9.628106790885865e-05,
+      "loss": 0.0002,
+      "step": 604
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 108.67857360839844,
+      "epoch": 0.8461538461538461,
+      "grad_norm": 1.0500525475797962e-05,
+      "kl": 0.220703125,
+      "learning_rate": 9.625793828724618e-05,
+      "loss": 0.0002,
+      "num_tokens": 56798.0,
+      "reward": 2.2515501976013184,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.7857142686843872,
+      "rewards/check_perplexity_diff_func": 0.9642857313156128,
+      "rewards/check_winston_local_func": 0.5015498399734497,
+      "step": 605
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8475524475524475,
+      "grad_norm": 9.962219746884584e-06,
+      "kl": 0.220703125,
+      "learning_rate": 9.62347397577553e-05,
+      "loss": 0.0002,
+      "step": 606
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 123.64286041259766,
+      "epoch": 0.848951048951049,
+      "grad_norm": 0.00014103568607896937,
+      "kl": 0.1728515625,
+      "learning_rate": 9.621147235494356e-05,
+      "loss": 0.0002,
+      "num_tokens": 79358.0,
+      "reward": 2.151411771774292,
+      "reward_std": 0.00029218001873232424,
+      "rewards/check_gptzero_func": 0.7142857313156128,
+      "rewards/check_perplexity_diff_func": 0.9285714030265808,
+      "rewards/check_winston_local_func": 0.5085545778274536,
+      "step": 607
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8503496503496504,
+      "grad_norm": 0.00012130440152598759,
+      "kl": 0.171875,
+      "learning_rate": 9.618813611347113e-05,
+      "loss": 0.0002,
+      "step": 608
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 118.71429443359375,
+      "epoch": 0.8517482517482518,
+      "grad_norm": 1.8287469066668684e-07,
+      "kl": 0.1494140625,
+      "learning_rate": 9.61647310681007e-05,
+      "loss": 0.0001,
+      "num_tokens": 101324.0,
+      "reward": 2.4129910469055176,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.9285714030265808,
+      "rewards/check_perplexity_diff_func": 0.9285714030265808,
+      "rewards/check_winston_local_func": 0.5558480024337769,
+      "step": 609
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8531468531468531,
+      "grad_norm": 1.9169052102503996e-07,
+      "kl": 0.1494140625,
+      "learning_rate": 9.614125725369747e-05,
+      "loss": 0.0001,
+      "step": 610
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 84.30357360839844,
+      "epoch": 0.8545454545454545,
+      "grad_norm": 0.0035849846804439827,
+      "kl": 0.2314453125,
+      "learning_rate": 9.611771470522908e-05,
+      "loss": 0.0002,
+      "num_tokens": 118947.0,
+      "reward": 2.3909339904785156,
+      "reward_std": 0.0009617910836823285,
+      "rewards/check_gptzero_func": 0.8571428656578064,
+      "rewards/check_perplexity_diff_func": 1.0,
+      "rewards/check_winston_local_func": 0.533791184425354,
+      "step": 611
+    },
+    {
+      "clip_ratio": 0.00020614307140931487,
+      "epoch": 0.855944055944056,
+      "grad_norm": 0.0024010748168965985,
+      "kl": 0.22265625,
+      "learning_rate": 9.609410345776554e-05,
+      "loss": 0.0002,
+      "step": 612
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 97.92857360839844,
+      "epoch": 0.8573426573426574,
+      "grad_norm": 0.0165782655855821,
+      "kl": 0.220703125,
+      "learning_rate": 9.607042354647924e-05,
+      "loss": -0.0006,
+      "num_tokens": 138117.0,
+      "reward": 2.496706485748291,
+      "reward_std": 0.025315813720226288,
+      "rewards/check_gptzero_func": 0.9642857313156128,
+      "rewards/check_perplexity_diff_func": 0.9464285969734192,
+      "rewards/check_winston_local_func": 0.585992157459259,
+      "step": 613
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8587412587412587,
+      "grad_norm": 0.005231318150303852,
+      "kl": 0.220703125,
+      "learning_rate": 9.60466750066448e-05,
+      "loss": -0.0006,
+      "step": 614
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 116.71429443359375,
+      "epoch": 0.8601398601398601,
+      "grad_norm": 1.4205544083992568e-05,
+      "kl": 0.1533203125,
+      "learning_rate": 9.60228578736391e-05,
+      "loss": 0.0002,
+      "num_tokens": 159543.0,
+      "reward": 2.470890998840332,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 1.0,
+      "rewards/check_perplexity_diff_func": 1.0,
+      "rewards/check_winston_local_func": 0.47089076042175293,
+      "step": 615
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8615384615384616,
+      "grad_norm": 1.6303392641980344e-05,
+      "kl": 0.1533203125,
+      "learning_rate": 9.599897218294122e-05,
+      "loss": 0.0002,
+      "step": 616
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 112.50000762939453,
+      "epoch": 0.862937062937063,
+      "grad_norm": 1.6234044022820597e-07,
+      "kl": 0.2021484375,
+      "learning_rate": 9.597501797013233e-05,
+      "loss": 0.0002,
+      "num_tokens": 180563.0,
+      "reward": 2.09031081199646,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.6428571343421936,
+      "rewards/check_perplexity_diff_func": 0.9285714030265808,
+      "rewards/check_winston_local_func": 0.5188822746276855,
+      "step": 617
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8643356643356643,
+      "grad_norm": 1.666191215120993e-07,
+      "kl": 0.2021484375,
+      "learning_rate": 9.595099527089569e-05,
+      "loss": 0.0002,
+      "step": 618
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 108.78572082519531,
+      "epoch": 0.8657342657342657,
+      "grad_norm": 8.531816841675701e-07,
+      "kl": 0.193359375,
+      "learning_rate": 9.592690412101658e-05,
+      "loss": 0.0002,
+      "num_tokens": 201177.0,
+      "reward": 1.9570515155792236,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.5714285969734192,
+      "rewards/check_perplexity_diff_func": 1.0,
+      "rewards/check_winston_local_func": 0.38562270998954773,
+      "step": 619
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8671328671328671,
+      "grad_norm": 8.842725127740978e-07,
+      "kl": 0.193359375,
+      "learning_rate": 9.590274455638225e-05,
+      "loss": 0.0002,
+      "step": 620
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 138.07144165039062,
+      "epoch": 0.8685314685314686,
+      "grad_norm": 5.1580717328778724e-05,
+      "kl": 0.1640625,
+      "learning_rate": 9.587851661298186e-05,
+      "loss": 0.0002,
+      "num_tokens": 225661.0,
+      "reward": 2.487313747406006,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 1.0714285373687744,
+      "rewards/check_perplexity_diff_func": 0.9642857313156128,
+      "rewards/check_winston_local_func": 0.4515993893146515,
+      "step": 621
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8699300699300699,
+      "grad_norm": 5.097158350926591e-05,
+      "kl": 0.1640625,
+      "learning_rate": 9.585422032690643e-05,
+      "loss": 0.0002,
+      "step": 622
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 126.30357360839844,
+      "epoch": 0.8713286713286713,
+      "grad_norm": 1.5371764428374234e-05,
+      "kl": 0.1494140625,
+      "learning_rate": 9.582985573434877e-05,
+      "loss": 0.0002,
+      "num_tokens": 248756.0,
+      "reward": 2.386916160583496,
+      "reward_std": 0.0005263118073344231,
+      "rewards/check_gptzero_func": 0.8571428656578064,
+      "rewards/check_perplexity_diff_func": 1.0,
+      "rewards/check_winston_local_func": 0.5297732949256897,
+      "step": 623
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8727272727272727,
+      "grad_norm": 1.5410923964835553e-05,
+      "kl": 0.1494140625,
+      "learning_rate": 9.580542287160348e-05,
+      "loss": 0.0002,
+      "step": 624
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 121.85714721679688,
+      "epoch": 0.8741258741258742,
+      "grad_norm": 0.0002966067751134756,
+      "kl": 0.173828125,
+      "learning_rate": 9.578092177506683e-05,
+      "loss": 0.0002,
+      "num_tokens": 270514.0,
+      "reward": 2.7340447902679443,
+      "reward_std": 6.457499694079161e-05,
+      "rewards/check_gptzero_func": 1.2142857313156128,
+      "rewards/check_perplexity_diff_func": 0.9642857313156128,
+      "rewards/check_winston_local_func": 0.555473268032074,
+      "step": 625
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8755244755244755,
+      "grad_norm": 0.0002795708879844016,
+      "kl": 0.173828125,
+      "learning_rate": 9.575635248123675e-05,
+      "loss": 0.0002,
+      "step": 626
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 97.89286041259766,
+      "epoch": 0.8769230769230769,
+      "grad_norm": 1.0675950775787285e-07,
+      "kl": 0.1904296875,
+      "learning_rate": 9.573171502671273e-05,
+      "loss": 0.0002,
+      "num_tokens": 289588.0,
+      "reward": 2.467723846435547,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.9285714030265808,
+      "rewards/check_perplexity_diff_func": 1.0,
+      "rewards/check_winston_local_func": 0.5391523838043213,
+      "step": 627
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8783216783216783,
+      "grad_norm": 1.0758265762547126e-07,
+      "kl": 0.1904296875,
+      "learning_rate": 9.570700944819584e-05,
+      "loss": 0.0002,
+      "step": 628
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 126.96429443359375,
+      "epoch": 0.8797202797202798,
+      "grad_norm": 1.8189249055345895e-06,
+      "kl": 0.150390625,
+      "learning_rate": 9.568223578248858e-05,
+      "loss": 0.0002,
+      "num_tokens": 312792.0,
+      "reward": 2.4109601974487305,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.9285714030265808,
+      "rewards/check_perplexity_diff_func": 0.9285714030265808,
+      "rewards/check_winston_local_func": 0.5538173913955688,
+      "step": 629
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8811188811188811,
+      "grad_norm": 1.817429271106412e-06,
+      "kl": 0.150390625,
+      "learning_rate": 9.565739406649493e-05,
+      "loss": 0.0002,
+      "step": 630
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 109.64286041259766,
+      "epoch": 0.8825174825174825,
+      "grad_norm": 7.209254474831735e-05,
+      "kl": 0.169921875,
+      "learning_rate": 9.563248433722019e-05,
+      "loss": 0.0002,
+      "num_tokens": 333908.0,
+      "reward": 2.1215410232543945,
+      "reward_std": 0.0032710533123463392,
+      "rewards/check_gptzero_func": 0.7142857313156128,
+      "rewards/check_perplexity_diff_func": 0.9285714030265808,
+      "rewards/check_winston_local_func": 0.4786837100982666,
+      "step": 631
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8839160839160839,
+      "grad_norm": 0.005778816018850288,
+      "kl": 0.169921875,
+      "learning_rate": 9.560750663177101e-05,
+      "loss": 0.0002,
+      "step": 632
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 101.0714340209961,
+      "epoch": 0.8853146853146853,
+      "grad_norm": 5.141706041264495e-06,
+      "kl": 0.1689453125,
+      "learning_rate": 9.558246098735528e-05,
+      "loss": 0.0002,
+      "num_tokens": 353808.0,
+      "reward": 2.3030316829681396,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.7857142686843872,
+      "rewards/check_perplexity_diff_func": 0.9642857313156128,
+      "rewards/check_winston_local_func": 0.5530317425727844,
+      "step": 633
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8867132867132868,
+      "grad_norm": 5.171400787940402e-06,
+      "kl": 0.1689453125,
+      "learning_rate": 9.55573474412821e-05,
+      "loss": 0.0002,
+      "step": 634
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 112.0714340209961,
+      "epoch": 0.8881118881118881,
+      "grad_norm": 2.5579574578411833e-07,
+      "kl": 0.16015625,
+      "learning_rate": 9.553216603096176e-05,
+      "loss": 0.0002,
+      "num_tokens": 375024.0,
+      "reward": 2.2100911140441895,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.7857142686843872,
+      "rewards/check_perplexity_diff_func": 1.0,
+      "rewards/check_winston_local_func": 0.42437654733657837,
+      "step": 635
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8895104895104895,
+      "grad_norm": 3.163234894184058e-07,
+      "kl": 0.16015625,
+      "learning_rate": 9.550691679390558e-05,
+      "loss": 0.0002,
+      "step": 636
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 78.39286041259766,
+      "epoch": 0.8909090909090909,
+      "grad_norm": 2.7422578147300963e-06,
+      "kl": 0.19921875,
+      "learning_rate": 9.548159976772592e-05,
+      "loss": 0.0002,
+      "num_tokens": 391944.0,
+      "reward": 2.234971523284912,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.7142857313156128,
+      "rewards/check_perplexity_diff_func": 0.9285714030265808,
+      "rewards/check_winston_local_func": 0.5921144485473633,
+      "step": 637
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8923076923076924,
+      "grad_norm": 2.3669860027714333e-06,
+      "kl": 0.19921875,
+      "learning_rate": 9.54562149901362e-05,
+      "loss": 0.0002,
+      "step": 638
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 120.75000762939453,
+      "epoch": 0.8937062937062937,
+      "grad_norm": 2.3171625543487874e-07,
+      "kl": 0.158203125,
+      "learning_rate": 9.543076249895066e-05,
+      "loss": 0.0002,
+      "num_tokens": 413760.0,
+      "reward": 2.152951955795288,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.7142857313156128,
+      "rewards/check_perplexity_diff_func": 1.0,
+      "rewards/check_winston_local_func": 0.43866607546806335,
+      "step": 639
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8951048951048951,
+      "grad_norm": 2.4228268027328343e-07,
+      "kl": 0.158203125,
+      "learning_rate": 9.540524233208448e-05,
+      "loss": 0.0002,
+      "step": 640
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 87.05357360839844,
+      "epoch": 0.8965034965034965,
+      "grad_norm": 0.008398279317441937,
+      "kl": 0.181640625,
+      "learning_rate": 9.537965452755365e-05,
+      "loss": 0.0094,
+      "num_tokens": 431959.0,
+      "reward": 2.0755865573883057,
+      "reward_std": 0.019971024245023727,
+      "rewards/check_gptzero_func": 0.7142857313156128,
+      "rewards/check_perplexity_diff_func": 0.8571428656578064,
+      "rewards/check_winston_local_func": 0.5041577219963074,
+      "step": 641
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.8979020979020979,
+      "grad_norm": 0.0032568280015785474,
+      "kl": 0.181640625,
+      "learning_rate": 9.535399912347489e-05,
+      "loss": 0.0094,
+      "step": 642
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 107.28572082519531,
+      "epoch": 0.8993006993006993,
+      "grad_norm": 4.5580158500525314e-05,
+      "kl": 0.150390625,
+      "learning_rate": 9.53282761580656e-05,
+      "loss": 0.0001,
+      "num_tokens": 452913.0,
+      "reward": 2.6801095008850098,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 1.1428571939468384,
+      "rewards/check_perplexity_diff_func": 0.9285714030265808,
+      "rewards/check_winston_local_func": 0.6086806058883667,
+      "step": 643
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.9006993006993007,
+      "grad_norm": 4.426477450510755e-05,
+      "kl": 0.150390625,
+      "learning_rate": 9.530248566964391e-05,
+      "loss": 0.0001,
+      "step": 644
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 105.25000762939453,
+      "epoch": 0.9020979020979021,
+      "grad_norm": 2.371230122251656e-05,
+      "kl": 0.1513671875,
+      "learning_rate": 9.527662769662841e-05,
+      "loss": 0.0002,
+      "num_tokens": 473141.0,
+      "reward": 2.1837356090545654,
+      "reward_std": 0.00018372414342593402,
+      "rewards/check_gptzero_func": 0.7142857313156128,
+      "rewards/check_perplexity_diff_func": 1.0,
+      "rewards/check_winston_local_func": 0.4694497585296631,
+      "step": 645
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.9034965034965035,
+      "grad_norm": 2.391891150952087e-05,
+      "kl": 0.1513671875,
+      "learning_rate": 9.525070227753834e-05,
+      "loss": 0.0002,
+      "step": 646
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 117.64286041259766,
+      "epoch": 0.9048951048951049,
+      "grad_norm": 1.3380347420537266e-07,
+      "kl": 0.150390625,
+      "learning_rate": 9.522470945099334e-05,
+      "loss": 0.0002,
+      "num_tokens": 494483.0,
+      "reward": 1.8219071626663208,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.5,
+      "rewards/check_perplexity_diff_func": 1.0,
+      "rewards/check_winston_local_func": 0.3219069838523865,
+      "step": 647
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.9062937062937063,
+      "grad_norm": 1.336307086326504e-07,
+      "kl": 0.150390625,
+      "learning_rate": 9.519864925571345e-05,
+      "loss": 0.0002,
+      "step": 648
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 131.0,
+      "epoch": 0.9076923076923077,
+      "grad_norm": 0.00012151914677644382,
+      "kl": 0.11474609375,
+      "learning_rate": 9.517252173051911e-05,
+      "loss": 0.0001,
+      "num_tokens": 518551.0,
+      "reward": 2.6524739265441895,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 1.2142857313156128,
+      "rewards/check_perplexity_diff_func": 0.8928571343421936,
+      "rewards/check_winston_local_func": 0.545330822467804,
+      "step": 649
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.9090909090909091,
+      "grad_norm": 5.8305084675906205e-05,
+      "kl": 0.11474609375,
+      "learning_rate": 9.514632691433107e-05,
+      "loss": 0.0001,
+      "step": 650
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 100.89286041259766,
+      "epoch": 0.9104895104895104,
+      "grad_norm": 5.230987317211898e-07,
+      "kl": 0.2119140625,
+      "learning_rate": 9.512006484617025e-05,
+      "loss": 0.0002,
+      "num_tokens": 538141.0,
+      "reward": 2.5491878986358643,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 1.0,
+      "rewards/check_perplexity_diff_func": 0.9642857313156128,
+      "rewards/check_winston_local_func": 0.5849019885063171,
+      "step": 651
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.9118881118881119,
+      "grad_norm": 5.241995408876847e-07,
+      "kl": 0.2119140625,
+      "learning_rate": 9.509373556515782e-05,
+      "loss": 0.0002,
+      "step": 652
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 87.41072082519531,
+      "epoch": 0.9132867132867133,
+      "grad_norm": 0.00010906936400705836,
+      "kl": 0.201171875,
+      "learning_rate": 9.506733911051502e-05,
+      "loss": 0.0002,
+      "num_tokens": 555810.0,
+      "reward": 2.1223483085632324,
+      "reward_std": 0.00045456583029590547,
+      "rewards/check_gptzero_func": 0.6428571343421936,
+      "rewards/check_perplexity_diff_func": 1.0,
+      "rewards/check_winston_local_func": 0.47949108481407166,
+      "step": 653
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.9146853146853147,
+      "grad_norm": 0.0001088298231648261,
+      "kl": 0.201171875,
+      "learning_rate": 9.50408755215632e-05,
+      "loss": 0.0002,
+      "step": 654
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 108.28572082519531,
+      "epoch": 0.916083916083916,
+      "grad_norm": 4.051397464310879e-07,
+      "kl": 0.16015625,
+      "learning_rate": 9.50143448377237e-05,
+      "loss": 0.0002,
+      "num_tokens": 576240.0,
+      "reward": 2.4261841773986816,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.9285714030265808,
+      "rewards/check_perplexity_diff_func": 0.9642857313156128,
+      "rewards/check_winston_local_func": 0.5333269238471985,
+      "step": 655
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.9174825174825175,
+      "grad_norm": 4.914401353891805e-07,
+      "kl": 0.16015625,
+      "learning_rate": 9.498774709851779e-05,
+      "loss": 0.0002,
+      "step": 656
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 103.35714721679688,
+      "epoch": 0.9188811188811189,
+      "grad_norm": 2.415453772573951e-07,
+      "kl": 0.1845703125,
+      "learning_rate": 9.496108234356666e-05,
+      "loss": 0.0002,
+      "num_tokens": 596032.0,
+      "reward": 2.0929291248321533,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.6428571343421936,
+      "rewards/check_perplexity_diff_func": 1.0,
+      "rewards/check_winston_local_func": 0.4500720500946045,
+      "step": 657
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.9202797202797203,
+      "grad_norm": 2.538209980883487e-07,
+      "kl": 0.1845703125,
+      "learning_rate": 9.493435061259131e-05,
+      "loss": 0.0002,
+      "step": 658
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 108.5714340209961,
+      "epoch": 0.9216783216783216,
+      "grad_norm": 0.0005472818191454855,
+      "kl": 0.15234375,
+      "learning_rate": 9.49075519454125e-05,
+      "loss": 0.0002,
+      "num_tokens": 616806.0,
+      "reward": 2.419811964035034,
+      "reward_std": 0.0021216755267232656,
+      "rewards/check_gptzero_func": 1.0,
+      "rewards/check_perplexity_diff_func": 0.9642857313156128,
+      "rewards/check_winston_local_func": 0.4555261731147766,
+      "step": 659
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.9230769230769231,
+      "grad_norm": 0.000555498769506101,
+      "kl": 0.15234375,
+      "learning_rate": 9.48806863819507e-05,
+      "loss": 0.0002,
+      "step": 660
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 99.50000762939453,
+      "epoch": 0.9244755244755245,
+      "grad_norm": 1.210593242712218e-05,
+      "kl": 0.185546875,
+      "learning_rate": 9.485375396222609e-05,
+      "loss": 0.0002,
+      "num_tokens": 636302.0,
+      "reward": 2.2581381797790527,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.7857142686843872,
+      "rewards/check_perplexity_diff_func": 0.9642857313156128,
+      "rewards/check_winston_local_func": 0.5081380605697632,
+      "step": 661
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.9258741258741259,
+      "grad_norm": 1.2079740491548147e-05,
+      "kl": 0.185546875,
+      "learning_rate": 9.482675472635837e-05,
+      "loss": 0.0002,
+      "step": 662
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 130.8928680419922,
+      "epoch": 0.9272727272727272,
+      "grad_norm": 6.087538896625874e-06,
+      "kl": 0.1640625,
+      "learning_rate": 9.479968871456679e-05,
+      "loss": 0.0002,
+      "num_tokens": 659494.0,
+      "reward": 2.4411208629608154,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 1.0,
+      "rewards/check_perplexity_diff_func": 0.9285714030265808,
+      "rewards/check_winston_local_func": 0.5125490427017212,
+      "step": 663
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.9286713286713286,
+      "grad_norm": 5.91002527093218e-06,
+      "kl": 0.1640625,
+      "learning_rate": 9.477255596717012e-05,
+      "loss": 0.0002,
+      "step": 664
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 106.78572082519531,
+      "epoch": 0.9300699300699301,
+      "grad_norm": 0.0006001697890594295,
+      "kl": 0.1572265625,
+      "learning_rate": 9.474535652458647e-05,
+      "loss": 0.0002,
+      "num_tokens": 679806.0,
+      "reward": 1.9322819709777832,
+      "reward_std": 0.0051432885229587555,
+      "rewards/check_gptzero_func": 0.5714285969734192,
+      "rewards/check_perplexity_diff_func": 0.9642857313156128,
+      "rewards/check_winston_local_func": 0.39656758308410645,
+      "step": 665
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.9314685314685315,
+      "grad_norm": 0.0006059012217339499,
+      "kl": 0.1572265625,
+      "learning_rate": 9.471809042733336e-05,
+      "loss": 0.0002,
+      "step": 666
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 93.46428680419922,
+      "epoch": 0.9328671328671329,
+      "grad_norm": 2.199744146458813e-06,
+      "kl": 0.197265625,
+      "learning_rate": 9.469075771602755e-05,
+      "loss": 0.0002,
+      "num_tokens": 698264.0,
+      "reward": 2.203263282775879,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.7142857313156128,
+      "rewards/check_perplexity_diff_func": 0.9642857313156128,
+      "rewards/check_winston_local_func": 0.524691641330719,
+      "step": 667
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.9342657342657342,
+      "grad_norm": 2.363657066653465e-06,
+      "kl": 0.197265625,
+      "learning_rate": 9.46633584313851e-05,
+      "loss": 0.0002,
+      "step": 668
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 121.28572082519531,
+      "epoch": 0.9356643356643357,
+      "grad_norm": 1.1463012847374795e-06,
+      "kl": 0.138671875,
+      "learning_rate": 9.463589261422116e-05,
+      "loss": 0.0001,
+      "num_tokens": 720488.0,
+      "reward": 2.178800582885742,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.8571428656578064,
+      "rewards/check_perplexity_diff_func": 0.9285714030265808,
+      "rewards/check_winston_local_func": 0.3930862247943878,
+      "step": 669
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.9370629370629371,
+      "grad_norm": 1.142739230420817e-06,
+      "kl": 0.138671875,
+      "learning_rate": 9.460836030545007e-05,
+      "loss": 0.0001,
+      "step": 670
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 110.00000762939453,
+      "epoch": 0.9384615384615385,
+      "grad_norm": 1.5883057563648785e-06,
+      "kl": 0.1748046875,
+      "learning_rate": 9.458076154608515e-05,
+      "loss": 0.0002,
+      "num_tokens": 741268.0,
+      "reward": 2.4054763317108154,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 1.0714285373687744,
+      "rewards/check_perplexity_diff_func": 0.8928571343421936,
+      "rewards/check_winston_local_func": 0.44119033217430115,
+      "step": 671
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.9398601398601398,
+      "grad_norm": 1.485387924830463e-06,
+      "kl": 0.1748046875,
+      "learning_rate": 9.455309637723875e-05,
+      "loss": 0.0002,
+      "step": 672
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 76.0,
+      "epoch": 0.9412587412587412,
+      "grad_norm": 5.476513075735238e-07,
+      "kl": 0.275390625,
+      "learning_rate": 9.452536484012212e-05,
+      "loss": 0.0003,
+      "num_tokens": 757900.0,
+      "reward": 2.1300675868988037,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.5714285969734192,
+      "rewards/check_perplexity_diff_func": 1.0,
+      "rewards/check_winston_local_func": 0.558638870716095,
+      "step": 673
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.9426573426573427,
+      "grad_norm": 5.390848738144119e-07,
+      "kl": 0.275390625,
+      "learning_rate": 9.44975669760454e-05,
+      "loss": 0.0003,
+      "step": 674
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 85.42857360839844,
+      "epoch": 0.9440559440559441,
+      "grad_norm": 0.09102872158611763,
+      "kl": 0.2197265625,
+      "learning_rate": 9.446970282641754e-05,
+      "loss": 0.0018,
+      "num_tokens": 775790.0,
+      "reward": 2.0888733863830566,
+      "reward_std": 0.06202990189194679,
+      "rewards/check_gptzero_func": 0.5357142686843872,
+      "rewards/check_perplexity_diff_func": 1.0,
+      "rewards/check_winston_local_func": 0.5531590580940247,
+      "step": 675
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.9454545454545454,
+      "grad_norm": 0.05645096724429603,
+      "kl": 0.220703125,
+      "learning_rate": 9.444177243274618e-05,
+      "loss": 0.0003,
+      "step": 676
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 109.25000762939453,
+      "epoch": 0.9468531468531468,
+      "grad_norm": 8.293674316919555e-06,
+      "kl": 0.1435546875,
+      "learning_rate": 9.441377583663768e-05,
+      "loss": 0.0001,
+      "num_tokens": 796550.0,
+      "reward": 2.231931209564209,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.8571428656578064,
+      "rewards/check_perplexity_diff_func": 0.9285714030265808,
+      "rewards/check_winston_local_func": 0.44621697068214417,
+      "step": 677
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.9482517482517483,
+      "grad_norm": 1.3063887978182609e-05,
+      "kl": 0.1435546875,
+      "learning_rate": 9.438571307979704e-05,
+      "loss": 0.0001,
+      "step": 678
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 116.10714721679688,
+      "epoch": 0.9496503496503497,
+      "grad_norm": 8.55167943307393e-05,
+      "kl": 0.171875,
+      "learning_rate": 9.435758420402778e-05,
+      "loss": 0.0002,
+      "num_tokens": 818208.0,
+      "reward": 2.5719223022460938,
+      "reward_std": 0.00022740935673937201,
+      "rewards/check_gptzero_func": 1.0714285373687744,
+      "rewards/check_perplexity_diff_func": 0.9642857313156128,
+      "rewards/check_winston_local_func": 0.5362080335617065,
+      "step": 679
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.951048951048951,
+      "grad_norm": 7.003771025309948e-05,
+      "kl": 0.171875,
+      "learning_rate": 9.43293892512319e-05,
+      "loss": 0.0002,
+      "step": 680
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 134.75,
+      "epoch": 0.9524475524475524,
+      "grad_norm": 0.000302463889717664,
+      "kl": 0.1435546875,
+      "learning_rate": 9.430112826340987e-05,
+      "loss": 0.0001,
+      "num_tokens": 841930.0,
+      "reward": 2.399691581726074,
+      "reward_std": 0.0017504910938441753,
+      "rewards/check_gptzero_func": 1.0714285373687744,
+      "rewards/check_perplexity_diff_func": 0.8928571343421936,
+      "rewards/check_winston_local_func": 0.4354057312011719,
+      "step": 681
+    },
+    {
+      "clip_ratio": 9.007386688608676e-05,
+      "epoch": 0.9538461538461539,
+      "grad_norm": 0.00031004535575776196,
+      "kl": 0.1435546875,
+      "learning_rate": 9.42728012826605e-05,
+      "loss": 0.0001,
+      "step": 682
+    },
+    {
+      "clip_ratio": 0.00011586143227759749,
+      "completion_length": 117.26786041259766,
+      "epoch": 0.9552447552447553,
+      "grad_norm": 0.0023255881114160465,
+      "kl": 0.146484375,
+      "learning_rate": 9.424440835118094e-05,
+      "loss": 0.0003,
+      "num_tokens": 863539.0,
+      "reward": 2.1542186737060547,
+      "reward_std": 0.02600925788283348,
+      "rewards/check_gptzero_func": 0.7857142686843872,
+      "rewards/check_perplexity_diff_func": 0.9464285969734192,
+      "rewards/check_winston_local_func": 0.42207565903663635,
+      "step": 683
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.9566433566433566,
+      "grad_norm": 0.003353044551999646,
+      "kl": 0.146484375,
+      "learning_rate": 9.421594951126653e-05,
+      "loss": 0.0003,
+      "step": 684
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 130.75,
+      "epoch": 0.958041958041958,
+      "grad_norm": 3.4853626956300013e-07,
+      "kl": 0.1376953125,
+      "learning_rate": 9.418742480531085e-05,
+      "loss": 0.0001,
+      "num_tokens": 886503.0,
+      "reward": 2.1976943016052246,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.7857142686843872,
+      "rewards/check_perplexity_diff_func": 0.9285714030265808,
+      "rewards/check_winston_local_func": 0.4834086000919342,
+      "step": 685
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.9594405594405594,
+      "grad_norm": 4.041143328184125e-07,
+      "kl": 0.1376953125,
+      "learning_rate": 9.415883427580556e-05,
+      "loss": 0.0001,
+      "step": 686
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 88.14286041259766,
+      "epoch": 0.9608391608391609,
+      "grad_norm": 0.0002925937022236145,
+      "kl": 0.1884765625,
+      "learning_rate": 9.413017796534038e-05,
+      "loss": 0.0002,
+      "num_tokens": 904263.0,
+      "reward": 2.0918164253234863,
+      "reward_std": 0.0009377764072269201,
+      "rewards/check_gptzero_func": 0.6428571343421936,
+      "rewards/check_perplexity_diff_func": 1.0,
+      "rewards/check_winston_local_func": 0.448959082365036,
+      "step": 687
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.9622377622377623,
+      "grad_norm": 0.00027650769429382713,
+      "kl": 0.1884765625,
+      "learning_rate": 9.410145591660301e-05,
+      "loss": 0.0002,
+      "step": 688
+    },
+    {
+      "clip_ratio": 0.00036945813917554915,
+      "completion_length": 111.60714721679688,
+      "epoch": 0.9636363636363636,
+      "grad_norm": 0.00019032746086675872,
+      "kl": 0.154296875,
+      "learning_rate": 9.407266817237911e-05,
+      "loss": 0.0002,
+      "num_tokens": 925343.0,
+      "reward": 2.2334210872650146,
+      "reward_std": 0.00015559412713628262,
+      "rewards/check_gptzero_func": 0.7142857313156128,
+      "rewards/check_perplexity_diff_func": 0.9642857313156128,
+      "rewards/check_winston_local_func": 0.5548495650291443,
+      "step": 689
+    },
+    {
+      "clip_ratio": 0.00024630542611703277,
+      "epoch": 0.965034965034965,
+      "grad_norm": 0.00013578009050933864,
+      "kl": 0.1533203125,
+      "learning_rate": 9.404381477555217e-05,
+      "loss": 0.0002,
+      "step": 690
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 113.14286041259766,
+      "epoch": 0.9664335664335665,
+      "grad_norm": 1.7240378740659336e-05,
+      "kl": 0.16015625,
+      "learning_rate": 9.401489576910349e-05,
+      "loss": 0.0002,
+      "num_tokens": 946805.0,
+      "reward": 2.417461395263672,
+      "reward_std": 0.00010864253272302449,
+      "rewards/check_gptzero_func": 0.9285714030265808,
+      "rewards/check_perplexity_diff_func": 0.9642857313156128,
+      "rewards/check_winston_local_func": 0.5246042609214783,
+      "step": 691
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.9678321678321679,
+      "grad_norm": 1.6290593947397482e-05,
+      "kl": 0.16015625,
+      "learning_rate": 9.398591119611211e-05,
+      "loss": 0.0002,
+      "step": 692
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 89.4464340209961,
+      "epoch": 0.9692307692307692,
+      "grad_norm": 0.0003060987829802528,
+      "kl": 0.19140625,
+      "learning_rate": 9.395686109975474e-05,
+      "loss": 0.0002,
+      "num_tokens": 965064.0,
+      "reward": 2.075838804244995,
+      "reward_std": 0.0029447702690958977,
+      "rewards/check_gptzero_func": 0.6428571343421936,
+      "rewards/check_perplexity_diff_func": 1.0,
+      "rewards/check_winston_local_func": 0.43298134207725525,
+      "step": 693
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.9706293706293706,
+      "grad_norm": 0.001874059602241207,
+      "kl": 0.19140625,
+      "learning_rate": 9.392774552330567e-05,
+      "loss": 0.0002,
+      "step": 694
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 126.10714721679688,
+      "epoch": 0.972027972027972,
+      "grad_norm": 8.509194894061606e-06,
+      "kl": 0.1640625,
+      "learning_rate": 9.38985645101368e-05,
+      "loss": 0.0002,
+      "num_tokens": 987280.0,
+      "reward": 2.1274285316467285,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.7857142686843872,
+      "rewards/check_perplexity_diff_func": 0.9285714030265808,
+      "rewards/check_winston_local_func": 0.41314274072647095,
+      "step": 695
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.9734265734265735,
+      "grad_norm": 8.720610779885554e-06,
+      "kl": 0.1640625,
+      "learning_rate": 9.386931810371742e-05,
+      "loss": 0.0002,
+      "step": 696
+    },
+    {
+      "clip_ratio": 0.00018198363250121474,
+      "completion_length": 104.37500762939453,
+      "epoch": 0.9748251748251748,
+      "grad_norm": 0.006044271030051565,
+      "kl": 0.162109375,
+      "learning_rate": 9.38400063476143e-05,
+      "loss": 0.0001,
+      "num_tokens": 1007329.0,
+      "reward": 2.1394481658935547,
+      "reward_std": 0.008249713107943535,
+      "rewards/check_gptzero_func": 0.7857142686843872,
+      "rewards/check_perplexity_diff_func": 0.9642857313156128,
+      "rewards/check_winston_local_func": 0.38944822549819946,
+      "step": 697
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.9762237762237762,
+      "grad_norm": 0.0001835404667097808,
+      "kl": 0.162109375,
+      "learning_rate": 9.381062928549153e-05,
+      "loss": 0.0001,
+      "step": 698
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 121.28572082519531,
+      "epoch": 0.9776223776223776,
+      "grad_norm": 2.5914502760266237e-06,
+      "kl": 0.1494140625,
+      "learning_rate": 9.378118696111047e-05,
+      "loss": 0.0001,
+      "num_tokens": 1029293.0,
+      "reward": 2.0580966472625732,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.6428571343421936,
+      "rewards/check_perplexity_diff_func": 1.0,
+      "rewards/check_winston_local_func": 0.4152393937110901,
+      "step": 699
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.9790209790209791,
+      "grad_norm": 2.773516504511405e-06,
+      "kl": 0.1494140625,
+      "learning_rate": 9.375167941832973e-05,
+      "loss": 0.0001,
+      "step": 700
+    },
+    {
+      "clip_ratio": 0.00013789300282951444,
+      "completion_length": 93.75000762939453,
+      "epoch": 0.9804195804195804,
+      "grad_norm": 0.013670676452353387,
+      "kl": 0.2041015625,
+      "learning_rate": 9.372210670110505e-05,
+      "loss": -0.0004,
+      "num_tokens": 1048027.0,
+      "reward": 2.084160327911377,
+      "reward_std": 0.011760223656892776,
+      "rewards/check_gptzero_func": 0.6428571343421936,
+      "rewards/check_perplexity_diff_func": 0.9464285969734192,
+      "rewards/check_winston_local_func": 0.49487441778182983,
+      "step": 701
+    },
+    {
+      "clip_ratio": 0.00013789300282951444,
+      "epoch": 0.9818181818181818,
+      "grad_norm": 0.01262284842044345,
+      "kl": 0.2041015625,
+      "learning_rate": 9.369246885348926e-05,
+      "loss": -0.0005,
+      "step": 702
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 113.28572082519531,
+      "epoch": 0.9832167832167832,
+      "grad_norm": 2.9830896158632583e-05,
+      "kl": 0.15625,
+      "learning_rate": 9.366276591963221e-05,
+      "loss": 0.0002,
+      "num_tokens": 1068897.0,
+      "reward": 2.2253122329711914,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.7857142686843872,
+      "rewards/check_perplexity_diff_func": 1.0,
+      "rewards/check_winston_local_func": 0.4395977854728699,
+      "step": 703
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.9846153846153847,
+      "grad_norm": 2.8538602263117158e-05,
+      "kl": 0.15625,
+      "learning_rate": 9.363299794378073e-05,
+      "loss": 0.0002,
+      "step": 704
+    },
+    {
+      "clip_ratio": 0.00022251892369240522,
+      "completion_length": 112.42857360839844,
+      "epoch": 0.986013986013986,
+      "grad_norm": 0.013827349944974443,
+      "kl": 0.1552734375,
+      "learning_rate": 9.360316497027848e-05,
+      "loss": 0.0006,
+      "num_tokens": 1089715.0,
+      "reward": 2.2796030044555664,
+      "reward_std": 0.036380112171173096,
+      "rewards/check_gptzero_func": 0.7857142686843872,
+      "rewards/check_perplexity_diff_func": 0.9642857313156128,
+      "rewards/check_winston_local_func": 0.529603123664856,
+      "step": 705
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.9874125874125874,
+      "grad_norm": 0.013400518475433424,
+      "kl": 0.1552734375,
+      "learning_rate": 9.357326704356602e-05,
+      "loss": 0.0005,
+      "step": 706
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 96.26786041259766,
+      "epoch": 0.9888111888111888,
+      "grad_norm": 0.013500774785978116,
+      "kl": 0.19140625,
+      "learning_rate": 9.35433042081806e-05,
+      "loss": 0.0003,
+      "num_tokens": 1108898.0,
+      "reward": 2.1743431091308594,
+      "reward_std": 0.025075625628232956,
+      "rewards/check_gptzero_func": 0.7857142686843872,
+      "rewards/check_perplexity_diff_func": 0.9464285969734192,
+      "rewards/check_winston_local_func": 0.442200243473053,
+      "step": 707
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.9902097902097902,
+      "grad_norm": 0.004775981640901926,
+      "kl": 0.1904296875,
+      "learning_rate": 9.351327650875624e-05,
+      "loss": 0.0002,
+      "step": 708
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 106.53572082519531,
+      "epoch": 0.9916083916083916,
+      "grad_norm": 5.729650858248644e-07,
+      "kl": 0.1650390625,
+      "learning_rate": 9.348318399002347e-05,
+      "loss": 0.0002,
+      "num_tokens": 1129126.0,
+      "reward": 2.423046588897705,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.9285714030265808,
+      "rewards/check_perplexity_diff_func": 1.0,
+      "rewards/check_winston_local_func": 0.49447494745254517,
+      "step": 709
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 0.993006993006993,
+      "grad_norm": 5.262172579879396e-07,
+      "kl": 0.1650390625,
+      "learning_rate": 9.345302669680946e-05,
+      "loss": 0.0002,
+      "step": 710
+    },
+    {
+      "clip_ratio": 0.00021829926117789,
+      "completion_length": 120.46429443359375,
+      "epoch": 0.9944055944055944,
+      "grad_norm": 0.05104704112053445,
+      "kl": 0.17578125,
+      "learning_rate": 9.342280467403788e-05,
+      "loss": -0.0017,
+      "num_tokens": 1151220.0,
+      "reward": 2.2311315536499023,
+      "reward_std": 0.04048382490873337,
+      "rewards/check_gptzero_func": 0.8214285969734192,
+      "rewards/check_perplexity_diff_func": 0.9464285969734192,
+      "rewards/check_winston_local_func": 0.4632743299007416,
+      "step": 711
+    },
+    {
+      "clip_ratio": 0.00010004002251662314,
+      "epoch": 0.9958041958041958,
+      "grad_norm": 0.03334333430793214,
+      "kl": 0.169921875,
+      "learning_rate": 9.339251796672877e-05,
+      "loss": -0.0022,
+      "step": 712
+    },
+    {
+      "clip_ratio": 0.00011548677866812795,
+      "completion_length": 106.26786041259766,
+      "epoch": 0.9972027972027973,
+      "grad_norm": 8.312891203940763e-05,
+      "kl": 0.15234375,
+      "learning_rate": 9.336216661999857e-05,
+      "loss": 0.0002,
+      "num_tokens": 1171757.0,
+      "reward": 2.4376866817474365,
+      "reward_std": 0.00044757546856999397,
+      "rewards/check_gptzero_func": 0.9285714030265808,
+      "rewards/check_perplexity_diff_func": 0.9642857313156128,
+      "rewards/check_winston_local_func": 0.5448295474052429,
+      "step": 713
+    },
+    {
+      "clip_ratio": 0.00011548677866812795,
+      "epoch": 0.9986013986013986,
+      "grad_norm": 7.041724475492408e-05,
+      "kl": 0.15234375,
+      "learning_rate": 9.333175067905997e-05,
+      "loss": 0.0002,
+      "step": 714
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 127.00000762939453,
+      "epoch": 1.0013986013986014,
+      "grad_norm": 7.04944892486657e-07,
+      "kl": 0.142578125,
+      "learning_rate": 9.330127018922194e-05,
+      "loss": 0.0001,
+      "num_tokens": 1194243.0,
+      "reward": 1.8460028171539307,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.5,
+      "rewards/check_perplexity_diff_func": 1.0,
+      "rewards/check_winston_local_func": 0.34600263833999634,
+      "step": 715
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 1.0027972027972027,
+      "grad_norm": 6.628698629011792e-07,
+      "kl": 0.142578125,
+      "learning_rate": 9.327072519588954e-05,
+      "loss": 0.0001,
+      "step": 716
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 94.66072082519531,
+      "epoch": 1.0041958041958041,
+      "grad_norm": 0.00044081574153375096,
+      "kl": 0.2099609375,
+      "learning_rate": 9.324011574456397e-05,
+      "loss": 0.0002,
+      "num_tokens": 1213238.0,
+      "reward": 2.0919501781463623,
+      "reward_std": 0.00025218268274329603,
+      "rewards/check_gptzero_func": 0.5714285969734192,
+      "rewards/check_perplexity_diff_func": 0.9285714030265808,
+      "rewards/check_winston_local_func": 0.5919499397277832,
+      "step": 717
+    },
+    {
+      "clip_ratio": 0.00019868866365868598,
+      "epoch": 1.0055944055944055,
+      "grad_norm": 0.000549077840173661,
+      "kl": 0.2109375,
+      "learning_rate": 9.320944188084242e-05,
+      "loss": 0.0002,
+      "step": 718
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 84.85714721679688,
+      "epoch": 1.006993006993007,
+      "grad_norm": 0.005163027828412511,
+      "kl": 0.1884765625,
+      "learning_rate": 9.317870365041804e-05,
+      "loss": 0.0002,
+      "num_tokens": 1230630.0,
+      "reward": 2.2233378887176514,
+      "reward_std": 0.02511097863316536,
+      "rewards/check_gptzero_func": 0.7857142686843872,
+      "rewards/check_perplexity_diff_func": 0.9821428656578064,
+      "rewards/check_winston_local_func": 0.45548057556152344,
+      "step": 719
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 1.0083916083916085,
+      "grad_norm": 0.00444116748229232,
+      "kl": 0.1884765625,
+      "learning_rate": 9.314790109907984e-05,
+      "loss": 0.0002,
+      "step": 720
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 80.75,
+      "epoch": 1.0097902097902098,
+      "grad_norm": 8.668316800887643e-07,
+      "kl": 0.216796875,
+      "learning_rate": 9.31170342727127e-05,
+      "loss": 0.0002,
+      "num_tokens": 1248084.0,
+      "reward": 2.440389394760132,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.9285714030265808,
+      "rewards/check_perplexity_diff_func": 1.0,
+      "rewards/check_winston_local_func": 0.5118179321289062,
+      "step": 721
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 1.0111888111888112,
+      "grad_norm": 7.749843514631877e-07,
+      "kl": 0.216796875,
+      "learning_rate": 9.30861032172972e-05,
+      "loss": 0.0002,
+      "step": 722
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 109.53572082519531,
+      "epoch": 1.0125874125874126,
+      "grad_norm": 1.4806859938034535e-05,
+      "kl": 0.19140625,
+      "learning_rate": 9.305510797890962e-05,
+      "loss": 0.0002,
+      "num_tokens": 1269030.0,
+      "reward": 2.161181688308716,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.7142857313156128,
+      "rewards/check_perplexity_diff_func": 1.0,
+      "rewards/check_winston_local_func": 0.44689592719078064,
+      "step": 723
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 1.013986013986014,
+      "grad_norm": 1.600117516634148e-05,
+      "kl": 0.19140625,
+      "learning_rate": 9.302404860372185e-05,
+      "loss": 0.0002,
+      "step": 724
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 105.42857360839844,
+      "epoch": 1.0153846153846153,
+      "grad_norm": 0.15558383270036835,
+      "kl": 0.21484375,
+      "learning_rate": 9.299292513800129e-05,
+      "loss": 0.05,
+      "num_tokens": 1289470.0,
+      "reward": 2.1427464485168457,
+      "reward_std": 0.07507836818695068,
+      "rewards/check_gptzero_func": 0.6785714030265808,
+      "rewards/check_perplexity_diff_func": 0.9285714030265808,
+      "rewards/check_winston_local_func": 0.535603404045105,
+      "step": 725
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 1.0167832167832167,
+      "grad_norm": 0.06803559596877588,
+      "kl": 0.2353515625,
+      "learning_rate": 9.296173762811085e-05,
+      "loss": 0.0475,
+      "step": 726
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 117.10714721679688,
+      "epoch": 1.018181818181818,
+      "grad_norm": 1.6687299566663977e-05,
+      "kl": 0.14453125,
+      "learning_rate": 9.293048612050883e-05,
+      "loss": 0.0001,
+      "num_tokens": 1311104.0,
+      "reward": 1.9664360284805298,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.7142857313156128,
+      "rewards/check_perplexity_diff_func": 0.8928571343421936,
+      "rewards/check_winston_local_func": 0.35929304361343384,
+      "step": 727
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 1.0195804195804197,
+      "grad_norm": 1.8557887240310947e-05,
+      "kl": 0.14453125,
+      "learning_rate": 9.289917066174886e-05,
+      "loss": 0.0001,
+      "step": 728
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 93.42857360839844,
+      "epoch": 1.020979020979021,
+      "grad_norm": 1.0271883224976578e-06,
+      "kl": 0.2314453125,
+      "learning_rate": 9.286779129847984e-05,
+      "loss": 0.0002,
+      "num_tokens": 1329690.0,
+      "reward": 2.0715172290802,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.5714285969734192,
+      "rewards/check_perplexity_diff_func": 1.0,
+      "rewards/check_winston_local_func": 0.5000885725021362,
+      "step": 729
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 1.0223776223776224,
+      "grad_norm": 1.0026902118997668e-06,
+      "kl": 0.2314453125,
+      "learning_rate": 9.283634807744586e-05,
+      "loss": 0.0002,
+      "step": 730
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 127.00000762939453,
+      "epoch": 1.0237762237762238,
+      "grad_norm": 1.674477493760208e-06,
+      "kl": 0.169921875,
+      "learning_rate": 9.280484104548615e-05,
+      "loss": 0.0002,
+      "num_tokens": 1352330.0,
+      "reward": 2.3377161026000977,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 1.0,
+      "rewards/check_perplexity_diff_func": 0.8928571343421936,
+      "rewards/check_winston_local_func": 0.4448588192462921,
+      "step": 731
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 1.0251748251748252,
+      "grad_norm": 1.927135056596767e-06,
+      "kl": 0.169921875,
+      "learning_rate": 9.277327024953496e-05,
+      "loss": 0.0002,
+      "step": 732
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 96.10714721679688,
+      "epoch": 1.0265734265734265,
+      "grad_norm": 1.21032862101872e-06,
+      "kl": 0.1689453125,
+      "learning_rate": 9.274163573662157e-05,
+      "loss": 0.0002,
+      "num_tokens": 1371384.0,
+      "reward": 2.437558889389038,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.9285714030265808,
+      "rewards/check_perplexity_diff_func": 1.0,
+      "rewards/check_winston_local_func": 0.508987307548523,
+      "step": 733
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 1.027972027972028,
+      "grad_norm": 1.7462301887618669e-06,
+      "kl": 0.1689453125,
+      "learning_rate": 9.270993755387015e-05,
+      "loss": 0.0002,
+      "step": 734
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 89.78572082519531,
+      "epoch": 1.0293706293706293,
+      "grad_norm": 0.0004304814492353597,
+      "kl": 0.2177734375,
+      "learning_rate": 9.26781757484997e-05,
+      "loss": 0.0002,
+      "num_tokens": 1389928.0,
+      "reward": 2.4899582862854004,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.8571428656578064,
+      "rewards/check_perplexity_diff_func": 1.0,
+      "rewards/check_winston_local_func": 0.6328152418136597,
+      "step": 735
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 1.0307692307692307,
+      "grad_norm": 0.00046915841230749046,
+      "kl": 0.2177734375,
+      "learning_rate": 9.264635036782405e-05,
+      "loss": 0.0002,
+      "step": 736
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 117.4464340209961,
+      "epoch": 1.0321678321678323,
+      "grad_norm": 0.003036344469002795,
+      "kl": 0.142578125,
+      "learning_rate": 9.261446145925167e-05,
+      "loss": -0.0041,
+      "num_tokens": 1411475.0,
+      "reward": 2.17564058303833,
+      "reward_std": 0.030071333050727844,
+      "rewards/check_gptzero_func": 0.8571428656578064,
+      "rewards/check_perplexity_diff_func": 0.9821428656578064,
+      "rewards/check_winston_local_func": 0.33635467290878296,
+      "step": 737
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 1.0335664335664336,
+      "grad_norm": 0.0027126180636015186,
+      "kl": 0.1435546875,
+      "learning_rate": 9.258250907028572e-05,
+      "loss": -0.0041,
+      "step": 738
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 104.73214721679688,
+      "epoch": 1.034965034965035,
+      "grad_norm": 0.0544058478741893,
+      "kl": 0.2080078125,
+      "learning_rate": 9.255049324852388e-05,
+      "loss": 0.0002,
+      "num_tokens": 1431764.0,
+      "reward": 2.090238571166992,
+      "reward_std": 0.08559178560972214,
+      "rewards/check_gptzero_func": 0.6785714030265808,
+      "rewards/check_perplexity_diff_func": 0.9821428656578064,
+      "rewards/check_winston_local_func": 0.42952439188957214,
+      "step": 739
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 1.0363636363636364,
+      "grad_norm": 0.006960547191239059,
+      "kl": 0.205078125,
+      "learning_rate": 9.251841404165835e-05,
+      "loss": 0.0001,
+      "step": 740
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 88.33928680419922,
+      "epoch": 1.0377622377622377,
+      "grad_norm": 0.00033693931372504076,
+      "kl": 0.185546875,
+      "learning_rate": 9.248627149747573e-05,
+      "loss": 0.0002,
+      "num_tokens": 1449951.0,
+      "reward": 2.0378341674804688,
+      "reward_std": 0.0007935017929412425,
+      "rewards/check_gptzero_func": 0.6428571343421936,
+      "rewards/check_perplexity_diff_func": 0.9285714030265808,
+      "rewards/check_winston_local_func": 0.4664055407047272,
+      "step": 741
+    },
+    {
+      "clip_ratio": 0.00013215278158895671,
+      "epoch": 1.0391608391608391,
+      "grad_norm": 0.00034490140909119695,
+      "kl": 0.185546875,
+      "learning_rate": 9.245406566385697e-05,
+      "loss": 0.0002,
+      "step": 742
+    },
+    {
+      "clip_ratio": 0.00015926103515084833,
+      "completion_length": 107.80357360839844,
+      "epoch": 1.0405594405594405,
+      "grad_norm": 0.013332167554109667,
+      "kl": 0.1865234375,
+      "learning_rate": 9.242179658877734e-05,
+      "loss": 0.0001,
+      "num_tokens": 1470538.0,
+      "reward": 2.4016566276550293,
+      "reward_std": 0.02674621157348156,
+      "rewards/check_gptzero_func": 0.8571428656578064,
+      "rewards/check_perplexity_diff_func": 0.9821428656578064,
+      "rewards/check_winston_local_func": 0.5623708963394165,
+      "step": 743
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 1.0419580419580419,
+      "grad_norm": 0.0029520677726348676,
+      "kl": 0.2001953125,
+      "learning_rate": 9.238946432030626e-05,
+      "loss": 0.0001,
+      "step": 744
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 83.35714721679688,
+      "epoch": 1.0433566433566432,
+      "grad_norm": 0.0005995187289984499,
+      "kl": 0.2255859375,
+      "learning_rate": 9.235706890660733e-05,
+      "loss": 0.0002,
+      "num_tokens": 1487784.0,
+      "reward": 2.067596435546875,
+      "reward_std": 0.00029761545010842383,
+      "rewards/check_gptzero_func": 0.5714285969734192,
+      "rewards/check_perplexity_diff_func": 0.9642857313156128,
+      "rewards/check_winston_local_func": 0.5318821668624878,
+      "step": 745
+    },
+    {
+      "clip_ratio": 0.0002088554756483063,
+      "epoch": 1.0447552447552448,
+      "grad_norm": 0.0005979581476299124,
+      "kl": 0.2265625,
+      "learning_rate": 9.232461039593819e-05,
+      "loss": 0.0002,
+      "step": 746
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 122.33929443359375,
+      "epoch": 1.0461538461538462,
+      "grad_norm": 1.3373485214002519e-05,
+      "kl": 0.1455078125,
+      "learning_rate": 9.229208883665047e-05,
+      "loss": 0.0001,
+      "num_tokens": 1510355.0,
+      "reward": 1.986084222793579,
+      "reward_std": 0.024681033566594124,
+      "rewards/check_gptzero_func": 0.7142857313156128,
+      "rewards/check_perplexity_diff_func": 0.8392857313156128,
+      "rewards/check_winston_local_func": 0.43251270055770874,
+      "step": 747
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 1.0475524475524476,
+      "grad_norm": 0.005145688764176549,
+      "kl": 0.1455078125,
+      "learning_rate": 9.225950427718975e-05,
+      "loss": 0.0001,
+      "step": 748
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 118.75000762939453,
+      "epoch": 1.048951048951049,
+      "grad_norm": 2.1467569659911603e-06,
+      "kl": 0.142578125,
+      "learning_rate": 9.22268567660954e-05,
+      "loss": 0.0001,
+      "num_tokens": 1532269.0,
+      "reward": 2.1966254711151123,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.7857142686843872,
+      "rewards/check_perplexity_diff_func": 1.0,
+      "rewards/check_winston_local_func": 0.4109109342098236,
+      "step": 749
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 1.0503496503496503,
+      "grad_norm": 2.040002552625244e-06,
+      "kl": 0.142578125,
+      "learning_rate": 9.219414635200065e-05,
+      "loss": 0.0001,
+      "step": 750
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 103.46428680419922,
+      "epoch": 1.0517482517482517,
+      "grad_norm": 0.0015423150610452264,
+      "kl": 0.2177734375,
+      "learning_rate": 9.216137308363233e-05,
+      "loss": 0.0002,
+      "num_tokens": 1552375.0,
+      "reward": 2.8162217140197754,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 1.3571428060531616,
+      "rewards/check_perplexity_diff_func": 1.0,
+      "rewards/check_winston_local_func": 0.45907872915267944,
+      "step": 751
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 1.053146853146853,
+      "grad_norm": 0.000923513262354566,
+      "kl": 0.2158203125,
+      "learning_rate": 9.2128537009811e-05,
+      "loss": 0.0002,
+      "step": 752
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 92.39286041259766,
+      "epoch": 1.0545454545454545,
+      "grad_norm": 1.100033680537667e-06,
+      "kl": 0.1689453125,
+      "learning_rate": 9.209563817945072e-05,
+      "loss": 0.0002,
+      "num_tokens": 1570845.0,
+      "reward": 1.9652879238128662,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.5714285969734192,
+      "rewards/check_perplexity_diff_func": 1.0,
+      "rewards/check_winston_local_func": 0.39385929703712463,
+      "step": 753
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 1.055944055944056,
+      "grad_norm": 1.1228592596360449e-06,
+      "kl": 0.1689453125,
+      "learning_rate": 9.206267664155907e-05,
+      "loss": 0.0002,
+      "step": 754
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 156.42857360839844,
+      "epoch": 1.0573426573426574,
+      "grad_norm": 2.057149136687796e-07,
+      "kl": 0.1171875,
+      "learning_rate": 9.202965244523695e-05,
+      "loss": 0.0001,
+      "num_tokens": 1596971.0,
+      "reward": 2.463409185409546,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 1.0714285373687744,
+      "rewards/check_perplexity_diff_func": 0.9642857313156128,
+      "rewards/check_winston_local_func": 0.4276948571205139,
+      "step": 755
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 1.0587412587412588,
+      "grad_norm": 1.931817225075422e-07,
+      "kl": 0.1171875,
+      "learning_rate": 9.199656563967875e-05,
+      "loss": 0.0001,
+      "step": 756
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 133.60714721679688,
+      "epoch": 1.0601398601398602,
+      "grad_norm": 7.066669417456737e-05,
+      "kl": 0.126953125,
+      "learning_rate": 9.1963416274172e-05,
+      "loss": 0.0001,
+      "num_tokens": 1621103.0,
+      "reward": 2.4794726371765137,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 1.0,
+      "rewards/check_perplexity_diff_func": 0.9285714030265808,
+      "rewards/check_winston_local_func": 0.5509011149406433,
+      "step": 757
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 1.0615384615384615,
+      "grad_norm": 5.5301588065085076e-05,
+      "kl": 0.126953125,
+      "learning_rate": 9.193020439809747e-05,
+      "loss": 0.0001,
+      "step": 758
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 128.57144165039062,
+      "epoch": 1.062937062937063,
+      "grad_norm": 1.0710398657530842e-06,
+      "kl": 0.1494140625,
+      "learning_rate": 9.189693006092907e-05,
+      "loss": 0.0001,
+      "num_tokens": 1644239.0,
+      "reward": 2.3191192150115967,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.9285714030265808,
+      "rewards/check_perplexity_diff_func": 0.9642857313156128,
+      "rewards/check_winston_local_func": 0.426261842250824,
+      "step": 759
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 1.0643356643356643,
+      "grad_norm": 1.0634008914464338e-06,
+      "kl": 0.1494140625,
+      "learning_rate": 9.186359331223369e-05,
+      "loss": 0.0001,
+      "step": 760
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 101.53572082519531,
+      "epoch": 1.0657342657342657,
+      "grad_norm": 1.016780218352338e-06,
+      "kl": 0.1728515625,
+      "learning_rate": 9.183019420167126e-05,
+      "loss": 0.0002,
+      "num_tokens": 1664379.0,
+      "reward": 2.527435779571533,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 1.0,
+      "rewards/check_perplexity_diff_func": 0.9642857313156128,
+      "rewards/check_winston_local_func": 0.5631500482559204,
+      "step": 761
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 1.067132867132867,
+      "grad_norm": 9.957050622798446e-07,
+      "kl": 0.1728515625,
+      "learning_rate": 9.179673277899456e-05,
+      "loss": 0.0002,
+      "step": 762
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 73.91072082519531,
+      "epoch": 1.0685314685314686,
+      "grad_norm": 0.003192954717479896,
+      "kl": 0.251953125,
+      "learning_rate": 9.176320909404924e-05,
+      "loss": 0.0002,
+      "num_tokens": 1680674.0,
+      "reward": 1.9061808586120605,
+      "reward_std": 0.007731279823929071,
+      "rewards/check_gptzero_func": 0.4285714328289032,
+      "rewards/check_perplexity_diff_func": 1.0,
+      "rewards/check_winston_local_func": 0.4776093661785126,
+      "step": 763
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 1.06993006993007,
+      "grad_norm": 0.0032145560494452733,
+      "kl": 0.251953125,
+      "learning_rate": 9.172962319677363e-05,
+      "loss": 0.0002,
+      "step": 764
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 114.78572082519531,
+      "epoch": 1.0713286713286714,
+      "grad_norm": 1.061638596684205e-06,
+      "kl": 0.15625,
+      "learning_rate": 9.169597513719881e-05,
+      "loss": 0.0002,
+      "num_tokens": 1702162.0,
+      "reward": 2.3026795387268066,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.8571428656578064,
+      "rewards/check_perplexity_diff_func": 0.9642857313156128,
+      "rewards/check_winston_local_func": 0.4812507629394531,
+      "step": 765
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 1.0727272727272728,
+      "grad_norm": 1.172228963353414e-06,
+      "kl": 0.15625,
+      "learning_rate": 9.166226496544839e-05,
+      "loss": 0.0002,
+      "step": 766
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 93.03572082519531,
+      "epoch": 1.0741258741258741,
+      "grad_norm": 2.1039959611311484e-05,
+      "kl": 0.1611328125,
+      "learning_rate": 9.162849273173857e-05,
+      "loss": 0.0002,
+      "num_tokens": 1721000.0,
+      "reward": 2.4015400409698486,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.8571428656578064,
+      "rewards/check_perplexity_diff_func": 1.0,
+      "rewards/check_winston_local_func": 0.5443968772888184,
+      "step": 767
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 1.0755244755244755,
+      "grad_norm": 1.7217131059909884e-05,
+      "kl": 0.1611328125,
+      "learning_rate": 9.159465848637795e-05,
+      "loss": 0.0002,
+      "step": 768
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 116.53572082519531,
+      "epoch": 1.0769230769230769,
+      "grad_norm": 2.8854421633379664e-07,
+      "kl": 0.15625,
+      "learning_rate": 9.156076227976752e-05,
+      "loss": 0.0002,
+      "num_tokens": 1742074.0,
+      "reward": 2.019984006881714,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.6428571343421936,
+      "rewards/check_perplexity_diff_func": 1.0,
+      "rewards/check_winston_local_func": 0.3771267533302307,
+      "step": 769
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 1.0783216783216782,
+      "grad_norm": 2.958786538763937e-07,
+      "kl": 0.15625,
+      "learning_rate": 9.152680416240059e-05,
+      "loss": 0.0002,
+      "step": 770
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 90.17857360839844,
+      "epoch": 1.0797202797202796,
+      "grad_norm": 1.3818538255152332e-07,
+      "kl": 0.2119140625,
+      "learning_rate": 9.149278418486267e-05,
+      "loss": 0.0002,
+      "num_tokens": 1760270.0,
+      "reward": 2.354346752166748,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 1.0,
+      "rewards/check_perplexity_diff_func": 0.9642857313156128,
+      "rewards/check_winston_local_func": 0.3900607228279114,
+      "step": 771
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 1.0811188811188812,
+      "grad_norm": 1.3653596591447927e-07,
+      "kl": 0.2119140625,
+      "learning_rate": 9.145870239783142e-05,
+      "loss": 0.0002,
+      "step": 772
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 113.96429443359375,
+      "epoch": 1.0825174825174826,
+      "grad_norm": 5.440872620013143e-05,
+      "kl": 0.138671875,
+      "learning_rate": 9.142455885207659e-05,
+      "loss": 0.0001,
+      "num_tokens": 1781844.0,
+      "reward": 1.8641180992126465,
+      "reward_std": 0.0003365546581335366,
+      "rewards/check_gptzero_func": 0.5,
+      "rewards/check_perplexity_diff_func": 0.9285714030265808,
+      "rewards/check_winston_local_func": 0.4355466663837433,
+      "step": 773
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 1.083916083916084,
+      "grad_norm": 5.372773059582913e-05,
+      "kl": 0.138671875,
+      "learning_rate": 9.139035359845994e-05,
+      "loss": 0.0001,
+      "step": 774
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 86.67857360839844,
+      "epoch": 1.0853146853146853,
+      "grad_norm": 2.3746587653962933e-07,
+      "kl": 0.1748046875,
+      "learning_rate": 9.135608668793511e-05,
+      "loss": 0.0002,
+      "num_tokens": 1799902.0,
+      "reward": 2.2892978191375732,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.7857142686843872,
+      "rewards/check_perplexity_diff_func": 1.0,
+      "rewards/check_winston_local_func": 0.5035834908485413,
+      "step": 775
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 1.0867132867132867,
+      "grad_norm": 2.3654979764665925e-07,
+      "kl": 0.1748046875,
+      "learning_rate": 9.132175817154763e-05,
+      "loss": 0.0002,
+      "step": 776
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 106.42857360839844,
+      "epoch": 1.088111888111888,
+      "grad_norm": 1.2723426089566611e-05,
+      "kl": 0.1728515625,
+      "learning_rate": 9.128736810043474e-05,
+      "loss": 0.0002,
+      "num_tokens": 1820052.0,
+      "reward": 2.0230119228363037,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.6428571343421936,
+      "rewards/check_perplexity_diff_func": 0.9642857313156128,
+      "rewards/check_winston_local_func": 0.4158690869808197,
+      "step": 777
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 1.0895104895104895,
+      "grad_norm": 1.0458212078617634e-05,
+      "kl": 0.1728515625,
+      "learning_rate": 9.125291652582549e-05,
+      "loss": 0.0002,
+      "step": 778
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 116.71429443359375,
+      "epoch": 1.0909090909090908,
+      "grad_norm": 3.995811344230737e-07,
+      "kl": 0.189453125,
+      "learning_rate": 9.121840349904039e-05,
+      "loss": 0.0002,
+      "num_tokens": 1841842.0,
+      "reward": 2.676417112350464,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 1.1428571939468384,
+      "rewards/check_perplexity_diff_func": 0.9642857313156128,
+      "rewards/check_winston_local_func": 0.5692740082740784,
+      "step": 779
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 1.0923076923076924,
+      "grad_norm": 4.3767238552948305e-07,
+      "kl": 0.189453125,
+      "learning_rate": 9.118382907149165e-05,
+      "loss": 0.0002,
+      "step": 780
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 89.85714721679688,
+      "epoch": 1.0937062937062938,
+      "grad_norm": 9.02834233189963e-08,
+      "kl": 0.154296875,
+      "learning_rate": 9.114919329468282e-05,
+      "loss": 0.0002,
+      "num_tokens": 1859828.0,
+      "reward": 1.9822884798049927,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.5714285969734192,
+      "rewards/check_perplexity_diff_func": 1.0,
+      "rewards/check_winston_local_func": 0.41085976362228394,
+      "step": 781
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 1.0951048951048952,
+      "grad_norm": 8.932188055681656e-08,
+      "kl": 0.154296875,
+      "learning_rate": 9.111449622020893e-05,
+      "loss": 0.0002,
+      "step": 782
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 136.46429443359375,
+      "epoch": 1.0965034965034965,
+      "grad_norm": 1.7587248831698471e-07,
+      "kl": 0.138671875,
+      "learning_rate": 9.107973789975626e-05,
+      "loss": 0.0001,
+      "num_tokens": 1883558.0,
+      "reward": 2.237004280090332,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.9285714030265808,
+      "rewards/check_perplexity_diff_func": 0.9285714030265808,
+      "rewards/check_winston_local_func": 0.37986141443252563,
+      "step": 783
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 1.097902097902098,
+      "grad_norm": 1.7072205199857748e-07,
+      "kl": 0.138671875,
+      "learning_rate": 9.104491838510235e-05,
+      "loss": 0.0001,
+      "step": 784
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 109.42857360839844,
+      "epoch": 1.0993006993006993,
+      "grad_norm": 4.723811406665928e-07,
+      "kl": 0.185546875,
+      "learning_rate": 9.101003772811589e-05,
+      "loss": 0.0002,
+      "num_tokens": 1904436.0,
+      "reward": 2.0636329650878906,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.5714285969734192,
+      "rewards/check_perplexity_diff_func": 1.0,
+      "rewards/check_winston_local_func": 0.492204487323761,
+      "step": 785
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 1.1006993006993007,
+      "grad_norm": 4.813313652081485e-07,
+      "kl": 0.185546875,
+      "learning_rate": 9.097509598075667e-05,
+      "loss": 0.0002,
+      "step": 786
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 100.75000762939453,
+      "epoch": 1.102097902097902,
+      "grad_norm": 1.2624745647996438e-07,
+      "kl": 0.15625,
+      "learning_rate": 9.094009319507547e-05,
+      "loss": 0.0002,
+      "num_tokens": 1924402.0,
+      "reward": 2.1538825035095215,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.7857142686843872,
+      "rewards/check_perplexity_diff_func": 0.9285714030265808,
+      "rewards/check_winston_local_func": 0.43959662318229675,
+      "step": 787
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 1.1034965034965034,
+      "grad_norm": 1.2855084178721202e-07,
+      "kl": 0.15625,
+      "learning_rate": 9.090502942321397e-05,
+      "loss": 0.0002,
+      "step": 788
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 101.26786041259766,
+      "epoch": 1.104895104895105,
+      "grad_norm": 0.002133418465001678,
+      "kl": 0.2109375,
+      "learning_rate": 9.086990471740473e-05,
+      "loss": -0.0001,
+      "num_tokens": 1944079.0,
+      "reward": 2.209216833114624,
+      "reward_std": 0.006096469704061747,
+      "rewards/check_gptzero_func": 0.7142857313156128,
+      "rewards/check_perplexity_diff_func": 0.9285714030265808,
+      "rewards/check_winston_local_func": 0.5663596987724304,
+      "step": 789
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 1.1062937062937064,
+      "grad_norm": 0.003624393686726291,
+      "kl": 0.2109375,
+      "learning_rate": 9.083471912997108e-05,
+      "loss": -0.0001,
+      "step": 790
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 119.39286041259766,
+      "epoch": 1.1076923076923078,
+      "grad_norm": 2.2118816611478268e-07,
+      "kl": 0.1328125,
+      "learning_rate": 9.079947271332706e-05,
+      "loss": 0.0001,
+      "num_tokens": 1965839.0,
+      "reward": 2.1228208541870117,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.7142857313156128,
+      "rewards/check_perplexity_diff_func": 1.0,
+      "rewards/check_winston_local_func": 0.40853506326675415,
+      "step": 791
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 1.1090909090909091,
+      "grad_norm": 2.2277128985396498e-07,
+      "kl": 0.1328125,
+      "learning_rate": 9.076416551997721e-05,
+      "loss": 0.0001,
+      "step": 792
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 88.03572082519531,
+      "epoch": 1.1104895104895105,
+      "grad_norm": 3.3242813310295524e-07,
+      "kl": 0.1640625,
+      "learning_rate": 9.072879760251679e-05,
+      "loss": 0.0002,
+      "num_tokens": 1983855.0,
+      "reward": 1.911961317062378,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.5714285969734192,
+      "rewards/check_perplexity_diff_func": 0.9642857313156128,
+      "rewards/check_winston_local_func": 0.3762470781803131,
+      "step": 793
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 1.1118881118881119,
+      "grad_norm": 3.866365448835279e-07,
+      "kl": 0.1640625,
+      "learning_rate": 9.069336901363137e-05,
+      "loss": 0.0002,
+      "step": 794
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 101.42857360839844,
+      "epoch": 1.1132867132867132,
+      "grad_norm": 4.186823728733852e-07,
+      "kl": 0.1435546875,
+      "learning_rate": 9.065787980609695e-05,
+      "loss": 0.0001,
+      "num_tokens": 2004327.0,
+      "reward": 2.2043607234954834,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.7857142686843872,
+      "rewards/check_perplexity_diff_func": 0.9285714030265808,
+      "rewards/check_winston_local_func": 0.49007490277290344,
+      "step": 795
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 1.1146853146853146,
+      "grad_norm": 5.030626047983616e-07,
+      "kl": 0.1435546875,
+      "learning_rate": 9.062233003277983e-05,
+      "loss": 0.0001,
+      "step": 796
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 103.03572082519531,
+      "epoch": 1.116083916083916,
+      "grad_norm": 0.0014210803266876898,
+      "kl": 0.166015625,
+      "learning_rate": 9.058671974663655e-05,
+      "loss": 0.0001,
+      "num_tokens": 2024235.0,
+      "reward": 2.1631124019622803,
+      "reward_std": 0.02190079353749752,
+      "rewards/check_gptzero_func": 0.6428571343421936,
+      "rewards/check_perplexity_diff_func": 0.9821428656578064,
+      "rewards/check_winston_local_func": 0.538112461566925,
+      "step": 797
+    },
+    {
+      "clip_ratio": 0.00044923630775883794,
+      "epoch": 1.1174825174825176,
+      "grad_norm": 1.0597804777807007,
+      "kl": 0.1669921875,
+      "learning_rate": 9.055104900071376e-05,
+      "loss": 0.0008,
+      "step": 798
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 112.67857360839844,
+      "epoch": 1.118881118881119,
+      "grad_norm": 2.8951990171744424e-07,
+      "kl": 0.1591796875,
+      "learning_rate": 9.051531784814817e-05,
+      "loss": 0.0002,
+      "num_tokens": 2045053.0,
+      "reward": 2.453437328338623,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 1.0,
+      "rewards/check_perplexity_diff_func": 1.0,
+      "rewards/check_winston_local_func": 0.45343732833862305,
+      "step": 799
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 1.1202797202797203,
+      "grad_norm": 6.334011517220965e-07,
+      "kl": 0.1591796875,
+      "learning_rate": 9.047952634216652e-05,
+      "loss": 0.0002,
+      "step": 800
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 130.5,
+      "epoch": 1.1216783216783217,
+      "grad_norm": 5.6671055086629736e-05,
+      "kl": 0.130859375,
+      "learning_rate": 9.044367453608542e-05,
+      "loss": 0.0001,
+      "num_tokens": 2068083.0,
+      "reward": 2.568925142288208,
+      "reward_std": 0.0004195176879875362,
+      "rewards/check_gptzero_func": 1.0714285373687744,
+      "rewards/check_perplexity_diff_func": 1.0,
+      "rewards/check_winston_local_func": 0.4974963665008545,
+      "step": 801
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 1.123076923076923,
+      "grad_norm": 6.526367004124096e-05,
+      "kl": 0.130859375,
+      "learning_rate": 9.040776248331129e-05,
+      "loss": 0.0001,
+      "step": 802
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 110.83929443359375,
+      "epoch": 1.1244755244755245,
+      "grad_norm": 0.0007158488471600192,
+      "kl": 0.1591796875,
+      "learning_rate": 9.037179023734035e-05,
+      "loss": -0.0001,
+      "num_tokens": 2089028.0,
+      "reward": 2.4878017902374268,
+      "reward_std": 0.0023686587810516357,
+      "rewards/check_gptzero_func": 1.0,
+      "rewards/check_perplexity_diff_func": 1.0,
+      "rewards/check_winston_local_func": 0.4878018796443939,
+      "step": 803
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 1.1258741258741258,
+      "grad_norm": 0.000683099567038594,
+      "kl": 0.1591796875,
+      "learning_rate": 9.033575785175842e-05,
+      "loss": -0.0001,
+      "step": 804
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 117.87500762939453,
+      "epoch": 1.1272727272727272,
+      "grad_norm": 0.007538791801225539,
+      "kl": 0.216796875,
+      "learning_rate": 9.029966538024097e-05,
+      "loss": 0.0001,
+      "num_tokens": 2110797.0,
+      "reward": 2.3355278968811035,
+      "reward_std": 0.011562000960111618,
+      "rewards/check_gptzero_func": 0.8571428656578064,
+      "rewards/check_perplexity_diff_func": 0.9642857313156128,
+      "rewards/check_winston_local_func": 0.5140992999076843,
+      "step": 805
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 1.1286713286713286,
+      "grad_norm": 0.006982956687476011,
+      "kl": 0.2177734375,
+      "learning_rate": 9.026351287655294e-05,
+      "loss": 0.0001,
+      "step": 806
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 84.78572082519531,
+      "epoch": 1.1300699300699302,
+      "grad_norm": 6.477926555870939e-06,
+      "kl": 0.21875,
+      "learning_rate": 9.02273003945487e-05,
+      "loss": 0.0002,
+      "num_tokens": 2128445.0,
+      "reward": 1.931990623474121,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.3571428656578064,
+      "rewards/check_perplexity_diff_func": 1.0,
+      "rewards/check_winston_local_func": 0.5748476386070251,
+      "step": 807
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 1.1314685314685315,
+      "grad_norm": 8.039632616266636e-06,
+      "kl": 0.21875,
+      "learning_rate": 9.019102798817197e-05,
+      "loss": 0.0002,
+      "step": 808
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 96.9464340209961,
+      "epoch": 1.132867132867133,
+      "grad_norm": 0.015062463501999659,
+      "kl": 0.2080078125,
+      "learning_rate": 9.015469571145572e-05,
+      "loss": -0.0003,
+      "num_tokens": 2147760.0,
+      "reward": 1.9913533926010132,
+      "reward_std": 0.013614636845886707,
+      "rewards/check_gptzero_func": 0.6428571343421936,
+      "rewards/check_perplexity_diff_func": 0.9642857313156128,
+      "rewards/check_winston_local_func": 0.3842104375362396,
+      "step": 809
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 1.1342657342657343,
+      "grad_norm": 0.016309061661870246,
+      "kl": 0.20703125,
+      "learning_rate": 9.011830361852216e-05,
+      "loss": -0.0003,
+      "step": 810
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 100.25000762939453,
+      "epoch": 1.1356643356643357,
+      "grad_norm": 2.5340229455467882e-05,
+      "kl": 0.1669921875,
+      "learning_rate": 9.008185176358256e-05,
+      "loss": 0.0002,
+      "num_tokens": 2167618.0,
+      "reward": 2.2566254138946533,
+      "reward_std": 0.0,
+      "rewards/check_gptzero_func": 0.8571428656578064,
+      "rewards/check_perplexity_diff_func": 0.9642857313156128,
+      "rewards/check_winston_local_func": 0.43519681692123413,
+      "step": 811
+    },
+    {
+      "clip_ratio": 0.0,
+      "epoch": 1.137062937062937,
+      "grad_norm": 7.422350668171382e-05,
+      "kl": 0.1669921875,
+      "learning_rate": 9.00453402009372e-05,
+      "loss": 0.0002,
+      "step": 812
+    },
+    {
+      "clip_ratio": 0.0003940887108910829,
+      "completion_length": 126.26786041259766,
+      "epoch": 1.1384615384615384,
+      "grad_norm": 0.1475191777101766,
+      "kl": 0.1787109375,
+      "learning_rate": 9.000876898497532e-05,
+      "loss": -0.0042,
+      "num_tokens": 2190531.0,
+      "reward": 2.253868579864502,
+      "reward_std": 0.04925652965903282,
+      "rewards/check_gptzero_func": 0.8928571343421936,
+      "rewards/check_perplexity_diff_func": 0.9107142686843872,
+      "rewards/check_winston_local_func": 0.45029687881469727,
+      "step": 813
+    },
+    {
+      "clip_ratio": 0.000985221704468131,
+      "epoch": 1.1398601398601398,
+      "grad_norm": 0.06884327492740326,
+      "kl": 0.1806640625,
+      "learning_rate": 8.997213817017507e-05,
+      "loss": -0.0052,
+      "step": 814
+    },
+    {
+      "clip_ratio": 0.002508617704734206,
+      "completion_length": 153.42857360839844,
+      "epoch": 1.1412587412587412,
+      "grad_norm": 0.3371983554749533,
+      "kl": 0.373046875,
+      "learning_rate": 8.993544781110327e-05,
+      "loss": 0.0302,
+      "num_tokens": 2217307.0,
+      "reward": 2.651003360748291,
+      "reward_std": 0.27240806818008423,
+      "rewards/check_gptzero_func": 1.3214285373687744,
+      "rewards/check_perplexity_diff_func": 0.7678571343421936,
+      "rewards/check_winston_local_func": 0.5617175102233887,
+      "step": 815
+    },
+    {
+      "clip_ratio": 0.007771393749862909,
+      "epoch": 1.1426573426573428,
+      "grad_norm": 0.18940625122936564,
+      "kl": 0.369140625,
+      "learning_rate": 8.989869796241559e-05,
+      "loss": 0.0269,
+      "step": 816
+    },
+    {
+      "clip_ratio": 0.007156358100473881,
+      "completion_length": 95.17857360839844,
+      "epoch": 1.1440559440559441,
+      "grad_norm": 0.4666536744917174,
+      "kl": 0.8046875,
+      "learning_rate": 8.986188867885617e-05,
+      "loss": -0.0305,
+      "num_tokens": 2236277.0,
+      "reward": 2.897188663482666,
+      "reward_std": 0.5179702043533325,
+      "rewards/check_gptzero_func": 1.5,
+      "rewards/check_perplexity_diff_func": 0.5714285969734192,
+      "rewards/check_winston_local_func": 0.8257598876953125,
+      "step": 817
+    },
+    {
+      "clip_ratio": 0.021209854632616043,
+      "epoch": 1.1454545454545455,
+      "grad_norm": 8.279801106797997,
+      "kl": 91.5,
+      "learning_rate": 8.982502001525778e-05,
+      "loss": 0.0702,
+      "step": 818
+    },
+    {
+      "clip_ratio": 0.007509272545576096,
+      "completion_length": 88.6964340209961,
+      "epoch": 1.1468531468531469,
+      "grad_norm": 0.35088469016777446,
+      "kl": 0.953125,
+      "learning_rate": 8.978809202654162e-05,
+      "loss": 0.0299,
+      "num_tokens": 2254384.0,
+      "reward": 3.254634141921997,
+      "reward_std": 0.37467071413993835,
+      "rewards/check_gptzero_func": 1.7857142686843872,
+      "rewards/check_perplexity_diff_func": 0.5535714030265808,
+      "rewards/check_winston_local_func": 0.91534823179245,
+      "step": 819
+    },
+    {
+      "clip_ratio": 0.031093057245016098,
+      "epoch": 1.1482517482517482,
+      "grad_norm": 0.24890907696684347,
+      "kl": 1.3125,
+      "learning_rate": 8.975110476771724e-05,
+      "loss": 0.0221,
+      "step": 820
+    },
+    {
+      "clip_ratio": 0.0030267522670328617,
+      "completion_length": 106.66072082519531,
+      "epoch": 1.1496503496503496,
+      "grad_norm": 0.22909787091102216,
+      "kl": 0.76171875,
+      "learning_rate": 8.971405829388253e-05,
+      "loss": 0.0243,
+      "num_tokens": 2274383.0,
+      "reward": 2.738980293273926,
+      "reward_std": 0.5776112079620361,
+      "rewards/check_gptzero_func": 1.5714285373687744,
+      "rewards/check_perplexity_diff_func": 0.3571428656578064,
+      "rewards/check_winston_local_func": 0.8104087710380554,
+      "step": 821
+    },
+    {
+      "clip_ratio": 0.024285081773996353,
+      "epoch": 1.151048951048951,
+      "grad_norm": 0.2030178324659267,
+      "kl": 0.73046875,
+      "learning_rate": 8.967695266022355e-05,
+      "loss": 0.019,
+      "step": 822
+    },
+    {
+      "clip_ratio": 0.004899229854345322,
+      "completion_length": 115.4464340209961,
+      "epoch": 1.1524475524475524,
+      "grad_norm": 0.15388182261983746,
+      "kl": 0.6796875,
+      "learning_rate": 8.963978792201449e-05,
+      "loss": -0.0315,
+      "num_tokens": 2295796.0,
+      "reward": 2.9562063217163086,
+      "reward_std": 0.46262314915657043,
+      "rewards/check_gptzero_func": 1.6428571939468384,
+      "rewards/check_perplexity_diff_func": 0.4642857015132904,
+      "rewards/check_winston_local_func": 0.8490633964538574,
+      "step": 823
+    },
+    {
+      "clip_ratio": 0.011996669694781303,
+      "epoch": 1.1538461538461537,
+      "grad_norm": 0.11115252524164657,
+      "kl": 0.69140625,
+      "learning_rate": 8.96025641346176e-05,
+      "loss": -0.0363,
+      "step": 824
+    },
+    {
+      "clip_ratio": 0.005713926628232002,
+      "completion_length": 100.67857360839844,
+      "epoch": 1.1552447552447553,
+      "grad_norm": 0.38159859212017105,
+      "kl": 1.0078125,
+      "learning_rate": 8.95652813534831e-05,
+      "loss": 0.0247,
+      "num_tokens": 2316068.0,
+      "reward": 2.900416612625122,
+      "reward_std": 0.6779367327690125,
+      "rewards/check_gptzero_func": 1.7142857313156128,
+      "rewards/check_perplexity_diff_func": 0.2857142984867096,
+      "rewards/check_winston_local_func": 0.9004164934158325,
+      "step": 825
+    },
+    {
+      "clip_ratio": 0.013081731274724007,
+      "epoch": 1.1566433566433567,
+      "grad_norm": 0.2076814708084972,
+      "kl": 0.98828125,
+      "learning_rate": 8.952793963414907e-05,
+      "loss": 0.0069,
+      "step": 826
+    },
+    {
+      "clip_ratio": 0.005952088627964258,
+      "completion_length": 108.83928680419922,
+      "epoch": 1.158041958041958,
+      "grad_norm": 0.25572967826609705,
+      "kl": 0.9140625,
+      "learning_rate": 8.949053903224137e-05,
+      "loss": -0.0056,
+      "num_tokens": 2337955.0,
+      "reward": 2.9091410636901855,
+      "reward_std": 0.4837423861026764,
+      "rewards/check_gptzero_func": 1.6071428060531616,
+      "rewards/check_perplexity_diff_func": 0.4285714328289032,
+      "rewards/check_winston_local_func": 0.8734264373779297,
+      "step": 827
+    },
+    {
+      "clip_ratio": 0.02270878292620182,
+      "epoch": 1.1594405594405595,
+      "grad_norm": 0.1976733952761619,
+      "kl": 1.0390625,
+      "learning_rate": 8.945307960347363e-05,
+      "loss": -0.013,
+      "step": 828
+    },
+    {
+      "clip_ratio": 0.004438840784132481,
+      "completion_length": 95.50000762939453,
+      "epoch": 1.1608391608391608,
+      "grad_norm": 0.32416668330490334,
+      "kl": 0.94921875,
+      "learning_rate": 8.941556140364706e-05,
+      "loss": -0.0013,
+      "num_tokens": 2356981.0,
+      "reward": 2.690964698791504,
+      "reward_std": 0.5679685473442078,
+      "rewards/check_gptzero_func": 1.5,
+      "rewards/check_perplexity_diff_func": 0.3928571343421936,
+      "rewards/check_winston_local_func": 0.7981075048446655,
+      "step": 829
+    },
+    {
+      "clip_ratio": 0.02389359660446644,
+      "epoch": 1.1622377622377622,
+      "grad_norm": 0.1674670942817363,
+      "kl": 1.109375,
+      "learning_rate": 8.937798448865044e-05,
+      "loss": -0.0101,
+      "step": 830
+    },
+    {
+      "clip_ratio": 0.0023547932505607605,
+      "completion_length": 70.05357360839844,
+      "epoch": 1.1636363636363636,
+      "grad_norm": 0.2984951445058083,
+      "kl": 1.03125,
+      "learning_rate": 8.934034891446e-05,
+      "loss": -0.0438,
+      "num_tokens": 2372552.0,
+      "reward": 2.58469557762146,
+      "reward_std": 0.4580416977405548,
+      "rewards/check_gptzero_func": 1.2857142686843872,
+      "rewards/check_perplexity_diff_func": 0.4821428656578064,
+      "rewards/check_winston_local_func": 0.8168383836746216,
+      "step": 831
+    },
+    {
+      "clip_ratio": 0.015139752998948097,
+      "epoch": 1.165034965034965,
+      "grad_norm": 0.7045953524167903,
+      "kl": 0.94921875,
+      "learning_rate": 8.930265473713938e-05,
+      "loss": -0.0504,
+      "step": 832
+    },
+    {
+      "clip_ratio": 0.004466689191758633,
+      "completion_length": 111.60714721679688,
+      "epoch": 1.1664335664335663,
+      "grad_norm": 0.18705400225320276,
+      "kl": 0.8828125,
+      "learning_rate": 8.926490201283948e-05,
+      "loss": 0.0038,
+      "num_tokens": 2392914.0,
+      "reward": 2.811556816101074,
+      "reward_std": 0.4326061010360718,
+      "rewards/check_gptzero_func": 1.6071428060531616,
+      "rewards/check_perplexity_diff_func": 0.3392857015132904,
+      "rewards/check_winston_local_func": 0.865128219127655,
+      "step": 833
+    },
+    {
+      "clip_ratio": 0.022041011601686478,
+      "epoch": 1.167832167832168,
+      "grad_norm": 0.12453141987495926,
+      "kl": 0.93359375,
+      "learning_rate": 8.922709079779846e-05,
+      "loss": -0.0003,
+      "step": 834
+    },
+    {
+      "clip_ratio": 0.0027127026114612818,
+      "completion_length": 109.12500762939453,
+      "epoch": 1.1692307692307693,
+      "grad_norm": 0.19278606652413194,
+      "kl": 0.9140625,
+      "learning_rate": 8.918922114834156e-05,
+      "loss": -0.0038,
+      "num_tokens": 2413065.0,
+      "reward": 2.8914380073547363,
+      "reward_std": 0.4447008967399597,
+      "rewards/check_gptzero_func": 1.5,
+      "rewards/check_perplexity_diff_func": 0.5357142686843872,
+      "rewards/check_winston_local_func": 0.8557236790657043,
+      "step": 835
+    },
+    {
+      "clip_ratio": 0.009510784409940243,
+      "epoch": 1.1706293706293707,
+      "grad_norm": 0.15050199959459193,
+      "kl": 0.859375,
+      "learning_rate": 8.915129312088112e-05,
+      "loss": -0.0097,
+      "step": 836
+    },
+    {
+      "clip_ratio": 0.004420355428010225,
+      "completion_length": 135.0,
+      "epoch": 1.172027972027972,
+      "grad_norm": 0.18992143693052943,
+      "kl": 1.34375,
+      "learning_rate": 8.911330677191638e-05,
+      "loss": -0.0214,
+      "num_tokens": 2436495.0,
+      "reward": 3.075127363204956,
+      "reward_std": 0.3562222421169281,
+      "rewards/check_gptzero_func": 1.7857142686843872,
+      "rewards/check_perplexity_diff_func": 0.4285714328289032,
+      "rewards/check_winston_local_func": 0.8608413338661194,
+      "step": 837
+    },
+    {
+      "clip_ratio": 0.013964839279651642,
+      "epoch": 1.1734265734265734,
+      "grad_norm": 0.12537032841396828,
+      "kl": 0.984375,
+      "learning_rate": 8.90752621580335e-05,
+      "loss": -0.0261,
+      "step": 838
+    },
+    {
+      "clip_ratio": 0.004812764469534159,
+      "completion_length": 106.42857360839844,
+      "epoch": 1.1748251748251748,
+      "grad_norm": 0.4607092699431397,
+      "kl": 0.92578125,
+      "learning_rate": 8.903715933590544e-05,
+      "loss": 0.0288,
+      "num_tokens": 2456367.0,
+      "reward": 2.8998942375183105,
+      "reward_std": 0.5146588087081909,
+      "rewards/check_gptzero_func": 1.5,
+      "rewards/check_perplexity_diff_func": 0.5178571343421936,
+      "rewards/check_winston_local_func": 0.8820368051528931,
+      "step": 839
+    },
+    {
+      "clip_ratio": 0.034036021679639816,
+      "epoch": 1.1762237762237762,
+      "grad_norm": 0.2716074424016385,
+      "kl": 0.87109375,
+      "learning_rate": 8.899899836229185e-05,
+      "loss": 0.0179,
+      "step": 840
+    },
+    {
+      "clip_ratio": 0.0062322416342794895,
+      "completion_length": 121.76786041259766,
+      "epoch": 1.1776223776223775,
+      "grad_norm": 0.13281232500681697,
+      "kl": 0.8671875,
+      "learning_rate": 8.896077929403901e-05,
+      "loss": -0.0058,
+      "num_tokens": 2478560.0,
+      "reward": 2.824719190597534,
+      "reward_std": 0.459418386220932,
+      "rewards/check_gptzero_func": 1.6428571939468384,
+      "rewards/check_perplexity_diff_func": 0.3214285671710968,
+      "rewards/check_winston_local_func": 0.8604331612586975,
+      "step": 841
+    },
+    {
+      "clip_ratio": 0.030564727261662483,
+      "epoch": 1.179020979020979,
+      "grad_norm": 0.14925851302897927,
+      "kl": 0.99609375,
+      "learning_rate": 8.892250218807974e-05,
+      "loss": -0.0085,
+      "step": 842
+    },
+    {
+      "clip_ratio": 0.005103765986859798,
+      "completion_length": 130.9107208251953,
+      "epoch": 1.1804195804195805,
+      "grad_norm": 0.13057549986848624,
+      "kl": 0.76953125,
+      "learning_rate": 8.88841671014333e-05,
+      "loss": -0.0254,
+      "num_tokens": 2501767.0,
+      "reward": 3.120969533920288,
+      "reward_std": 0.3832559883594513,
+      "rewards/check_gptzero_func": 1.8214285373687744,
+      "rewards/check_perplexity_diff_func": 0.4642857015132904,
+      "rewards/check_winston_local_func": 0.8352552056312561,
+      "step": 843
+    },
+    {
+      "clip_ratio": 0.00968684908002615,
+      "epoch": 1.1818181818181819,
+      "grad_norm": 0.10659044072025987,
+      "kl": 0.77734375,
+      "learning_rate": 8.884577409120535e-05,
+      "loss": -0.0288,
+      "step": 844
+    },
+    {
+      "clip_ratio": 0.006903046276420355,
+      "completion_length": 121.48214721679688,
+      "epoch": 1.1832167832167833,
+      "grad_norm": 0.1792345300425977,
+      "kl": 0.95703125,
+      "learning_rate": 8.880732321458784e-05,
+      "loss": -0.0268,
+      "num_tokens": 2524422.0,
+      "reward": 3.155743360519409,
+      "reward_std": 0.3429409861564636,
+      "rewards/check_gptzero_func": 1.9285714626312256,
+      "rewards/check_perplexity_diff_func": 0.3392857015132904,
+      "rewards/check_winston_local_func": 0.8878862261772156,
+      "step": 845
+    },
+    {
+      "clip_ratio": 0.010083622299134731,
+      "epoch": 1.1846153846153846,
+      "grad_norm": 0.1567593685401593,
+      "kl": 0.9609375,
+      "learning_rate": 8.876881452885889e-05,
+      "loss": -0.0313,
+      "step": 846
+    },
+    {
+      "clip_ratio": 0.0059029581025242805,
+      "completion_length": 91.89286041259766,
+      "epoch": 1.186013986013986,
+      "grad_norm": 0.15058698267173182,
+      "kl": 0.875,
+      "learning_rate": 8.873024809138272e-05,
+      "loss": -0.0043,
+      "num_tokens": 2543122.0,
+      "reward": 3.241715908050537,
+      "reward_std": 0.34929126501083374,
+      "rewards/check_gptzero_func": 1.9642857313156128,
+      "rewards/check_perplexity_diff_func": 0.3928571343421936,
+      "rewards/check_winston_local_func": 0.8845729231834412,
+      "step": 847
+    },
+    {
+      "clip_ratio": 0.016130443662405014,
+      "epoch": 1.1874125874125874,
+      "grad_norm": 0.11770457997063148,
+      "kl": 0.87109375,
+      "learning_rate": 8.869162395960966e-05,
+      "loss": -0.0068,
+      "step": 848
+    },
+    {
+      "clip_ratio": 0.005722040310502052,
+      "completion_length": 96.9464340209961,
+      "epoch": 1.1888111888111887,
+      "grad_norm": 0.20581876847454508,
+      "kl": 0.94921875,
+      "learning_rate": 8.86529421910759e-05,
+      "loss": -0.0095,
+      "num_tokens": 2562419.0,
+      "reward": 3.0281920433044434,
+      "reward_std": 0.39599350094795227,
+      "rewards/check_gptzero_func": 1.75,
+      "rewards/check_perplexity_diff_func": 0.4107142984867096,
+      "rewards/check_winston_local_func": 0.8674777150154114,
+      "step": 849
+    },
+    {
+      "clip_ratio": 0.009945889003574848,
+      "epoch": 1.1902097902097901,
+      "grad_norm": 0.17347879325044108,
+      "kl": 0.93359375,
+      "learning_rate": 8.861420284340352e-05,
+      "loss": -0.0159,
+      "step": 850
+    },
+    {
+      "clip_ratio": 0.009763496927917004,
+      "completion_length": 92.17857360839844,
+      "epoch": 1.1916083916083915,
+      "grad_norm": 0.25864509842699746,
+      "kl": 1.1328125,
+      "learning_rate": 8.857540597430036e-05,
+      "loss": -0.0206,
+      "num_tokens": 2581123.0,
+      "reward": 3.1546006202697754,
+      "reward_std": 0.39692866802215576,
+      "rewards/check_gptzero_func": 1.9285714626312256,
+      "rewards/check_perplexity_diff_func": 0.2857142984867096,
+      "rewards/check_winston_local_func": 0.940314769744873,
+      "step": 851
+    },
+    {
+      "clip_ratio": 0.017916982993483543,
+      "epoch": 1.193006993006993,
+      "grad_norm": 0.17087058222849108,
+      "kl": 1.1640625,
+      "learning_rate": 8.853655164155998e-05,
+      "loss": -0.0291,
+      "step": 852
+    },
+    {
+      "clip_ratio": 0.006903782952576876,
+      "completion_length": 111.30357360839844,
+      "epoch": 1.1944055944055945,
+      "grad_norm": 0.16071423284080857,
+      "kl": 0.93359375,
+      "learning_rate": 8.849763990306152e-05,
+      "loss": 0.0205,
+      "num_tokens": 2601408.0,
+      "reward": 3.3505136966705322,
+      "reward_std": 0.22764992713928223,
+      "rewards/check_gptzero_func": 2.0,
+      "rewards/check_perplexity_diff_func": 0.4285714328289032,
+      "rewards/check_winston_local_func": 0.921941876411438,
+      "step": 853
+    },
+    {
+      "clip_ratio": 0.013233030214905739,
+      "epoch": 1.1958041958041958,
+      "grad_norm": 0.12706931182647724,
+      "kl": 0.953125,
+      "learning_rate": 8.845867081676962e-05,
+      "loss": 0.0165,
+      "step": 854
+    },
+    {
+      "clip_ratio": 0.009717343375086784,
+      "completion_length": 85.5714340209961,
+      "epoch": 1.1972027972027972,
+      "grad_norm": 0.474864564292558,
+      "kl": 1.359375,
+      "learning_rate": 8.841964444073437e-05,
+      "loss": -0.0038,
+      "num_tokens": 2619040.0,
+      "reward": 3.180955410003662,
+      "reward_std": 0.24645240604877472,
+      "rewards/check_gptzero_func": 1.8571428060531616,
+      "rewards/check_perplexity_diff_func": 0.4107142984867096,
+      "rewards/check_winston_local_func": 0.9130982160568237,
+      "step": 855
+    },
+    {
+      "clip_ratio": 0.031235292553901672,
+      "epoch": 1.1986013986013986,
+      "grad_norm": 115099.93064186395,
+      "kl": 1261568.0,
+      "learning_rate": 8.838056083309118e-05,
+      "loss": 1267.3759,
+      "step": 856
+    },
+    {
+      "clip_ratio": 0.008160261437296867,
+      "completion_length": 95.42857360839844,
+      "epoch": 1.2,
+      "grad_norm": 0.32932688116557046,
+      "kl": 1.140625,
+      "learning_rate": 8.834142005206075e-05,
+      "loss": -0.0163,
+      "num_tokens": 2638598.0,
+      "reward": 3.330536127090454,
+      "reward_std": 0.5345718264579773,
+      "rewards/check_gptzero_func": 1.8928571939468384,
+      "rewards/check_perplexity_diff_func": 0.5,
+      "rewards/check_winston_local_func": 0.9376789927482605,
+      "step": 857
+    },
+    {
+      "clip_ratio": 0.06036671996116638,
+      "epoch": 1.2013986013986013,
+      "grad_norm": 0.40835843242634484,
+      "kl": 1.0625,
+      "learning_rate": 8.83022221559489e-05,
+      "loss": -0.0227,
+      "step": 858
+    },
+    {
+      "clip_ratio": 0.0028968648985028267,
+      "completion_length": 84.76786041259766,
+      "epoch": 1.2027972027972027,
+      "grad_norm": 0.4064982927186033,
+      "kl": 1.2890625,
+      "learning_rate": 8.826296720314657e-05,
+      "loss": -0.0066,
+      "num_tokens": 2658457.0,
+      "reward": 3.166985034942627,
+      "reward_std": 0.3735349774360657,
+      "rewards/check_gptzero_func": 1.8928571939468384,
+      "rewards/check_perplexity_diff_func": 0.3214285671710968,
+      "rewards/check_winston_local_func": 0.9526991248130798,
+      "step": 859
+    },
+    {
+      "clip_ratio": 0.05851617828011513,
+      "epoch": 1.204195804195804,
+      "grad_norm": 0.20749144960095425,
+      "kl": 1.3515625,
+      "learning_rate": 8.822365525212968e-05,
+      "loss": -0.0176,
+      "step": 860
+    },
+    {
+      "clip_ratio": 0.0031495015136897564,
+      "completion_length": 80.1964340209961,
+      "epoch": 1.2055944055944057,
+      "grad_norm": 0.15053519471673255,
+      "kl": 1.125,
+      "learning_rate": 8.818428636145906e-05,
+      "loss": -0.018,
+      "num_tokens": 2678678.0,
+      "reward": 3.2977185249328613,
+      "reward_std": 0.2966470420360565,
+      "rewards/check_gptzero_func": 2.0,
+      "rewards/check_perplexity_diff_func": 0.3392857015132904,
+      "rewards/check_winston_local_func": 0.9584324955940247,
+      "step": 861
+    },
+    {
+      "clip_ratio": 0.016123468056321144,
+      "epoch": 1.206993006993007,
+      "grad_norm": 0.18455354967998963,
+      "kl": 1.125,
+      "learning_rate": 8.814486058978035e-05,
+      "loss": -0.0224,
+      "step": 862
+    },
+    {
+      "clip_ratio": 0.004086082335561514,
+      "completion_length": 92.3214340209961,
+      "epoch": 1.2083916083916084,
+      "grad_norm": 0.27228134170848783,
+      "kl": 1.296875,
+      "learning_rate": 8.810537799582393e-05,
+      "loss": -0.023,
+      "num_tokens": 2699946.0,
+      "reward": 3.193101406097412,
+      "reward_std": 0.6046900749206543,
+      "rewards/check_gptzero_func": 1.8214285373687744,
+      "rewards/check_perplexity_diff_func": 0.4285714328289032,
+      "rewards/check_winston_local_func": 0.9431011080741882,
+      "step": 863
+    },
+    {
+      "clip_ratio": 0.018903842195868492,
+      "epoch": 1.2097902097902098,
+      "grad_norm": 0.22986574920613892,
+      "kl": 1.3203125,
+      "learning_rate": 8.806583863840482e-05,
+      "loss": -0.037,
+      "step": 864
+    },
+    {
+      "clip_ratio": 0.003315736772492528,
+      "completion_length": 93.60714721679688,
+      "epoch": 1.2111888111888112,
+      "grad_norm": 0.172235850839695,
+      "kl": 1.1875,
+      "learning_rate": 8.802624257642261e-05,
+      "loss": -0.0352,
+      "num_tokens": 2721026.0,
+      "reward": 3.263352155685425,
+      "reward_std": 0.44618692994117737,
+      "rewards/check_gptzero_func": 1.9642857313156128,
+      "rewards/check_perplexity_diff_func": 0.375,
+      "rewards/check_winston_local_func": 0.9240660071372986,
+      "step": 865
+    },
+    {
+      "clip_ratio": 0.011954267509281635,
+      "epoch": 1.2125874125874125,
+      "grad_norm": 0.14917383286155952,
+      "kl": 1.1875,
+      "learning_rate": 8.798658986886135e-05,
+      "loss": -0.0419,
+      "step": 866
+    },
+    {
+      "clip_ratio": 0.004093995317816734,
+      "completion_length": 80.4464340209961,
+      "epoch": 1.213986013986014,
+      "grad_norm": 0.18510792865464507,
+      "kl": 1.25,
+      "learning_rate": 8.794688057478948e-05,
+      "loss": -0.0487,
+      "num_tokens": 2739883.0,
+      "reward": 3.1514322757720947,
+      "reward_std": 0.2983841896057129,
+      "rewards/check_gptzero_func": 1.9642857313156128,
+      "rewards/check_perplexity_diff_func": 0.2321428507566452,
+      "rewards/check_winston_local_func": 0.9550036191940308,
+      "step": 867
+    },
+    {
+      "clip_ratio": 0.013391178101301193,
+      "epoch": 1.2153846153846155,
+      "grad_norm": 0.1438802997383443,
+      "kl": 1.25,
+      "learning_rate": 8.790711475335971e-05,
+      "loss": -0.0558,
+      "step": 868
+    },
+    {
+      "clip_ratio": 0.005463641602545977,
+      "completion_length": 81.05357360839844,
+      "epoch": 1.2167832167832167,
+      "grad_norm": 0.18670367539464772,
+      "kl": 1.2890625,
+      "learning_rate": 8.786729246380901e-05,
+      "loss": 0.0148,
+      "num_tokens": 2757716.0,
+      "reward": 3.0032906532287598,
+      "reward_std": 0.28309959173202515,
+      "rewards/check_gptzero_func": 1.8571428060531616,
+      "rewards/check_perplexity_diff_func": 0.1964285671710968,
+      "rewards/check_winston_local_func": 0.9497188925743103,
+      "step": 869
+    },
+    {
+      "clip_ratio": 0.016334589570760727,
+      "epoch": 1.2181818181818183,
+      "grad_norm": 0.15273097718440906,
+      "kl": 1.34375,
+      "learning_rate": 8.782741376545838e-05,
+      "loss": 0.0072,
+      "step": 870
+    },
+    {
+      "clip_ratio": 0.00400504469871521,
+      "completion_length": 113.25000762939453,
+      "epoch": 1.2195804195804196,
+      "grad_norm": 0.20246778640237242,
+      "kl": 1.265625,
+      "learning_rate": 8.778747871771292e-05,
+      "loss": -0.021,
+      "num_tokens": 2780456.0,
+      "reward": 3.18583083152771,
+      "reward_std": 0.41532081365585327,
+      "rewards/check_gptzero_func": 1.8928571939468384,
+      "rewards/check_perplexity_diff_func": 0.3214285671710968,
+      "rewards/check_winston_local_func": 0.9715452194213867,
+      "step": 871
+    },
+    {
+      "clip_ratio": 0.01152871921658516,
+      "epoch": 1.220979020979021,
+      "grad_norm": 0.15664458772417972,
+      "kl": 1.2578125,
+      "learning_rate": 8.774748738006167e-05,
+      "loss": -0.0293,
+      "step": 872
+    },
+    {
+      "clip_ratio": 0.0038485296536237,
+      "completion_length": 82.21428680419922,
+      "epoch": 1.2223776223776224,
+      "grad_norm": 0.25567822509642146,
+      "kl": 1.28125,
+      "learning_rate": 8.770743981207747e-05,
+      "loss": -0.0049,
+      "num_tokens": 2798202.0,
+      "reward": 3.195807456970215,
+      "reward_std": 0.32202741503715515,
+      "rewards/check_gptzero_func": 1.8928571939468384,
+      "rewards/check_perplexity_diff_func": 0.3571428656578064,
+      "rewards/check_winston_local_func": 0.9458074569702148,
+      "step": 873
+    },
+    {
+      "clip_ratio": 0.021505463868379593,
+      "epoch": 1.2237762237762237,
+      "grad_norm": 0.19814335325949858,
+      "kl": 1.3125,
+      "learning_rate": 8.766733607341698e-05,
+      "loss": -0.0148,
+      "step": 874
+    },
+    {
+      "clip_ratio": 0.003552226349711418,
+      "completion_length": 83.28572082519531,
+      "epoch": 1.2251748251748251,
+      "grad_norm": 0.20458282916731427,
+      "kl": 1.28125,
+      "learning_rate": 8.762717622382051e-05,
+      "loss": 0.0091,
+      "num_tokens": 2816740.0,
+      "reward": 3.0412774085998535,
+      "reward_std": 0.45299655199050903,
+      "rewards/check_gptzero_func": 1.7857142686843872,
+      "rewards/check_perplexity_diff_func": 0.3392857015132904,
+      "rewards/check_winston_local_func": 0.9162774682044983,
+      "step": 875
+    },
+    {
+      "clip_ratio": 0.031206289306282997,
+      "epoch": 1.2265734265734265,
+      "grad_norm": 0.1895907465531785,
+      "kl": 1.2578125,
+      "learning_rate": 8.758696032311192e-05,
+      "loss": 0.0022,
+      "step": 876
+    },
+    {
+      "clip_ratio": 0.005021117627620697,
+      "completion_length": 97.14286041259766,
+      "epoch": 1.227972027972028,
+      "grad_norm": 0.1341097478485543,
+      "kl": 1.1484375,
+      "learning_rate": 8.754668843119864e-05,
+      "loss": 0.0109,
+      "num_tokens": 2837326.0,
+      "reward": 2.85992169380188,
+      "reward_std": 0.3086967468261719,
+      "rewards/check_gptzero_func": 1.6428571939468384,
+      "rewards/check_perplexity_diff_func": 0.25,
+      "rewards/check_winston_local_func": 0.9670643210411072,
+      "step": 877
+    },
+    {
+      "clip_ratio": 0.013011510483920574,
+      "epoch": 1.2293706293706295,
+      "grad_norm": 0.1100489441974723,
+      "kl": 1.15625,
+      "learning_rate": 8.750636060807146e-05,
+      "loss": 0.0068,
+      "step": 878
+    },
+    {
+      "clip_ratio": 0.004042315296828747,
+      "completion_length": 92.75000762939453,
+      "epoch": 1.2307692307692308,
+      "grad_norm": 0.20521102088717552,
+      "kl": 1.28125,
+      "learning_rate": 8.746597691380448e-05,
+      "loss": -0.0462,
+      "num_tokens": 2857790.0,
+      "reward": 3.2252542972564697,
+      "reward_std": 0.3173792064189911,
+      "rewards/check_gptzero_func": 1.8928571939468384,
+      "rewards/check_perplexity_diff_func": 0.3571428656578064,
+      "rewards/check_winston_local_func": 0.9752540588378906,
+      "step": 879
+    },
+    {
+      "clip_ratio": 0.01367542240768671,
+      "epoch": 1.2321678321678322,
+      "grad_norm": 0.16635756076828898,
+      "kl": 1.265625,
+      "learning_rate": 8.742553740855506e-05,
+      "loss": -0.0555,
+      "step": 880
+    },
+    {
+      "clip_ratio": 0.004514673724770546,
+      "completion_length": 73.3214340209961,
+      "epoch": 1.2335664335664336,
+      "grad_norm": 0.20550785653096285,
+      "kl": 1.3515625,
+      "learning_rate": 8.738504215256367e-05,
+      "loss": -0.0198,
+      "num_tokens": 2874956.0,
+      "reward": 3.0314509868621826,
+      "reward_std": 0.4046560227870941,
+      "rewards/check_gptzero_func": 1.8571428060531616,
+      "rewards/check_perplexity_diff_func": 0.2321428507566452,
+      "rewards/check_winston_local_func": 0.9421650171279907,
+      "step": 881
+    },
+    {
+      "clip_ratio": 0.013386914506554604,
+      "epoch": 1.234965034965035,
+      "grad_norm": 0.1514209552309319,
+      "kl": 1.390625,
+      "learning_rate": 8.734449120615385e-05,
+      "loss": -0.0281,
+      "step": 882
+    },
+    {
+      "clip_ratio": 0.004252195358276367,
+      "completion_length": 91.23214721679688,
+      "epoch": 1.2363636363636363,
+      "grad_norm": 0.26914170730768494,
+      "kl": 1.625,
+      "learning_rate": 8.730388462973208e-05,
+      "loss": -0.0356,
+      "num_tokens": 2894865.0,
+      "reward": 3.176909923553467,
+      "reward_std": 0.48494505882263184,
+      "rewards/check_gptzero_func": 1.8928571939468384,
+      "rewards/check_perplexity_diff_func": 0.3214285671710968,
+      "rewards/check_winston_local_func": 0.9626240134239197,
+      "step": 883
+    },
+    {
+      "clip_ratio": 0.029839998111128807,
+      "epoch": 1.2377622377622377,
+      "grad_norm": 0.2358577901886362,
+      "kl": 1.71875,
+      "learning_rate": 8.726322248378775e-05,
+      "loss": -0.0484,
+      "step": 884
+    },
+    {
+      "clip_ratio": 0.0061589390970766544,
+      "completion_length": 116.39286041259766,
+      "epoch": 1.239160839160839,
+      "grad_norm": 0.22649788384805056,
+      "kl": 1.296875,
+      "learning_rate": 8.722250482889294e-05,
+      "loss": -0.0477,
+      "num_tokens": 2917669.0,
+      "reward": 3.1957309246063232,
+      "reward_std": 0.4417218565940857,
+      "rewards/check_gptzero_func": 1.9285714626312256,
+      "rewards/check_perplexity_diff_func": 0.3035714328289032,
+      "rewards/check_winston_local_func": 0.963587760925293,
+      "step": 885
+    },
+    {
+      "clip_ratio": 0.02660902589559555,
+      "epoch": 1.2405594405594407,
+      "grad_norm": 0.21091675537901597,
+      "kl": 1.2421875,
+      "learning_rate": 8.718173172570254e-05,
+      "loss": -0.0567,
+      "step": 886
+    },
+    {
+      "clip_ratio": 0.004395401570945978,
+      "completion_length": 111.12500762939453,
+      "epoch": 1.241958041958042,
+      "grad_norm": 0.17935096954048527,
+      "kl": 1.21875,
+      "learning_rate": 8.714090323495392e-05,
+      "loss": -0.0197,
+      "num_tokens": 2938724.0,
+      "reward": 3.2519590854644775,
+      "reward_std": 0.3486594557762146,
+      "rewards/check_gptzero_func": 1.9285714626312256,
+      "rewards/check_perplexity_diff_func": 0.3571428656578064,
+      "rewards/check_winston_local_func": 0.9662445783615112,
+      "step": 887
+    },
+    {
+      "clip_ratio": 0.015748703852295876,
+      "epoch": 1.2433566433566434,
+      "grad_norm": 0.15491390668659677,
+      "kl": 1.234375,
+      "learning_rate": 8.710001941746704e-05,
+      "loss": -0.0268,
+      "step": 888
+    },
+    {
+      "clip_ratio": 0.0043277433142066,
+      "completion_length": 109.14286041259766,
+      "epoch": 1.2447552447552448,
+      "grad_norm": 0.20367981624275092,
+      "kl": 1.296875,
+      "learning_rate": 8.705908033414425e-05,
+      "loss": -0.0039,
+      "num_tokens": 2959316.0,
+      "reward": 3.349128484725952,
+      "reward_std": 0.4315777122974396,
+      "rewards/check_gptzero_func": 2.0,
+      "rewards/check_perplexity_diff_func": 0.375,
+      "rewards/check_winston_local_func": 0.9741284251213074,
+      "step": 889
+    },
+    {
+      "clip_ratio": 0.010803469456732273,
+      "epoch": 1.2461538461538462,
+      "grad_norm": 0.16095041335848162,
+      "kl": 1.296875,
+      "learning_rate": 8.701808604597022e-05,
+      "loss": -0.0125,
+      "step": 890
+    },
+    {
+      "clip_ratio": 0.007922212593257427,
+      "completion_length": 92.91072082519531,
+      "epoch": 1.2475524475524475,
+      "grad_norm": 0.3135038237758286,
+      "kl": 1.421875,
+      "learning_rate": 8.697703661401186e-05,
+      "loss": -0.0084,
+      "num_tokens": 2977813.0,
+      "reward": 3.1613879203796387,
+      "reward_std": 0.3479323089122772,
+      "rewards/check_gptzero_func": 1.9285714626312256,
+      "rewards/check_perplexity_diff_func": 0.2678571343421936,
+      "rewards/check_winston_local_func": 0.964959442615509,
+      "step": 891
+    },
+    {
+      "clip_ratio": 0.0198274664580822,
+      "epoch": 1.248951048951049,
+      "grad_norm": 0.20882817360023168,
+      "kl": 1.4375,
+      "learning_rate": 8.693593209941825e-05,
+      "loss": -0.0206,
+      "step": 892
+    },
+    {
+      "clip_ratio": 0.006882136687636375,
+      "completion_length": 104.03572082519531,
+      "epoch": 1.2503496503496503,
+      "grad_norm": 0.24252685088697998,
+      "kl": 1.2265625,
+      "learning_rate": 8.689477256342046e-05,
+      "loss": 0.0095,
+      "num_tokens": 2997551.0,
+      "reward": 3.286836862564087,
+      "reward_std": 0.3823050558567047,
+      "rewards/check_gptzero_func": 1.9642857313156128,
+      "rewards/check_perplexity_diff_func": 0.3571428656578064,
+      "rewards/check_winston_local_func": 0.9654080271720886,
+      "step": 893
+    },
+    {
+      "clip_ratio": 0.020427286624908447,
+      "epoch": 1.2517482517482517,
+      "grad_norm": 0.2062850181216717,
+      "kl": 1.328125,
+      "learning_rate": 8.685355806733158e-05,
+      "loss": -0.0013,
+      "step": 894
+    },
+    {
+      "clip_ratio": 0.004929638002067804,
+      "completion_length": 103.17857360839844,
+      "epoch": 1.2531468531468533,
+      "grad_norm": 0.1630398997063258,
+      "kl": 1.484375,
+      "learning_rate": 8.681228867254655e-05,
+      "loss": 0.0141,
+      "num_tokens": 3017147.0,
+      "reward": 3.0600602626800537,
+      "reward_std": 0.3837727904319763,
+      "rewards/check_gptzero_func": 1.8571428060531616,
+      "rewards/check_perplexity_diff_func": 0.2678571343421936,
+      "rewards/check_winston_local_func": 0.9350600838661194,
+      "step": 895
+    },
+    {
+      "clip_ratio": 0.027192149311304092,
+      "epoch": 1.2545454545454544,
+      "grad_norm": 0.13164071058906224,
+      "kl": 1.3984375,
+      "learning_rate": 8.677096444054213e-05,
+      "loss": 0.0089,
+      "step": 896
+    },
+    {
+      "clip_ratio": 0.007074595894664526,
+      "completion_length": 103.6964340209961,
+      "epoch": 1.255944055944056,
+      "grad_norm": 0.2371375881462628,
+      "kl": 1.359375,
+      "learning_rate": 8.672958543287666e-05,
+      "loss": -0.0156,
+      "num_tokens": 3037542.0,
+      "reward": 3.138697624206543,
+      "reward_std": 0.4123023450374603,
+      "rewards/check_gptzero_func": 1.8928571939468384,
+      "rewards/check_perplexity_diff_func": 0.2857142984867096,
+      "rewards/check_winston_local_func": 0.9601259231567383,
+      "step": 897
+    },
+    {
+      "clip_ratio": 0.014699095860123634,
+      "epoch": 1.2573426573426574,
+      "grad_norm": 0.17763525209114453,
+      "kl": 1.34375,
+      "learning_rate": 8.668815171119021e-05,
+      "loss": -0.0255,
+      "step": 898
+    },
+    {
+      "clip_ratio": 0.006138819735497236,
+      "completion_length": 100.75000762939453,
+      "epoch": 1.2587412587412588,
+      "grad_norm": 0.21431434335739866,
+      "kl": 1.390625,
+      "learning_rate": 8.664666333720425e-05,
+      "loss": -0.0293,
+      "num_tokens": 3057876.0,
+      "reward": 3.079195976257324,
+      "reward_std": 0.43540167808532715,
+      "rewards/check_gptzero_func": 1.8571428060531616,
+      "rewards/check_perplexity_diff_func": 0.2678571343421936,
+      "rewards/check_winston_local_func": 0.9541959762573242,
+      "step": 899
+    },
+    {
+      "clip_ratio": 0.017809001728892326,
+      "epoch": 1.2601398601398601,
+      "grad_norm": 0.16147149125792792,
+      "kl": 1.421875,
+      "learning_rate": 8.660512037272172e-05,
+      "loss": -0.0385,
+      "step": 900
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 2860,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 4,
+  "save_steps": 100,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 0.0,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/checkpoint-900/training_args.bin b/checkpoint-900/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..14e2d3ed154ee910c6c1698e855fc17a04eb6193
--- /dev/null
+++ b/checkpoint-900/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2c2b399ccd68443f851f396da81b1c9d32709e5be46a59f8f9eb00238e3b493d
+size 7480
diff --git a/checkpoint-900/zero_to_fp32.py b/checkpoint-900/zero_to_fp32.py
new file mode 100644
index 0000000000000000000000000000000000000000..0e759146cadd92ddfefab3680146c2bd6a2b5c04
--- /dev/null
+++ b/checkpoint-900/zero_to_fp32.py
@@ -0,0 +1,760 @@
+#!/usr/bin/env python
+
+# Copyright (c) Microsoft Corporation.
+# SPDX-License-Identifier: Apache-2.0
+
+# DeepSpeed Team
+
+# This script extracts fp32 consolidated weights from a zero 1, 2 and 3 DeepSpeed checkpoints. It gets
+# copied into the top level checkpoint dir, so the user can easily do the conversion at any point in
+# the future. Once extracted, the weights don't require DeepSpeed and can be used in any
+# application.
+#
+# example:
+#   python zero_to_fp32.py . output_dir/
+#   or
+#   python zero_to_fp32.py . output_dir/ --safe_serialization
+
+import argparse
+import torch
+import glob
+import math
+import os
+import re
+import gc
+import json
+import numpy as np
+from tqdm import tqdm
+from collections import OrderedDict
+from dataclasses import dataclass
+
+# while this script doesn't use deepspeed to recover data, since the checkpoints are pickled with
+# DeepSpeed data structures it has to be available in the current python environment.
+from deepspeed.utils import logger
+from deepspeed.checkpoint.constants import (DS_VERSION, OPTIMIZER_STATE_DICT, SINGLE_PARTITION_OF_FP32_GROUPS,
+                                            FP32_FLAT_GROUPS, ZERO_STAGE, PARTITION_COUNT, PARAM_SHAPES, BUFFER_NAMES,
+                                            FROZEN_PARAM_SHAPES, FROZEN_PARAM_FRAGMENTS)
+
+
+@dataclass
+class zero_model_state:
+    buffers: dict()
+    param_shapes: dict()
+    shared_params: list
+    ds_version: int
+    frozen_param_shapes: dict()
+    frozen_param_fragments: dict()
+
+
+debug = 0
+
+# load to cpu
+device = torch.device('cpu')
+
+
+def atoi(text):
+    return int(text) if text.isdigit() else text
+
+
+def natural_keys(text):
+    '''
+    alist.sort(key=natural_keys) sorts in human order
+    http://nedbatchelder.com/blog/200712/human_sorting.html
+    (See Toothy's implementation in the comments)
+    '''
+    return [atoi(c) for c in re.split(r'(\d+)', text)]
+
+
+def get_model_state_file(checkpoint_dir, zero_stage):
+    if not os.path.isdir(checkpoint_dir):
+        raise FileNotFoundError(f"Directory '{checkpoint_dir}' doesn't exist")
+
+    # there should be only one file
+    if zero_stage <= 2:
+        file = os.path.join(checkpoint_dir, "mp_rank_00_model_states.pt")
+    elif zero_stage == 3:
+        file = os.path.join(checkpoint_dir, "zero_pp_rank_0_mp_rank_00_model_states.pt")
+
+    if not os.path.exists(file):
+        raise FileNotFoundError(f"can't find model states file at '{file}'")
+
+    return file
+
+
+def get_checkpoint_files(checkpoint_dir, glob_pattern):
+    # XXX: need to test that this simple glob rule works for multi-node setup too
+    ckpt_files = sorted(glob.glob(os.path.join(checkpoint_dir, glob_pattern)), key=natural_keys)
+
+    if len(ckpt_files) == 0:
+        raise FileNotFoundError(f"can't find {glob_pattern} files in directory '{checkpoint_dir}'")
+
+    return ckpt_files
+
+
+def get_optim_files(checkpoint_dir):
+    return get_checkpoint_files(checkpoint_dir, "*_optim_states.pt")
+
+
+def get_model_state_files(checkpoint_dir):
+    return get_checkpoint_files(checkpoint_dir, "*_model_states.pt")
+
+
+def parse_model_states(files):
+    zero_model_states = []
+    for file in files:
+        state_dict = torch.load(file, map_location=device, weights_only=False)
+
+        if BUFFER_NAMES not in state_dict:
+            raise ValueError(f"{file} is not a model state checkpoint")
+        buffer_names = state_dict[BUFFER_NAMES]
+        if debug:
+            print("Found buffers:", buffer_names)
+
+        # recover just the buffers while restoring them to fp32 if they were saved in fp16
+        buffers = {k: v.float() for k, v in state_dict["module"].items() if k in buffer_names}
+        param_shapes = state_dict[PARAM_SHAPES]
+
+        # collect parameters that are included in param_shapes
+        param_names = []
+        for s in param_shapes:
+            for name in s.keys():
+                param_names.append(name)
+
+        # update with frozen parameters
+        frozen_param_shapes = state_dict.get(FROZEN_PARAM_SHAPES, None)
+        if frozen_param_shapes is not None:
+            if debug:
+                print(f"Found frozen_param_shapes: {frozen_param_shapes}")
+            param_names += list(frozen_param_shapes.keys())
+
+        # handle shared params
+        shared_params = [[k, v] for k, v in state_dict["shared_params"].items()]
+
+        ds_version = state_dict.get(DS_VERSION, None)
+
+        frozen_param_fragments = state_dict.get(FROZEN_PARAM_FRAGMENTS, None)
+
+        z_model_state = zero_model_state(buffers=buffers,
+                                         param_shapes=param_shapes,
+                                         shared_params=shared_params,
+                                         ds_version=ds_version,
+                                         frozen_param_shapes=frozen_param_shapes,
+                                         frozen_param_fragments=frozen_param_fragments)
+        zero_model_states.append(z_model_state)
+
+    return zero_model_states
+
+
+def parse_optim_states(files, ds_checkpoint_dir):
+    total_files = len(files)
+    state_dicts = []
+    for f in tqdm(files, desc='Loading checkpoint shards'):
+        state_dict = torch.load(f, map_location=device, mmap=True, weights_only=False)
+        # immediately discard the potentially huge 2 optimizer states as we only care for fp32 master weights
+        # and also handle the case where it was already removed by another helper script
+        state_dict["optimizer_state_dict"].pop("optimizer_state_dict", None)
+        state_dicts.append(state_dict)
+
+    if not ZERO_STAGE in state_dicts[0][OPTIMIZER_STATE_DICT]:
+        raise ValueError(f"{files[0]} is not a zero checkpoint")
+    zero_stage = state_dicts[0][OPTIMIZER_STATE_DICT][ZERO_STAGE]
+    world_size = state_dicts[0][OPTIMIZER_STATE_DICT][PARTITION_COUNT]
+
+    # For ZeRO-2 each param group can have different partition_count as data parallelism for expert
+    # parameters can be different from data parallelism for non-expert parameters. So we can just
+    # use the max of the partition_count to get the dp world_size.
+
+    if type(world_size) is list:
+        world_size = max(world_size)
+
+    if world_size != total_files:
+        raise ValueError(
+            f"Expected {world_size} of '*_optim_states.pt' under '{ds_checkpoint_dir}' but found {total_files} files. "
+            "Possibly due to an overwrite of an old checkpoint, or a checkpoint didn't get saved by one or more processes."
+        )
+
+    # the groups are named differently in each stage
+    if zero_stage <= 2:
+        fp32_groups_key = SINGLE_PARTITION_OF_FP32_GROUPS
+    elif zero_stage == 3:
+        fp32_groups_key = FP32_FLAT_GROUPS
+    else:
+        raise ValueError(f"unknown zero stage {zero_stage}")
+
+    fp32_flat_groups = [state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key] for i in range(len(state_dicts))]
+    return zero_stage, world_size, fp32_flat_groups
+
+
+def _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters):
+    """
+    Returns fp32 state_dict reconstructed from ds checkpoint
+
+    Args:
+        - ``ds_checkpoint_dir``: path to the deepspeed checkpoint folder (where the optimizer files are)
+
+    """
+    print(f"Processing zero checkpoint '{ds_checkpoint_dir}'")
+
+    optim_files = get_optim_files(ds_checkpoint_dir)
+    zero_stage, world_size, fp32_flat_groups = parse_optim_states(optim_files, ds_checkpoint_dir)
+    print(f"Detected checkpoint of type zero stage {zero_stage}, world_size: {world_size}")
+
+    model_files = get_model_state_files(ds_checkpoint_dir)
+
+    zero_model_states = parse_model_states(model_files)
+    print(f'Parsing checkpoint created by deepspeed=={zero_model_states[0].ds_version}')
+
+    if zero_stage <= 2:
+        return _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states,
+                                                          exclude_frozen_parameters)
+    elif zero_stage == 3:
+        return _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states,
+                                                          exclude_frozen_parameters)
+
+
+def _zero2_merge_frozen_params(state_dict, zero_model_states):
+    if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0:
+        return
+
+    frozen_param_shapes = zero_model_states[0].frozen_param_shapes
+    frozen_param_fragments = zero_model_states[0].frozen_param_fragments
+
+    if debug:
+        num_elem = sum(s.numel() for s in frozen_param_shapes.values())
+        print(f'rank 0: {FROZEN_PARAM_SHAPES}.numel = {num_elem}')
+
+        wanted_params = len(frozen_param_shapes)
+        wanted_numel = sum(s.numel() for s in frozen_param_shapes.values())
+        avail_numel = sum([p.numel() for p in frozen_param_fragments.values()])
+        print(f'Frozen params: Have {avail_numel} numels to process.')
+        print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params')
+
+    total_params = 0
+    total_numel = 0
+    for name, shape in frozen_param_shapes.items():
+        total_params += 1
+        unpartitioned_numel = shape.numel()
+        total_numel += unpartitioned_numel
+
+        state_dict[name] = frozen_param_fragments[name]
+
+        if debug:
+            print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ")
+
+    print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements")
+
+
+def _has_callable(obj, fn):
+    attr = getattr(obj, fn, None)
+    return callable(attr)
+
+
+def _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states):
+    param_shapes = zero_model_states[0].param_shapes
+
+    # Reconstruction protocol:
+    #
+    # XXX: document this
+
+    if debug:
+        for i in range(world_size):
+            for j in range(len(fp32_flat_groups[0])):
+                print(f"{FP32_FLAT_GROUPS}[{i}][{j}].shape={fp32_flat_groups[i][j].shape}")
+
+    # XXX: memory usage doubles here (zero2)
+    num_param_groups = len(fp32_flat_groups[0])
+    merged_single_partition_of_fp32_groups = []
+    for i in range(num_param_groups):
+        merged_partitions = [sd[i] for sd in fp32_flat_groups]
+        full_single_fp32_vector = torch.cat(merged_partitions, 0)
+        merged_single_partition_of_fp32_groups.append(full_single_fp32_vector)
+    avail_numel = sum(
+        [full_single_fp32_vector.numel() for full_single_fp32_vector in merged_single_partition_of_fp32_groups])
+
+    if debug:
+        wanted_params = sum([len(shapes) for shapes in param_shapes])
+        wanted_numel = sum([sum(shape.numel() for shape in shapes.values()) for shapes in param_shapes])
+        # not asserting if there is a mismatch due to possible padding
+        print(f"Have {avail_numel} numels to process.")
+        print(f"Need {wanted_numel} numels in {wanted_params} params.")
+
+    # params
+    # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support
+    # out-of-core computing solution
+    total_numel = 0
+    total_params = 0
+    for shapes, full_single_fp32_vector in zip(param_shapes, merged_single_partition_of_fp32_groups):
+        offset = 0
+        avail_numel = full_single_fp32_vector.numel()
+        for name, shape in shapes.items():
+
+            unpartitioned_numel = shape.numel() if _has_callable(shape, 'numel') else math.prod(shape)
+            total_numel += unpartitioned_numel
+            total_params += 1
+
+            if debug:
+                print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ")
+            state_dict[name] = full_single_fp32_vector.narrow(0, offset, unpartitioned_numel).view(shape)
+            offset += unpartitioned_numel
+
+        # Z2 started to align to 2*world_size to improve nccl performance. Therefore both offset and
+        # avail_numel can differ by anywhere between 0..2*world_size. Due to two unrelated complex
+        # paddings performed in the code it's almost impossible to predict the exact numbers w/o the
+        # live optimizer object, so we are checking that the numbers are within the right range
+        align_to = 2 * world_size
+
+        def zero2_align(x):
+            return align_to * math.ceil(x / align_to)
+
+        if debug:
+            print(f"original offset={offset}, avail_numel={avail_numel}")
+
+        offset = zero2_align(offset)
+        avail_numel = zero2_align(avail_numel)
+
+        if debug:
+            print(f"aligned  offset={offset}, avail_numel={avail_numel}")
+
+        # Sanity check
+        if offset != avail_numel:
+            raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong")
+
+    print(f"Reconstructed fp32 state dict with {total_params} params {total_numel} elements")
+
+
+def _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states,
+                                               exclude_frozen_parameters):
+    state_dict = OrderedDict()
+
+    # buffers
+    buffers = zero_model_states[0].buffers
+    state_dict.update(buffers)
+    if debug:
+        print(f"added {len(buffers)} buffers")
+
+    if not exclude_frozen_parameters:
+        _zero2_merge_frozen_params(state_dict, zero_model_states)
+
+    _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states)
+
+    # recover shared parameters
+    for pair in zero_model_states[0].shared_params:
+        if pair[1] in state_dict:
+            state_dict[pair[0]] = state_dict[pair[1]]
+
+    return state_dict
+
+
+def zero3_partitioned_param_info(unpartitioned_numel, world_size):
+    remainder = unpartitioned_numel % world_size
+    padding_numel = (world_size - remainder) if remainder else 0
+    partitioned_numel = math.ceil(unpartitioned_numel / world_size)
+    return partitioned_numel, padding_numel
+
+
+def _zero3_merge_frozen_params(state_dict, world_size, zero_model_states):
+    if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0:
+        return
+
+    if debug:
+        for i in range(world_size):
+            num_elem = sum(s.numel() for s in zero_model_states[i].frozen_param_fragments.values())
+            print(f'rank {i}: {FROZEN_PARAM_SHAPES}.numel = {num_elem}')
+
+        frozen_param_shapes = zero_model_states[0].frozen_param_shapes
+        wanted_params = len(frozen_param_shapes)
+        wanted_numel = sum(s.numel() for s in frozen_param_shapes.values())
+        avail_numel = sum([p.numel() for p in zero_model_states[0].frozen_param_fragments.values()]) * world_size
+        print(f'Frozen params: Have {avail_numel} numels to process.')
+        print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params')
+
+    total_params = 0
+    total_numel = 0
+    for name, shape in zero_model_states[0].frozen_param_shapes.items():
+        total_params += 1
+        unpartitioned_numel = shape.numel()
+        total_numel += unpartitioned_numel
+
+        param_frags = tuple(model_state.frozen_param_fragments[name] for model_state in zero_model_states)
+        state_dict[name] = torch.cat(param_frags, 0).narrow(0, 0, unpartitioned_numel).view(shape)
+
+        partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size)
+
+        if debug:
+            print(
+                f"Frozen params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}"
+            )
+
+    print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements")
+
+
+class GatheredTensor:
+    """
+    A pseudo tensor that collects partitioned weights.
+    It is more memory efficient when there are multiple groups.
+    """
+
+    def __init__(self, flat_groups, flat_groups_offset, offset, partitioned_numel, shape):
+        self.flat_groups = flat_groups
+        self.flat_groups_offset = flat_groups_offset
+        self.offset = offset
+        self.partitioned_numel = partitioned_numel
+        self.shape = shape
+        self.dtype = self.flat_groups[0][0].dtype
+
+    def contiguous(self):
+        """
+        Merge partitioned weights from flat_groups into a single tensor.
+        """
+        end_idx = self.offset + self.partitioned_numel
+        world_size = len(self.flat_groups)
+        pad_flat_param_chunks = []
+
+        for rank_i in range(world_size):
+            # for each rank, we need to collect weights from related group/groups
+            flat_groups_at_rank_i = self.flat_groups[rank_i]
+            start_group_id = None
+            end_group_id = None
+            for group_id in range(len(self.flat_groups_offset)):
+                if self.flat_groups_offset[group_id] <= self.offset < self.flat_groups_offset[group_id + 1]:
+                    start_group_id = group_id
+                if self.flat_groups_offset[group_id] < end_idx <= self.flat_groups_offset[group_id + 1]:
+                    end_group_id = group_id
+                    break
+            # collect weights from related group/groups
+            for group_id in range(start_group_id, end_group_id + 1):
+                flat_tensor = flat_groups_at_rank_i[group_id]
+                start_offset = self.offset - self.flat_groups_offset[group_id]
+                end_offset = min(end_idx, self.flat_groups_offset[group_id + 1]) - self.flat_groups_offset[group_id]
+                pad_flat_param_chunks.append(flat_tensor[start_offset:end_offset])
+
+        # collect weights from all ranks
+        pad_flat_param = torch.cat(pad_flat_param_chunks, dim=0)
+        param = pad_flat_param[:self.shape.numel()].view(self.shape).contiguous()
+        return param
+
+
+def _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states):
+    param_shapes = zero_model_states[0].param_shapes
+    avail_numel = sum([flat_group.numel() for flat_group in fp32_flat_groups[0]]) * world_size
+
+    # Reconstruction protocol: For zero3 we need to zip the partitions together at boundary of each
+    # param, re-consolidating each param, while dealing with padding if any
+
+    # merge list of dicts, preserving order
+    param_shapes = {k: v for d in param_shapes for k, v in d.items()}
+
+    if debug:
+        for i in range(world_size):
+            print(f"{FP32_FLAT_GROUPS}[{i}].shape={fp32_flat_groups[i].shape}")
+
+        wanted_params = len(param_shapes)
+        wanted_numel = sum(shape.numel() for shape in param_shapes.values())
+        # not asserting if there is a mismatch due to possible padding
+        avail_numel = fp32_flat_groups[0].numel() * world_size
+        print(f"Trainable params: Have {avail_numel} numels to process.")
+        print(f"Trainable params: Need {wanted_numel} numels in {wanted_params} params.")
+
+    # params
+    # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support
+    # out-of-core computing solution
+    offset = 0
+    total_numel = 0
+    total_params = 0
+    flat_groups_offset = [0] + list(np.cumsum([flat_tensor.numel() for flat_tensor in fp32_flat_groups[0]]))
+    for name, shape in tqdm(param_shapes.items(), desc='Gathering sharded weights'):
+        unpartitioned_numel = shape.numel()
+        total_numel += unpartitioned_numel
+        total_params += 1
+        partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size)
+
+        if debug:
+            print(
+                f"Trainable params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}"
+            )
+
+        # memory efficient tensor
+        tensor = GatheredTensor(fp32_flat_groups, flat_groups_offset, offset, partitioned_numel, shape)
+        state_dict[name] = tensor
+        offset += partitioned_numel
+
+    offset *= world_size
+
+    # Sanity check
+    if offset != avail_numel:
+        raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong")
+
+    print(f"Reconstructed Trainable fp32 state dict with {total_params} params {total_numel} elements")
+
+
+def _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states,
+                                               exclude_frozen_parameters):
+    state_dict = OrderedDict()
+
+    # buffers
+    buffers = zero_model_states[0].buffers
+    state_dict.update(buffers)
+    if debug:
+        print(f"added {len(buffers)} buffers")
+
+    if not exclude_frozen_parameters:
+        _zero3_merge_frozen_params(state_dict, world_size, zero_model_states)
+
+    _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states)
+
+    # recover shared parameters
+    for pair in zero_model_states[0].shared_params:
+        if pair[1] in state_dict:
+            state_dict[pair[0]] = state_dict[pair[1]]
+
+    return state_dict
+
+
+def to_torch_tensor(state_dict, return_empty_tensor=False):
+    """
+    Convert state_dict of GatheredTensor to torch tensor
+    """
+    torch_state_dict = {}
+    converted_tensors = {}
+    for name, tensor in state_dict.items():
+        tensor_id = id(tensor)
+        if tensor_id in converted_tensors:  # shared tensors
+            shared_tensor = torch_state_dict[converted_tensors[tensor_id]]
+            torch_state_dict[name] = shared_tensor
+        else:
+            converted_tensors[tensor_id] = name
+            if return_empty_tensor:
+                torch_state_dict[name] = torch.empty(tensor.shape, dtype=tensor.dtype)
+            else:
+                torch_state_dict[name] = tensor.contiguous()
+    return torch_state_dict
+
+
+def get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir,
+                                             tag=None,
+                                             exclude_frozen_parameters=False,
+                                             lazy_mode=False):
+    """
+    Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated state_dict that can be loaded with
+    ``load_state_dict()`` and used for training without DeepSpeed or shared with others, for example
+    via a model hub.
+
+    Args:
+        - ``checkpoint_dir``: path to the desired checkpoint folder
+        - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in 'latest' file. e.g., ``global_step14``
+        - ``exclude_frozen_parameters``: exclude frozen parameters
+        - ``lazy_mode``: get state_dict in lazy mode. It returns a dict of pesduo tensor instead of torch tensor, which is more memory efficient.
+          Convert the pesduo tensor to torch tensor by ``.contiguous()``
+
+    Returns:
+        - pytorch ``state_dict``
+
+    A typical usage might be ::
+
+        from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint
+        # do the training and checkpoint saving
+        state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir) # already on cpu
+        model = model.cpu() # move to cpu
+        model.load_state_dict(state_dict)
+        # submit to model hub or save the model to share with others
+
+    In this example the ``model`` will no longer be usable in the deepspeed context of the same
+    application. i.e. you will need to re-initialize the deepspeed engine, since
+    ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it.
+
+    If you want it all done for you, use ``load_state_dict_from_zero_checkpoint`` instead.
+
+    Note: the above usage may not work if your application doesn't have sufficient free CPU memory.
+    You may need to use the offline approach using the ``zero_to_fp32.py`` script that is saved with
+    the checkpoint. Or you can load state_dict in lazy mode ::
+
+        from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint
+        state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, lazy_mode=True) # not on cpu
+        for name, lazy_tensor in state_dict.item():
+            tensor = lazy_tensor.contiguous()  # to cpu
+            print(name, tensor)
+            # del tensor to release memory if it no longer in use
+    """
+    if tag is None:
+        latest_path = os.path.join(checkpoint_dir, 'latest')
+        if os.path.isfile(latest_path):
+            with open(latest_path, 'r') as fd:
+                tag = fd.read().strip()
+        else:
+            raise ValueError(f"Unable to find 'latest' file at {latest_path}")
+
+    ds_checkpoint_dir = os.path.join(checkpoint_dir, tag)
+
+    if not os.path.isdir(ds_checkpoint_dir):
+        raise FileNotFoundError(f"Directory '{ds_checkpoint_dir}' doesn't exist")
+
+    state_dict = _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters)
+    if lazy_mode:
+        return state_dict
+    else:
+        return to_torch_tensor(state_dict)
+
+
+def convert_zero_checkpoint_to_fp32_state_dict(checkpoint_dir,
+                                               output_dir,
+                                               max_shard_size="5GB",
+                                               safe_serialization=False,
+                                               tag=None,
+                                               exclude_frozen_parameters=False):
+    """
+    Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` file that can be
+    loaded with ``torch.load(file)`` + ``load_state_dict()`` and used for training without DeepSpeed.
+
+    Args:
+        - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``)
+        - ``output_dir``: directory to the pytorch fp32 state_dict output files
+        - ``max_shard_size``: the maximum size for a checkpoint before being sharded, default value is 5GB
+        - ``safe_serialization``:  whether to save the model using `safetensors` or the traditional PyTorch way (that uses `pickle`).
+        - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14``
+        - ``exclude_frozen_parameters``: exclude frozen parameters
+    """
+
+    # Dependency pre-check
+    if safe_serialization:
+        try:
+            from safetensors.torch import save_file
+        except ImportError:
+            print('If you want to use `safe_serialization`, please `pip install safetensors`')
+            raise
+    if max_shard_size is not None:
+        try:
+            from huggingface_hub import split_torch_state_dict_into_shards
+        except ImportError:
+            print('If you want to use `max_shard_size`, please `pip install huggingface_hub`')
+            raise
+
+    # Convert zero checkpoint to state_dict
+    state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir,
+                                                          tag,
+                                                          exclude_frozen_parameters,
+                                                          lazy_mode=True)
+
+    # Shard the model if it is too big.
+    weights_name = "model.safetensors" if safe_serialization else "pytorch_model.bin"
+    if max_shard_size is not None:
+        filename_pattern = weights_name.replace(".bin", "{suffix}.bin").replace(".safetensors", "{suffix}.safetensors")
+        # an memory-efficient approach for sharding
+        empty_state_dict = to_torch_tensor(state_dict, return_empty_tensor=True)
+        state_dict_split = split_torch_state_dict_into_shards(empty_state_dict,
+                                                              filename_pattern=filename_pattern,
+                                                              max_shard_size=max_shard_size)
+    else:
+        from collections import namedtuple
+        StateDictSplit = namedtuple("StateDictSplit", ["is_sharded", "filename_to_tensors"])
+        state_dict_split = StateDictSplit(is_sharded=False,
+                                          filename_to_tensors={weights_name: list(state_dict.keys())})
+
+    # Save the model by shard
+    os.makedirs(output_dir, exist_ok=True)
+    filename_to_tensors = state_dict_split.filename_to_tensors.items()
+    for shard_file, tensors in tqdm(filename_to_tensors, desc="Saving checkpoint shards"):
+        shard_state_dict = {tensor_name: state_dict[tensor_name] for tensor_name in tensors}
+        shard_state_dict = to_torch_tensor(shard_state_dict)
+        output_path = os.path.join(output_dir, shard_file)
+        if safe_serialization:
+            save_file(shard_state_dict, output_path, metadata={"format": "pt"})
+        else:
+            torch.save(shard_state_dict, output_path)
+        # release the memory of current shard
+        for tensor_name in list(shard_state_dict.keys()):
+            del state_dict[tensor_name]
+            del shard_state_dict[tensor_name]
+        del shard_state_dict
+        gc.collect()
+
+    # Save index if sharded
+    if state_dict_split.is_sharded:
+        index = {
+            "metadata": state_dict_split.metadata,
+            "weight_map": state_dict_split.tensor_to_filename,
+        }
+        save_index_file = "model.safetensors.index.json" if safe_serialization else "pytorch_model.bin.index.json"
+        save_index_file = os.path.join(output_dir, save_index_file)
+        with open(save_index_file, "w", encoding="utf-8") as f:
+            content = json.dumps(index, indent=2, sort_keys=True) + "\n"
+            f.write(content)
+
+
+def load_state_dict_from_zero_checkpoint(model, checkpoint_dir, tag=None):
+    """
+    1. Put the provided model to cpu
+    2. Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict``
+    3. Load it into the provided model
+
+    Args:
+        - ``model``: the model object to update
+        - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``)
+        - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14``
+
+    Returns:
+        - ``model`: modified model
+
+    Make sure you have plenty of CPU memory available before you call this function. If you don't
+    have enough use the ``zero_to_fp32.py`` utility to do the conversion. You will find it
+    conveniently placed for you in the checkpoint folder.
+
+    A typical usage might be ::
+
+        from deepspeed.utils.zero_to_fp32 import load_state_dict_from_zero_checkpoint
+        model = load_state_dict_from_zero_checkpoint(trainer.model, checkpoint_dir)
+        # submit to model hub or save the model to share with others
+
+    Note, that once this was run, the ``model`` will no longer be usable in the deepspeed context
+    of the same application. i.e. you will need to re-initialize the deepspeed engine, since
+    ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it.
+
+    """
+    logger.info(f"Extracting fp32 weights")
+    state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag)
+
+    logger.info(f"Overwriting model with fp32 weights")
+    model = model.cpu()
+    model.load_state_dict(state_dict, strict=False)
+
+    return model
+
+
+if __name__ == "__main__":
+    parser = argparse.ArgumentParser()
+    parser.add_argument("checkpoint_dir",
+                        type=str,
+                        help="path to the desired checkpoint folder, e.g., path/checkpoint-12")
+    parser.add_argument("output_dir",
+                        type=str,
+                        help="directory to the pytorch fp32 state_dict output files"
+                        "(e.g. path/checkpoint-12-output/)")
+    parser.add_argument(
+        "--max_shard_size",
+        type=str,
+        default="5GB",
+        help="The maximum size for a checkpoint before being sharded. Checkpoints shard will then be each of size"
+        "lower than this size. If expressed as a string, needs to be digits followed by a unit (like `5MB`"
+        "We default it to 5GB in order for models to be able to run easily on free-tier google colab instances"
+        "without CPU OOM issues.")
+    parser.add_argument(
+        "--safe_serialization",
+        default=False,
+        action='store_true',
+        help="Whether to save the model using `safetensors` or the traditional PyTorch way (that uses `pickle`).")
+    parser.add_argument("-t",
+                        "--tag",
+                        type=str,
+                        default=None,
+                        help="checkpoint tag used as a unique identifier for checkpoint. e.g., global_step1")
+    parser.add_argument("--exclude_frozen_parameters", action='store_true', help="exclude frozen parameters")
+    parser.add_argument("-d", "--debug", action='store_true', help="enable debug")
+    args = parser.parse_args()
+
+    debug = args.debug
+
+    convert_zero_checkpoint_to_fp32_state_dict(args.checkpoint_dir,
+                                               args.output_dir,
+                                               max_shard_size=args.max_shard_size,
+                                               safe_serialization=args.safe_serialization,
+                                               tag=args.tag,
+                                               exclude_frozen_parameters=args.exclude_frozen_parameters)