Model save

README.md

@@ -0,0 +1,81 @@
+---
+license: apache-2.0
+base_model: alignment-handbook/zephyr-7b-sft-full
+tags:
+- trl
+- dpo
+- generated_from_trainer
+model-index:
+- name: zephyr-7b-dpo-full
+  results: []
+---
+
+<!-- This model card has been generated automatically according to the information the Trainer had access to. You
+should probably proofread and complete it, then remove this comment. -->
+
+# zephyr-7b-dpo-full
+
+This model is a fine-tuned version of [alignment-handbook/zephyr-7b-sft-full](https://huggingface.co/alignment-handbook/zephyr-7b-sft-full) on an unknown dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.4906
+- Rewards/chosen: -1.5814
+- Rewards/rejected: -2.6280
+- Rewards/accuracies: 0.7698
+- Rewards/margins: 1.0467
+- Logps/rejected: -523.0108
+- Logps/chosen: -440.1050
+- Logits/rejected: 0.7152
+- Logits/chosen: -0.3568
+
+## Model description
+
+More information needed
+
+## Intended uses & limitations
+
+More information needed
+
+## Training and evaluation data
+
+More information needed
+
+## Training procedure
+
+### Training hyperparameters
+
+The following hyperparameters were used during training:
+- learning_rate: 5e-07
+- train_batch_size: 8
+- eval_batch_size: 8
+- seed: 42
+- distributed_type: multi-GPU
+- num_devices: 4
+- gradient_accumulation_steps: 2
+- total_train_batch_size: 64
+- total_eval_batch_size: 32
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: cosine
+- lr_scheduler_warmup_ratio: 0.1
+- num_epochs: 1
+
+### Training results
+
+| Training Loss | Epoch  | Step | Validation Loss | Rewards/chosen | Rewards/rejected | Rewards/accuracies | Rewards/margins | Logps/rejected | Logps/chosen | Logits/rejected | Logits/chosen |
+|:-------------:|:------:|:----:|:---------------:|:--------------:|:----------------:|:------------------:|:---------------:|:--------------:|:------------:|:---------------:|:-------------:|
+| 0.6267        | 0.1047 | 100  | 0.6375          | 0.0143         | -0.1443          | 0.7063             | 0.1587          | -274.6408      | -280.5349    | -2.4244         | -2.4989       |
+| 0.578         | 0.2093 | 200  | 0.5544          | -0.7566        | -1.3732          | 0.7480             | 0.6167          | -397.5314      | -357.6234    | -1.3956         | -1.7000       |
+| 0.5301        | 0.3140 | 300  | 0.5198          | -0.9545        | -1.7880          | 0.7758             | 0.8335          | -439.0061      | -377.4185    | -0.0799         | -0.7774       |
+| 0.5386        | 0.4186 | 400  | 0.5105          | -1.3034        | -2.1080          | 0.7718             | 0.8046          | -471.0027      | -412.3046    | 0.9407          | 0.0250        |
+| 0.4996        | 0.5233 | 500  | 0.4953          | -1.7843        | -2.7473          | 0.7679             | 0.9630          | -534.9354      | -460.4000    | 1.1533          | 0.0059        |
+| 0.4664        | 0.6279 | 600  | 0.4944          | -1.6883        | -2.7073          | 0.7698             | 1.0191          | -530.9401      | -450.7948    | 1.0557          | -0.2689       |
+| 0.4716        | 0.7326 | 700  | 0.4929          | -1.5680        | -2.6003          | 0.7778             | 1.0324          | -520.2406      | -438.7653    | 0.7091          | -0.3818       |
+| 0.4741        | 0.8373 | 800  | 0.4907          | -1.5504        | -2.5743          | 0.7778             | 1.0239          | -517.6415      | -437.0119    | 0.6599          | -0.3941       |
+| 0.4741        | 0.9419 | 900  | 0.4906          | -1.5814        | -2.6280          | 0.7698             | 1.0467          | -523.0108      | -440.1050    | 0.7152          | -0.3568       |
+
+
+### Framework versions
+
+- Transformers 4.40.1
+- Pytorch 2.1.2+cu121
+- Datasets 2.19.0
+- Tokenizers 0.19.1

all_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 0.9994767137624281,
+    "total_flos": 0.0,
+    "train_loss": 0.5247097397349891,
+    "train_runtime": 18219.8315,
+    "train_samples": 61134,
+    "train_samples_per_second": 3.355,
+    "train_steps_per_second": 0.052
+}

generation_config.json

@@ -0,0 +1,6 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 1,
+  "eos_token_id": 2,
+  "transformers_version": "4.40.1"
+}

model-00001-of-00003.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9bc27b979262e05c38af1babd0a4a1d2741b1c9eb96d39e73f9768fa7eb5c3bb
+size 4943162336

model-00002-of-00003.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8563e99eeea3128a306b0d0ebcbdd4223b20a4f83ae5ce6c88a293c5e10b96c2
+size 4999819336

model-00003-of-00003.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dd18869baf9c51e0521149959977886cf788ec73ec4d5b01c4799c9d08c6ddee
+size 4540516344

model.safetensors.index.json

@@ -0,0 +1,298 @@
+{
+  "metadata": {
+    "total_size": 14483464192
+  },
+  "weight_map": {
+    "lm_head.weight": "model-00003-of-00003.safetensors",
+    "model.embed_tokens.weight": "model-00001-of-00003.safetensors",
+    "model.layers.0.input_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.0.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.0.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.0.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.1.input_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.1.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.1.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.1.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.1.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.1.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.1.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.10.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.10.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.10.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.10.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.10.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.10.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.10.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.10.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.10.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.11.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.11.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.11.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.11.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.11.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.11.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.11.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.11.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.11.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.12.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.12.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.12.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.12.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.12.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.12.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.12.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.12.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.12.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.13.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.13.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.13.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.13.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.13.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.13.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.13.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.13.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.13.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.14.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.14.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.14.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.14.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.14.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.14.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.14.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.14.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.14.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.15.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.15.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.15.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.15.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.15.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.15.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.15.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.15.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.15.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.16.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.16.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.16.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.16.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.16.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.16.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.16.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.16.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.16.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.17.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.17.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.17.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.17.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.17.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.17.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.17.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.17.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.17.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.18.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.18.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.18.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.18.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.18.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.18.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.18.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.18.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.18.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.19.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.19.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.19.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.19.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.19.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.19.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.19.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.19.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.19.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.2.input_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.2.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.2.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.2.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.2.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.2.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.2.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.2.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.2.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.20.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.20.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.20.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.20.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.20.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.20.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.20.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.20.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.20.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.21.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.21.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.21.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.21.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.21.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.21.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.21.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.21.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.21.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.22.input_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.22.mlp.down_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.22.mlp.gate_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.22.mlp.up_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.22.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.22.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.22.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.22.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.22.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.23.input_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.23.mlp.down_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.23.mlp.gate_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.23.mlp.up_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.23.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.23.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.23.self_attn.o_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.23.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.23.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.24.input_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.24.mlp.down_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.24.mlp.gate_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.24.mlp.up_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.24.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.24.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.24.self_attn.o_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.24.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.24.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.25.input_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.25.mlp.down_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.25.mlp.gate_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.25.mlp.up_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.25.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.25.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.25.self_attn.o_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.25.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.25.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.26.input_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.26.mlp.down_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.26.mlp.gate_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.26.mlp.up_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.26.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.26.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.26.self_attn.o_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.26.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.26.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.27.input_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.27.mlp.down_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.27.mlp.gate_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.27.mlp.up_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.27.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.27.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.27.self_attn.o_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.27.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.27.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.28.input_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.28.mlp.down_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.28.mlp.gate_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.28.mlp.up_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.28.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.28.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.28.self_attn.o_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.28.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.28.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.29.input_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.29.mlp.down_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.29.mlp.gate_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.29.mlp.up_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.29.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.29.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.29.self_attn.o_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.29.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.29.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.3.input_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.3.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.3.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.3.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.3.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.3.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.3.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.3.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.3.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.30.input_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.30.mlp.down_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.30.mlp.gate_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.30.mlp.up_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.30.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.30.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.30.self_attn.o_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.30.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.30.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.31.input_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.31.mlp.down_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.31.mlp.gate_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.31.mlp.up_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.31.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.31.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.31.self_attn.o_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.31.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.31.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.4.input_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.4.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.4.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.4.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.4.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.4.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.4.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.4.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.4.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.5.input_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.5.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.5.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.5.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.5.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.5.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.5.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.5.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.5.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.6.input_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.6.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.6.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.6.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.6.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.6.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.6.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.6.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.6.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.7.input_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.7.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.7.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.7.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.7.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.7.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.7.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.7.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.7.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.8.input_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.8.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.8.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.8.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.8.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.8.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.8.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.8.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.8.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.9.input_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.9.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.9.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.9.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.9.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.9.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.9.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.9.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.9.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "model.norm.weight": "model-00003-of-00003.safetensors"
+  }
+}

runs/Apr25_13-59-39_dsw-3393-565b8f59b6-l5p7k/events.out.tfevents.1714054068.dsw-3393-565b8f59b6-l5p7k.12328.0

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:00855262bccd96c73c624a2908a4bbddb39c4fe490b7a8df7f112d1811bbed6b
-size 73579
+oid sha256:95527bae5d18de11d7166531f381a0093ce883b752f3c8fcbf34e1b2aac94ac7
+size 77373

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 0.9994767137624281,
+    "total_flos": 0.0,
+    "train_loss": 0.5247097397349891,
+    "train_runtime": 18219.8315,
+    "train_samples": 61134,
+    "train_samples_per_second": 3.355,
+    "train_steps_per_second": 0.052
+}

trainer_state.json

@@ -0,0 +1,1614 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 0.9994767137624281,
+  "eval_steps": 100,
+  "global_step": 955,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0010465724751439038,
+      "grad_norm": 12.735772520921401,
+      "learning_rate": 5.208333333333333e-09,
+      "logits/chosen": -2.924262046813965,
+      "logits/rejected": -2.7925047874450684,
+      "logps/chosen": -380.8447570800781,
+      "logps/rejected": -358.51123046875,
+      "loss": 0.6931,
+      "rewards/accuracies": 0.0,
+      "rewards/chosen": 0.0,
+      "rewards/margins": 0.0,
+      "rewards/rejected": 0.0,
+      "step": 1
+    },
+    {
+      "epoch": 0.010465724751439037,
+      "grad_norm": 10.939955073523928,
+      "learning_rate": 5.208333333333333e-08,
+      "logits/chosen": -2.5958893299102783,
+      "logits/rejected": -2.5695536136627197,
+      "logps/chosen": -256.6239318847656,
+      "logps/rejected": -234.886962890625,
+      "loss": 0.6932,
+      "rewards/accuracies": 0.5,
+      "rewards/chosen": 0.00025410810485482216,
+      "rewards/margins": 0.00034702278207987547,
+      "rewards/rejected": -9.291467722505331e-05,
+      "step": 10
+    },
+    {
+      "epoch": 0.020931449502878074,
+      "grad_norm": 12.592291847861857,
+      "learning_rate": 1.0416666666666667e-07,
+      "logits/chosen": -2.6138570308685303,
+      "logits/rejected": -2.5763792991638184,
+      "logps/chosen": -283.1133728027344,
+      "logps/rejected": -282.29644775390625,
+      "loss": 0.693,
+      "rewards/accuracies": 0.5249999761581421,
+      "rewards/chosen": -0.00030150412931106985,
+      "rewards/margins": -8.710605470696464e-05,
+      "rewards/rejected": -0.00021439809643197805,
+      "step": 20
+    },
+    {
+      "epoch": 0.03139717425431711,
+      "grad_norm": 12.15555855567297,
+      "learning_rate": 1.5624999999999999e-07,
+      "logits/chosen": -2.691681385040283,
+      "logits/rejected": -2.6676158905029297,
+      "logps/chosen": -270.2156982421875,
+      "logps/rejected": -276.73724365234375,
+      "loss": 0.6923,
+      "rewards/accuracies": 0.543749988079071,
+      "rewards/chosen": 0.0006642484804615378,
+      "rewards/margins": 0.0012884512543678284,
+      "rewards/rejected": -0.0006242028321139514,
+      "step": 30
+    },
+    {
+      "epoch": 0.04186289900575615,
+      "grad_norm": 11.841392011083105,
+      "learning_rate": 2.0833333333333333e-07,
+      "logits/chosen": -2.6653332710266113,
+      "logits/rejected": -2.590430736541748,
+      "logps/chosen": -290.4707946777344,
+      "logps/rejected": -282.1141662597656,
+      "loss": 0.6902,
+      "rewards/accuracies": 0.637499988079071,
+      "rewards/chosen": 0.00528017058968544,
+      "rewards/margins": 0.0057797241024672985,
+      "rewards/rejected": -0.0004995539784431458,
+      "step": 40
+    },
+    {
+      "epoch": 0.052328623757195186,
+      "grad_norm": 13.24083046891695,
+      "learning_rate": 2.604166666666667e-07,
+      "logits/chosen": -2.6722922325134277,
+      "logits/rejected": -2.588569164276123,
+      "logps/chosen": -266.1282958984375,
+      "logps/rejected": -236.51809692382812,
+      "loss": 0.6859,
+      "rewards/accuracies": 0.675000011920929,
+      "rewards/chosen": 0.014677075669169426,
+      "rewards/margins": 0.013995910063385963,
+      "rewards/rejected": 0.0006811673520132899,
+      "step": 50
+    },
+    {
+      "epoch": 0.06279434850863422,
+      "grad_norm": 11.856942211355204,
+      "learning_rate": 3.1249999999999997e-07,
+      "logits/chosen": -2.627357244491577,
+      "logits/rejected": -2.591691493988037,
+      "logps/chosen": -299.61175537109375,
+      "logps/rejected": -274.43731689453125,
+      "loss": 0.6791,
+      "rewards/accuracies": 0.706250011920929,
+      "rewards/chosen": 0.044125042855739594,
+      "rewards/margins": 0.030969683080911636,
+      "rewards/rejected": 0.013155360706150532,
+      "step": 60
+    },
+    {
+      "epoch": 0.07326007326007326,
+      "grad_norm": 11.929671055235316,
+      "learning_rate": 3.645833333333333e-07,
+      "logits/chosen": -2.5305747985839844,
+      "logits/rejected": -2.523099422454834,
+      "logps/chosen": -257.43914794921875,
+      "logps/rejected": -262.67779541015625,
+      "loss": 0.6702,
+      "rewards/accuracies": 0.6812499761581421,
+      "rewards/chosen": 0.021997224539518356,
+      "rewards/margins": 0.06655967980623245,
+      "rewards/rejected": -0.044562458992004395,
+      "step": 70
+    },
+    {
+      "epoch": 0.0837257980115123,
+      "grad_norm": 13.864057015926953,
+      "learning_rate": 4.1666666666666667e-07,
+      "logits/chosen": -2.5468287467956543,
+      "logits/rejected": -2.474470376968384,
+      "logps/chosen": -274.47784423828125,
+      "logps/rejected": -261.062255859375,
+      "loss": 0.6503,
+      "rewards/accuracies": 0.706250011920929,
+      "rewards/chosen": 0.01564904674887657,
+      "rewards/margins": 0.11314449459314346,
+      "rewards/rejected": -0.097495436668396,
+      "step": 80
+    },
+    {
+      "epoch": 0.09419152276295134,
+      "grad_norm": 15.875862157352557,
+      "learning_rate": 4.6874999999999996e-07,
+      "logits/chosen": -2.488557815551758,
+      "logits/rejected": -2.4535651206970215,
+      "logps/chosen": -261.57366943359375,
+      "logps/rejected": -278.4251403808594,
+      "loss": 0.642,
+      "rewards/accuracies": 0.6499999761581421,
+      "rewards/chosen": -0.059757404029369354,
+      "rewards/margins": 0.07676272094249725,
+      "rewards/rejected": -0.1365201324224472,
+      "step": 90
+    },
+    {
+      "epoch": 0.10465724751439037,
+      "grad_norm": 13.065702068285603,
+      "learning_rate": 4.999732492681437e-07,
+      "logits/chosen": -2.4728801250457764,
+      "logits/rejected": -2.390676498413086,
+      "logps/chosen": -304.7428894042969,
+      "logps/rejected": -304.18231201171875,
+      "loss": 0.6267,
+      "rewards/accuracies": 0.7437499761581421,
+      "rewards/chosen": -0.0491873137652874,
+      "rewards/margins": 0.19260282814502716,
+      "rewards/rejected": -0.24179014563560486,
+      "step": 100
+    },
+    {
+      "epoch": 0.10465724751439037,
+      "eval_logits/chosen": -2.4988505840301514,
+      "eval_logits/rejected": -2.424436569213867,
+      "eval_logps/chosen": -280.5349426269531,
+      "eval_logps/rejected": -274.64080810546875,
+      "eval_loss": 0.6375061273574829,
+      "eval_rewards/accuracies": 0.7063491940498352,
+      "eval_rewards/chosen": 0.014330551959574223,
+      "eval_rewards/margins": 0.1586705446243286,
+      "eval_rewards/rejected": -0.14433999359607697,
+      "eval_runtime": 175.7842,
+      "eval_samples_per_second": 11.378,
+      "eval_steps_per_second": 0.358,
+      "step": 100
+    },
+    {
+      "epoch": 0.1151229722658294,
+      "grad_norm": 18.15644072773089,
+      "learning_rate": 4.996723692767926e-07,
+      "logits/chosen": -2.5023550987243652,
+      "logits/rejected": -2.432821273803711,
+      "logps/chosen": -249.1396484375,
+      "logps/rejected": -264.470458984375,
+      "loss": 0.6297,
+      "rewards/accuracies": 0.731249988079071,
+      "rewards/chosen": -0.07685011625289917,
+      "rewards/margins": 0.19513027369976044,
+      "rewards/rejected": -0.2719804048538208,
+      "step": 110
+    },
+    {
+      "epoch": 0.12558869701726844,
+      "grad_norm": 22.023309150811986,
+      "learning_rate": 4.990375746213598e-07,
+      "logits/chosen": -2.3616247177124023,
+      "logits/rejected": -2.281648874282837,
+      "logps/chosen": -285.5514831542969,
+      "logps/rejected": -339.8910217285156,
+      "loss": 0.5961,
+      "rewards/accuracies": 0.737500011920929,
+      "rewards/chosen": -0.2297312319278717,
+      "rewards/margins": 0.2847149968147278,
+      "rewards/rejected": -0.5144462585449219,
+      "step": 120
+    },
+    {
+      "epoch": 0.1360544217687075,
+      "grad_norm": 19.678835809215887,
+      "learning_rate": 4.980697142834314e-07,
+      "logits/chosen": -2.283674478530884,
+      "logits/rejected": -2.192523956298828,
+      "logps/chosen": -294.853515625,
+      "logps/rejected": -303.66986083984375,
+      "loss": 0.5883,
+      "rewards/accuracies": 0.7437499761581421,
+      "rewards/chosen": -0.1358092576265335,
+      "rewards/margins": 0.3562476634979248,
+      "rewards/rejected": -0.4920569360256195,
+      "step": 130
+    },
+    {
+      "epoch": 0.14652014652014653,
+      "grad_norm": 18.045058125702276,
+      "learning_rate": 4.967700826904229e-07,
+      "logits/chosen": -2.1272222995758057,
+      "logits/rejected": -2.050865888595581,
+      "logps/chosen": -321.29254150390625,
+      "logps/rejected": -360.5655517578125,
+      "loss": 0.5733,
+      "rewards/accuracies": 0.7749999761581421,
+      "rewards/chosen": -0.3020511567592621,
+      "rewards/margins": 0.41385501623153687,
+      "rewards/rejected": -0.7159062027931213,
+      "step": 140
+    },
+    {
+      "epoch": 0.15698587127158556,
+      "grad_norm": 23.733353319304864,
+      "learning_rate": 4.951404179843962e-07,
+      "logits/chosen": -1.8601700067520142,
+      "logits/rejected": -1.6911884546279907,
+      "logps/chosen": -328.0176696777344,
+      "logps/rejected": -331.93365478515625,
+      "loss": 0.5724,
+      "rewards/accuracies": 0.7124999761581421,
+      "rewards/chosen": -0.40121254324913025,
+      "rewards/margins": 0.44814401865005493,
+      "rewards/rejected": -0.849356472492218,
+      "step": 150
+    },
+    {
+      "epoch": 0.1674515960230246,
+      "grad_norm": 28.013542721553446,
+      "learning_rate": 4.931828996974498e-07,
+      "logits/chosen": -0.7096320390701294,
+      "logits/rejected": -0.25428909063339233,
+      "logps/chosen": -341.81439208984375,
+      "logps/rejected": -363.79217529296875,
+      "loss": 0.5414,
+      "rewards/accuracies": 0.737500011920929,
+      "rewards/chosen": -0.46001678705215454,
+      "rewards/margins": 0.6181891560554504,
+      "rewards/rejected": -1.0782058238983154,
+      "step": 160
+    },
+    {
+      "epoch": 0.17791732077446362,
+      "grad_norm": 23.342052779259543,
+      "learning_rate": 4.909001458367866e-07,
+      "logits/chosen": -0.7341230511665344,
+      "logits/rejected": -0.3530420660972595,
+      "logps/chosen": -352.09222412109375,
+      "logps/rejected": -376.5685119628906,
+      "loss": 0.5581,
+      "rewards/accuracies": 0.7124999761581421,
+      "rewards/chosen": -0.7362300753593445,
+      "rewards/margins": 0.6209839582443237,
+      "rewards/rejected": -1.3572142124176025,
+      "step": 170
+    },
+    {
+      "epoch": 0.18838304552590268,
+      "grad_norm": 28.21709060740068,
+      "learning_rate": 4.882952093833627e-07,
+      "logits/chosen": -0.876774787902832,
+      "logits/rejected": -0.6448796391487122,
+      "logps/chosen": -335.0474548339844,
+      "logps/rejected": -404.52728271484375,
+      "loss": 0.5274,
+      "rewards/accuracies": 0.75,
+      "rewards/chosen": -0.9509947896003723,
+      "rewards/margins": 0.6264718174934387,
+      "rewards/rejected": -1.5774667263031006,
+      "step": 180
+    },
+    {
+      "epoch": 0.1988487702773417,
+      "grad_norm": 40.43719634141344,
+      "learning_rate": 4.853715742087946e-07,
+      "logits/chosen": -0.12615634500980377,
+      "logits/rejected": 0.2832840085029602,
+      "logps/chosen": -344.6008605957031,
+      "logps/rejected": -427.69354248046875,
+      "loss": 0.5309,
+      "rewards/accuracies": 0.762499988079071,
+      "rewards/chosen": -1.0862140655517578,
+      "rewards/margins": 0.6919594407081604,
+      "rewards/rejected": -1.7781736850738525,
+      "step": 190
+    },
+    {
+      "epoch": 0.20931449502878074,
+      "grad_norm": 24.48228039656553,
+      "learning_rate": 4.821331504159906e-07,
+      "logits/chosen": -1.134807825088501,
+      "logits/rejected": -0.7409021258354187,
+      "logps/chosen": -399.9931640625,
+      "logps/rejected": -406.24249267578125,
+      "loss": 0.578,
+      "rewards/accuracies": 0.7124999761581421,
+      "rewards/chosen": -0.893362820148468,
+      "rewards/margins": 0.5979009866714478,
+      "rewards/rejected": -1.49126398563385,
+      "step": 200
+    },
+    {
+      "epoch": 0.20931449502878074,
+      "eval_logits/chosen": -1.7000207901000977,
+      "eval_logits/rejected": -1.3956148624420166,
+      "eval_logps/chosen": -357.6234130859375,
+      "eval_logps/rejected": -397.53143310546875,
+      "eval_loss": 0.5543879270553589,
+      "eval_rewards/accuracies": 0.7480158805847168,
+      "eval_rewards/chosen": -0.7565548419952393,
+      "eval_rewards/margins": 0.616691529750824,
+      "eval_rewards/rejected": -1.3732463121414185,
+      "eval_runtime": 175.6227,
+      "eval_samples_per_second": 11.388,
+      "eval_steps_per_second": 0.359,
+      "step": 200
+    },
+    {
+      "epoch": 0.21978021978021978,
+      "grad_norm": 33.0816384461788,
+      "learning_rate": 4.785842691097342e-07,
+      "logits/chosen": -1.2056543827056885,
+      "logits/rejected": -0.7488024830818176,
+      "logps/chosen": -379.0580139160156,
+      "logps/rejected": -430.2911682128906,
+      "loss": 0.5639,
+      "rewards/accuracies": 0.668749988079071,
+      "rewards/chosen": -0.9624747037887573,
+      "rewards/margins": 0.6178582906723022,
+      "rewards/rejected": -1.5803329944610596,
+      "step": 210
+    },
+    {
+      "epoch": 0.2302459445316588,
+      "grad_norm": 33.16914134802434,
+      "learning_rate": 4.7472967660421603e-07,
+      "logits/chosen": -0.9634332656860352,
+      "logits/rejected": -0.056523989886045456,
+      "logps/chosen": -388.16888427734375,
+      "logps/rejected": -407.8970642089844,
+      "loss": 0.5222,
+      "rewards/accuracies": 0.762499988079071,
+      "rewards/chosen": -1.0096079111099243,
+      "rewards/margins": 0.7075966000556946,
+      "rewards/rejected": -1.7172044515609741,
+      "step": 220
+    },
+    {
+      "epoch": 0.24071166928309787,
+      "grad_norm": 35.571946638008065,
+      "learning_rate": 4.705745280752585e-07,
+      "logits/chosen": -0.3167392611503601,
+      "logits/rejected": 0.18852970004081726,
+      "logps/chosen": -397.49847412109375,
+      "logps/rejected": -440.3253479003906,
+      "loss": 0.5468,
+      "rewards/accuracies": 0.7437499761581421,
+      "rewards/chosen": -1.0877156257629395,
+      "rewards/margins": 0.6301809549331665,
+      "rewards/rejected": -1.7178964614868164,
+      "step": 230
+    },
+    {
+      "epoch": 0.25117739403453687,
+      "grad_norm": 32.242793295380665,
+      "learning_rate": 4.6612438066572555e-07,
+      "logits/chosen": -0.004159653093665838,
+      "logits/rejected": 0.8073694109916687,
+      "logps/chosen": -364.86566162109375,
+      "logps/rejected": -433.24072265625,
+      "loss": 0.5162,
+      "rewards/accuracies": 0.7437499761581421,
+      "rewards/chosen": -0.9729146957397461,
+      "rewards/margins": 0.8177644610404968,
+      "rewards/rejected": -1.7906792163848877,
+      "step": 240
+    },
+    {
+      "epoch": 0.2616431187859759,
+      "grad_norm": 29.7046569499327,
+      "learning_rate": 4.6138518605333664e-07,
+      "logits/chosen": -0.3281826674938202,
+      "logits/rejected": 0.29218000173568726,
+      "logps/chosen": -366.93865966796875,
+      "logps/rejected": -417.32666015625,
+      "loss": 0.5058,
+      "rewards/accuracies": 0.71875,
+      "rewards/chosen": -0.8915438652038574,
+      "rewards/margins": 0.7059476375579834,
+      "rewards/rejected": -1.5974915027618408,
+      "step": 250
+    },
+    {
+      "epoch": 0.272108843537415,
+      "grad_norm": 26.07189300308382,
+      "learning_rate": 4.5636328249082514e-07,
+      "logits/chosen": -0.7655261158943176,
+      "logits/rejected": 0.05788875371217728,
+      "logps/chosen": -377.3932800292969,
+      "logps/rejected": -431.24603271484375,
+      "loss": 0.5333,
+      "rewards/accuracies": 0.675000011920929,
+      "rewards/chosen": -0.8842679262161255,
+      "rewards/margins": 0.6056793928146362,
+      "rewards/rejected": -1.4899474382400513,
+      "step": 260
+    },
+    {
+      "epoch": 0.282574568288854,
+      "grad_norm": 27.48585519105196,
+      "learning_rate": 4.510653863290871e-07,
+      "logits/chosen": -0.8150612711906433,
+      "logits/rejected": -0.19155649840831757,
+      "logps/chosen": -376.1461486816406,
+      "logps/rejected": -399.7369689941406,
+      "loss": 0.5074,
+      "rewards/accuracies": 0.737500011920929,
+      "rewards/chosen": -0.7649620175361633,
+      "rewards/margins": 0.6968373656272888,
+      "rewards/rejected": -1.4617992639541626,
+      "step": 270
+    },
+    {
+      "epoch": 0.29304029304029305,
+      "grad_norm": 28.38677400170819,
+      "learning_rate": 4.4549858303465737e-07,
+      "logits/chosen": 0.3504285514354706,
+      "logits/rejected": 0.9512192606925964,
+      "logps/chosen": -415.9444274902344,
+      "logps/rejected": -479.0673828125,
+      "loss": 0.5416,
+      "rewards/accuracies": 0.7124999761581421,
+      "rewards/chosen": -1.1970031261444092,
+      "rewards/margins": 0.7609006762504578,
+      "rewards/rejected": -1.9579038619995117,
+      "step": 280
+    },
+    {
+      "epoch": 0.3035060177917321,
+      "grad_norm": 32.744553637286586,
+      "learning_rate": 4.396703177135261e-07,
+      "logits/chosen": -0.09853874146938324,
+      "logits/rejected": 0.8018854260444641,
+      "logps/chosen": -399.70263671875,
+      "logps/rejected": -441.5960388183594,
+      "loss": 0.5186,
+      "rewards/accuracies": 0.731249988079071,
+      "rewards/chosen": -1.269182801246643,
+      "rewards/margins": 0.7766303420066833,
+      "rewards/rejected": -2.0458133220672607,
+      "step": 290
+    },
+    {
+      "epoch": 0.3139717425431711,
+      "grad_norm": 36.30101198384572,
+      "learning_rate": 4.335883851539693e-07,
+      "logits/chosen": -1.0064457654953003,
+      "logits/rejected": -0.4348925054073334,
+      "logps/chosen": -373.8060302734375,
+      "logps/rejected": -438.51513671875,
+      "loss": 0.5301,
+      "rewards/accuracies": 0.7250000238418579,
+      "rewards/chosen": -1.0478389263153076,
+      "rewards/margins": 0.6600475907325745,
+      "rewards/rejected": -1.7078864574432373,
+      "step": 300
+    },
+    {
+      "epoch": 0.3139717425431711,
+      "eval_logits/chosen": -0.7774365544319153,
+      "eval_logits/rejected": -0.07988195866346359,
+      "eval_logps/chosen": -377.41851806640625,
+      "eval_logps/rejected": -439.006103515625,
+      "eval_loss": 0.519775927066803,
+      "eval_rewards/accuracies": 0.7757936716079712,
+      "eval_rewards/chosen": -0.9545055627822876,
+      "eval_rewards/margins": 0.8334872126579285,
+      "eval_rewards/rejected": -1.7879927158355713,
+      "eval_runtime": 175.6202,
+      "eval_samples_per_second": 11.388,
+      "eval_steps_per_second": 0.359,
+      "step": 300
+    },
+    {
+      "epoch": 0.32443746729461015,
+      "grad_norm": 34.23825403742458,
+      "learning_rate": 4.272609194017105e-07,
+      "logits/chosen": -0.798404335975647,
+      "logits/rejected": 0.40799885988235474,
+      "logps/chosen": -376.5071716308594,
+      "logps/rejected": -409.09600830078125,
+      "loss": 0.5113,
+      "rewards/accuracies": 0.75,
+      "rewards/chosen": -0.9274541735649109,
+      "rewards/margins": 0.8360295295715332,
+      "rewards/rejected": -1.7634836435317993,
+      "step": 310
+    },
+    {
+      "epoch": 0.3349031920460492,
+      "grad_norm": 39.68212351207402,
+      "learning_rate": 4.2069638288135547e-07,
+      "logits/chosen": 0.02141362428665161,
+      "logits/rejected": 0.8004009127616882,
+      "logps/chosen": -371.86358642578125,
+      "logps/rejected": -460.26971435546875,
+      "loss": 0.5256,
+      "rewards/accuracies": 0.7250000238418579,
+      "rewards/chosen": -1.1884684562683105,
+      "rewards/margins": 0.9082374572753906,
+      "rewards/rejected": -2.096705913543701,
+      "step": 320
+    },
+    {
+      "epoch": 0.3453689167974882,
+      "grad_norm": 37.95982237233792,
+      "learning_rate": 4.139035550786494e-07,
+      "logits/chosen": 0.24720951914787292,
+      "logits/rejected": 0.8929317593574524,
+      "logps/chosen": -380.25689697265625,
+      "logps/rejected": -435.5292053222656,
+      "loss": 0.5049,
+      "rewards/accuracies": 0.731249988079071,
+      "rewards/chosen": -1.2391283512115479,
+      "rewards/margins": 0.795093297958374,
+      "rewards/rejected": -2.034221649169922,
+      "step": 330
+    },
+    {
+      "epoch": 0.35583464154892724,
+      "grad_norm": 31.44829466131635,
+      "learning_rate": 4.0689152079869306e-07,
+      "logits/chosen": 0.19869890809059143,
+      "logits/rejected": 0.7656279802322388,
+      "logps/chosen": -384.75897216796875,
+      "logps/rejected": -426.62274169921875,
+      "loss": 0.5341,
+      "rewards/accuracies": 0.71875,
+      "rewards/chosen": -1.1527179479599,
+      "rewards/margins": 0.7536069750785828,
+      "rewards/rejected": -1.9063247442245483,
+      "step": 340
+    },
+    {
+      "epoch": 0.3663003663003663,
+      "grad_norm": 33.60100508993655,
+      "learning_rate": 3.99669658015821e-07,
+      "logits/chosen": -0.23042722046375275,
+      "logits/rejected": 0.44881439208984375,
+      "logps/chosen": -416.906982421875,
+      "logps/rejected": -460.3355407714844,
+      "loss": 0.502,
+      "rewards/accuracies": 0.71875,
+      "rewards/chosen": -1.2567191123962402,
+      "rewards/margins": 0.7857658863067627,
+      "rewards/rejected": -2.042484998703003,
+      "step": 350
+    },
+    {
+      "epoch": 0.37676609105180536,
+      "grad_norm": 35.86790303415799,
+      "learning_rate": 3.92247625331392e-07,
+      "logits/chosen": -0.2522183060646057,
+      "logits/rejected": 0.6683695316314697,
+      "logps/chosen": -416.01641845703125,
+      "logps/rejected": -479.52294921875,
+      "loss": 0.4805,
+      "rewards/accuracies": 0.8062499761581421,
+      "rewards/chosen": -1.3013213872909546,
+      "rewards/margins": 0.8662745356559753,
+      "rewards/rejected": -2.1675961017608643,
+      "step": 360
+    },
+    {
+      "epoch": 0.3872318158032444,
+      "grad_norm": 34.95835364160841,
+      "learning_rate": 3.846353490562664e-07,
+      "logits/chosen": 0.0829664096236229,
+      "logits/rejected": 0.9096466898918152,
+      "logps/chosen": -406.54913330078125,
+      "logps/rejected": -456.1133728027344,
+      "loss": 0.5136,
+      "rewards/accuracies": 0.6937500238418579,
+      "rewards/chosen": -1.4327296018600464,
+      "rewards/margins": 0.7719124555587769,
+      "rewards/rejected": -2.2046420574188232,
+      "step": 370
+    },
+    {
+      "epoch": 0.3976975405546834,
+      "grad_norm": 33.96402149473364,
+      "learning_rate": 3.768430099352445e-07,
+      "logits/chosen": -0.14455661177635193,
+      "logits/rejected": 0.8097721934318542,
+      "logps/chosen": -388.0363464355469,
+      "logps/rejected": -460.75048828125,
+      "loss": 0.5138,
+      "rewards/accuracies": 0.7437499761581421,
+      "rewards/chosen": -1.333893060684204,
+      "rewards/margins": 0.8102533221244812,
+      "rewards/rejected": -2.14414644241333,
+      "step": 380
+    },
+    {
+      "epoch": 0.40816326530612246,
+      "grad_norm": 32.427509955053814,
+      "learning_rate": 3.6888102953122304e-07,
+      "logits/chosen": 0.7358514070510864,
+      "logits/rejected": 1.5359573364257812,
+      "logps/chosen": -425.21014404296875,
+      "logps/rejected": -492.5250549316406,
+      "loss": 0.5231,
+      "rewards/accuracies": 0.7124999761581421,
+      "rewards/chosen": -1.4359853267669678,
+      "rewards/margins": 0.8055831789970398,
+      "rewards/rejected": -2.2415683269500732,
+      "step": 390
+    },
+    {
+      "epoch": 0.4186289900575615,
+      "grad_norm": 36.13318678339082,
+      "learning_rate": 3.607600562872785e-07,
+      "logits/chosen": 0.8064459562301636,
+      "logits/rejected": 1.430558681488037,
+      "logps/chosen": -414.16839599609375,
+      "logps/rejected": -482.3439025878906,
+      "loss": 0.5386,
+      "rewards/accuracies": 0.7250000238418579,
+      "rewards/chosen": -1.4417721033096313,
+      "rewards/margins": 0.7046888470649719,
+      "rewards/rejected": -2.146461009979248,
+      "step": 400
+    },
+    {
+      "epoch": 0.4186289900575615,
+      "eval_logits/chosen": 0.024975202977657318,
+      "eval_logits/rejected": 0.9406751394271851,
+      "eval_logps/chosen": -412.3045654296875,
+      "eval_logps/rejected": -471.00274658203125,
+      "eval_loss": 0.510511040687561,
+      "eval_rewards/accuracies": 0.77182537317276,
+      "eval_rewards/chosen": -1.3033660650253296,
+      "eval_rewards/margins": 0.8045932054519653,
+      "eval_rewards/rejected": -2.107959032058716,
+      "eval_runtime": 177.0547,
+      "eval_samples_per_second": 11.296,
+      "eval_steps_per_second": 0.356,
+      "step": 400
+    },
+    {
+      "epoch": 0.4290947148090005,
+      "grad_norm": 26.612881775481124,
+      "learning_rate": 3.5249095128531856e-07,
+      "logits/chosen": 0.2300519496202469,
+      "logits/rejected": 0.9924400448799133,
+      "logps/chosen": -398.78729248046875,
+      "logps/rejected": -436.8857421875,
+      "loss": 0.5301,
+      "rewards/accuracies": 0.7250000238418579,
+      "rewards/chosen": -1.270517110824585,
+      "rewards/margins": 0.6298418045043945,
+      "rewards/rejected": -1.9003589153289795,
+      "step": 410
+    },
+    {
+      "epoch": 0.43956043956043955,
+      "grad_norm": 39.714482709669014,
+      "learning_rate": 3.4408477372034736e-07,
+      "logits/chosen": -0.2808998227119446,
+      "logits/rejected": 0.9214981198310852,
+      "logps/chosen": -379.78924560546875,
+      "logps/rejected": -428.7146911621094,
+      "loss": 0.5265,
+      "rewards/accuracies": 0.793749988079071,
+      "rewards/chosen": -0.9761778116226196,
+      "rewards/margins": 0.7907910346984863,
+      "rewards/rejected": -1.7669687271118164,
+      "step": 420
+    },
+    {
+      "epoch": 0.4500261643118786,
+      "grad_norm": 36.7812247576203,
+      "learning_rate": 3.3555276610977276e-07,
+      "logits/chosen": 1.0532963275909424,
+      "logits/rejected": 1.682213544845581,
+      "logps/chosen": -372.61004638671875,
+      "logps/rejected": -455.31524658203125,
+      "loss": 0.507,
+      "rewards/accuracies": 0.7749999761581421,
+      "rewards/chosen": -1.18856680393219,
+      "rewards/margins": 0.8627974390983582,
+      "rewards/rejected": -2.0513641834259033,
+      "step": 430
+    },
+    {
+      "epoch": 0.4604918890633176,
+      "grad_norm": 37.84672118674808,
+      "learning_rate": 3.269063392575352e-07,
+      "logits/chosen": 2.015638828277588,
+      "logits/rejected": 2.8407273292541504,
+      "logps/chosen": -472.79931640625,
+      "logps/rejected": -504.94561767578125,
+      "loss": 0.5227,
+      "rewards/accuracies": 0.6937500238418579,
+      "rewards/chosen": -1.7762467861175537,
+      "rewards/margins": 0.7188802361488342,
+      "rewards/rejected": -2.4951272010803223,
+      "step": 440
+    },
+    {
+      "epoch": 0.47095761381475665,
+      "grad_norm": 30.609934320862045,
+      "learning_rate": 3.1815705699316964e-07,
+      "logits/chosen": 1.3560642004013062,
+      "logits/rejected": 2.127135753631592,
+      "logps/chosen": -465.49188232421875,
+      "logps/rejected": -571.1328125,
+      "loss": 0.5099,
+      "rewards/accuracies": 0.800000011920929,
+      "rewards/chosen": -1.8310394287109375,
+      "rewards/margins": 1.0971113443374634,
+      "rewards/rejected": -2.9281508922576904,
+      "step": 450
+    },
+    {
+      "epoch": 0.48142333856619574,
+      "grad_norm": 28.715258464621037,
+      "learning_rate": 3.0931662070620794e-07,
+      "logits/chosen": 0.8564838171005249,
+      "logits/rejected": 1.5636537075042725,
+      "logps/chosen": -452.06951904296875,
+      "logps/rejected": -517.620361328125,
+      "loss": 0.5258,
+      "rewards/accuracies": 0.7250000238418579,
+      "rewards/chosen": -1.9461126327514648,
+      "rewards/margins": 0.7798872590065002,
+      "rewards/rejected": -2.725999593734741,
+      "step": 460
+    },
+    {
+      "epoch": 0.49188906331763477,
+      "grad_norm": 43.25596740234111,
+      "learning_rate": 3.003968536966078e-07,
+      "logits/chosen": 0.4564128816127777,
+      "logits/rejected": 1.1015666723251343,
+      "logps/chosen": -456.0911560058594,
+      "logps/rejected": -495.827880859375,
+      "loss": 0.4802,
+      "rewards/accuracies": 0.7124999761581421,
+      "rewards/chosen": -1.7230240106582642,
+      "rewards/margins": 0.7358741164207458,
+      "rewards/rejected": -2.458897829055786,
+      "step": 470
+    },
+    {
+      "epoch": 0.5023547880690737,
+      "grad_norm": 36.20202372184976,
+      "learning_rate": 2.9140968536213693e-07,
+      "logits/chosen": 0.7777734994888306,
+      "logits/rejected": 1.9267253875732422,
+      "logps/chosen": -460.71783447265625,
+      "logps/rejected": -554.2379150390625,
+      "loss": 0.5012,
+      "rewards/accuracies": 0.8125,
+      "rewards/chosen": -1.935050368309021,
+      "rewards/margins": 1.0943626165390015,
+      "rewards/rejected": -3.0294129848480225,
+      "step": 480
+    },
+    {
+      "epoch": 0.5128205128205128,
+      "grad_norm": 32.834417417864124,
+      "learning_rate": 2.823671352438608e-07,
+      "logits/chosen": 0.05315951257944107,
+      "logits/rejected": 2.070735454559326,
+      "logps/chosen": -481.5361328125,
+      "logps/rejected": -537.8575439453125,
+      "loss": 0.4958,
+      "rewards/accuracies": 0.824999988079071,
+      "rewards/chosen": -1.6960468292236328,
+      "rewards/margins": 1.1183526515960693,
+      "rewards/rejected": -2.814399242401123,
+      "step": 490
+    },
+    {
+      "epoch": 0.5232862375719518,
+      "grad_norm": 37.03566337015388,
+      "learning_rate": 2.73281296951072e-07,
+      "logits/chosen": 0.2395954132080078,
+      "logits/rejected": 1.218072772026062,
+      "logps/chosen": -431.42626953125,
+      "logps/rejected": -487.06756591796875,
+      "loss": 0.4996,
+      "rewards/accuracies": 0.7749999761581421,
+      "rewards/chosen": -1.7278209924697876,
+      "rewards/margins": 0.868577778339386,
+      "rewards/rejected": -2.5963988304138184,
+      "step": 500
+    },
+    {
+      "epoch": 0.5232862375719518,
+      "eval_logits/chosen": 0.005896765738725662,
+      "eval_logits/rejected": 1.1533339023590088,
+      "eval_logps/chosen": -460.39996337890625,
+      "eval_logps/rejected": -534.9353637695312,
+      "eval_loss": 0.49528467655181885,
+      "eval_rewards/accuracies": 0.7678571343421936,
+      "eval_rewards/chosen": -1.7843199968338013,
+      "eval_rewards/margins": 0.9629656076431274,
+      "eval_rewards/rejected": -2.7472856044769287,
+      "eval_runtime": 175.591,
+      "eval_samples_per_second": 11.39,
+      "eval_steps_per_second": 0.359,
+      "step": 500
+    },
+    {
+      "epoch": 0.533751962323391,
+      "grad_norm": 31.83813721910833,
+      "learning_rate": 2.641643219871597e-07,
+      "logits/chosen": -0.09106893837451935,
+      "logits/rejected": 1.0047601461410522,
+      "logps/chosen": -448.24365234375,
+      "logps/rejected": -547.2485961914062,
+      "loss": 0.4759,
+      "rewards/accuracies": 0.824999988079071,
+      "rewards/chosen": -1.6430542469024658,
+      "rewards/margins": 1.1913459300994873,
+      "rewards/rejected": -2.834400177001953,
+      "step": 510
+    },
+    {
+      "epoch": 0.54421768707483,
+      "grad_norm": 31.094283760600373,
+      "learning_rate": 2.550284034980507e-07,
+      "logits/chosen": 0.5583597421646118,
+      "logits/rejected": 1.752502202987671,
+      "logps/chosen": -487.286376953125,
+      "logps/rejected": -552.1371459960938,
+      "loss": 0.5187,
+      "rewards/accuracies": 0.706250011920929,
+      "rewards/chosen": -2.1521472930908203,
+      "rewards/margins": 0.8697667121887207,
+      "rewards/rejected": -3.021914005279541,
+      "step": 520
+    },
+    {
+      "epoch": 0.554683411826269,
+      "grad_norm": 46.54511021652677,
+      "learning_rate": 2.4588575996495794e-07,
+      "logits/chosen": -0.278386652469635,
+      "logits/rejected": 0.8032172918319702,
+      "logps/chosen": -446.83624267578125,
+      "logps/rejected": -526.9205322265625,
+      "loss": 0.4996,
+      "rewards/accuracies": 0.75,
+      "rewards/chosen": -1.7808927297592163,
+      "rewards/margins": 0.9380749464035034,
+      "rewards/rejected": -2.7189676761627197,
+      "step": 530
+    },
+    {
+      "epoch": 0.565149136577708,
+      "grad_norm": 32.98937848769124,
+      "learning_rate": 2.367486188632446e-07,
+      "logits/chosen": -0.7310466170310974,
+      "logits/rejected": 0.3242616355419159,
+      "logps/chosen": -452.5191345214844,
+      "logps/rejected": -511.3818359375,
+      "loss": 0.513,
+      "rewards/accuracies": 0.731249988079071,
+      "rewards/chosen": -1.6216903924942017,
+      "rewards/margins": 0.8862103223800659,
+      "rewards/rejected": -2.5079007148742676,
+      "step": 540
+    },
+    {
+      "epoch": 0.5756148613291471,
+      "grad_norm": 28.08745252820734,
+      "learning_rate": 2.276292003092593e-07,
+      "logits/chosen": -0.819686770439148,
+      "logits/rejected": 0.32188865542411804,
+      "logps/chosen": -443.1897888183594,
+      "logps/rejected": -494.9161071777344,
+      "loss": 0.526,
+      "rewards/accuracies": 0.7437499761581421,
+      "rewards/chosen": -1.719175100326538,
+      "rewards/margins": 0.9371669888496399,
+      "rewards/rejected": -2.656342029571533,
+      "step": 550
+    },
+    {
+      "epoch": 0.5860805860805861,
+      "grad_norm": 34.81864414663528,
+      "learning_rate": 2.185397007170141e-07,
+      "logits/chosen": -0.32582345604896545,
+      "logits/rejected": 1.0776915550231934,
+      "logps/chosen": -462.23406982421875,
+      "logps/rejected": -525.7540893554688,
+      "loss": 0.5104,
+      "rewards/accuracies": 0.7749999761581421,
+      "rewards/chosen": -1.7941694259643555,
+      "rewards/margins": 0.9143456220626831,
+      "rewards/rejected": -2.708515167236328,
+      "step": 560
+    },
+    {
+      "epoch": 0.5965463108320251,
+      "grad_norm": 32.06431987316443,
+      "learning_rate": 2.094922764865619e-07,
+      "logits/chosen": -0.5136072635650635,
+      "logits/rejected": 0.8532153367996216,
+      "logps/chosen": -462.86260986328125,
+      "logps/rejected": -532.7152099609375,
+      "loss": 0.4848,
+      "rewards/accuracies": 0.7562500238418579,
+      "rewards/chosen": -1.8336808681488037,
+      "rewards/margins": 0.8313940167427063,
+      "rewards/rejected": -2.6650748252868652,
+      "step": 570
+    },
+    {
+      "epoch": 0.6070120355834642,
+      "grad_norm": 38.10097117955034,
+      "learning_rate": 2.0049902774588797e-07,
+      "logits/chosen": -0.3451794385910034,
+      "logits/rejected": 0.9854526519775391,
+      "logps/chosen": -424.631103515625,
+      "logps/rejected": -478.0337829589844,
+      "loss": 0.5039,
+      "rewards/accuracies": 0.7749999761581421,
+      "rewards/chosen": -1.7129182815551758,
+      "rewards/margins": 0.9391329884529114,
+      "rewards/rejected": -2.6520514488220215,
+      "step": 580
+    },
+    {
+      "epoch": 0.6174777603349032,
+      "grad_norm": 31.542114777873778,
+      "learning_rate": 1.9157198216806238e-07,
+      "logits/chosen": -0.5171893239021301,
+      "logits/rejected": 0.9677211046218872,
+      "logps/chosen": -454.8963928222656,
+      "logps/rejected": -536.4180908203125,
+      "loss": 0.4838,
+      "rewards/accuracies": 0.6937500238418579,
+      "rewards/chosen": -1.7694385051727295,
+      "rewards/margins": 0.9072266817092896,
+      "rewards/rejected": -2.6766653060913086,
+      "step": 590
+    },
+    {
+      "epoch": 0.6279434850863422,
+      "grad_norm": 29.346076446463687,
+      "learning_rate": 1.8272307888529274e-07,
+      "logits/chosen": -0.03304073214530945,
+      "logits/rejected": 1.3204997777938843,
+      "logps/chosen": -449.006103515625,
+      "logps/rejected": -523.1787719726562,
+      "loss": 0.4664,
+      "rewards/accuracies": 0.800000011920929,
+      "rewards/chosen": -1.6648248434066772,
+      "rewards/margins": 0.9661164283752441,
+      "rewards/rejected": -2.630941390991211,
+      "step": 600
+    },
+    {
+      "epoch": 0.6279434850863422,
+      "eval_logits/chosen": -0.26891058683395386,
+      "eval_logits/rejected": 1.055680513381958,
+      "eval_logps/chosen": -450.7947692871094,
+      "eval_logps/rejected": -530.9400634765625,
+      "eval_loss": 0.49440109729766846,
+      "eval_rewards/accuracies": 0.7698412537574768,
+      "eval_rewards/chosen": -1.6882680654525757,
+      "eval_rewards/margins": 1.01906418800354,
+      "eval_rewards/rejected": -2.7073323726654053,
+      "eval_runtime": 176.7399,
+      "eval_samples_per_second": 11.316,
+      "eval_steps_per_second": 0.356,
+      "step": 600
+    },
+    {
+      "epoch": 0.6384092098377813,
+      "grad_norm": 37.553002444239766,
+      "learning_rate": 1.7396415252139288e-07,
+      "logits/chosen": -0.38198137283325195,
+      "logits/rejected": 1.3065571784973145,
+      "logps/chosen": -473.23046875,
+      "logps/rejected": -518.6094360351562,
+      "loss": 0.4964,
+      "rewards/accuracies": 0.762499988079071,
+      "rewards/chosen": -1.6766010522842407,
+      "rewards/margins": 1.062415361404419,
+      "rewards/rejected": -2.73901629447937,
+      "step": 610
+    },
+    {
+      "epoch": 0.6488749345892203,
+      "grad_norm": 42.193220673193395,
+      "learning_rate": 1.6530691736402316e-07,
+      "logits/chosen": -0.21174755692481995,
+      "logits/rejected": 0.9794837236404419,
+      "logps/chosen": -429.8130798339844,
+      "logps/rejected": -492.4026794433594,
+      "loss": 0.4665,
+      "rewards/accuracies": 0.800000011920929,
+      "rewards/chosen": -1.5878998041152954,
+      "rewards/margins": 0.9720605611801147,
+      "rewards/rejected": -2.5599606037139893,
+      "step": 620
+    },
+    {
+      "epoch": 0.6593406593406593,
+      "grad_norm": 38.97702529967974,
+      "learning_rate": 1.5676295169786864e-07,
+      "logits/chosen": -0.6409791707992554,
+      "logits/rejected": 0.6868582367897034,
+      "logps/chosen": -423.3990783691406,
+      "logps/rejected": -487.27117919921875,
+      "loss": 0.4718,
+      "rewards/accuracies": 0.78125,
+      "rewards/chosen": -1.4871389865875244,
+      "rewards/margins": 1.025777816772461,
+      "rewards/rejected": -2.5129168033599854,
+      "step": 630
+    },
+    {
+      "epoch": 0.6698063840920984,
+      "grad_norm": 47.68708831419142,
+      "learning_rate": 1.483436823197092e-07,
+      "logits/chosen": -0.141208216547966,
+      "logits/rejected": 1.2582366466522217,
+      "logps/chosen": -447.75701904296875,
+      "logps/rejected": -507.4087829589844,
+      "loss": 0.4792,
+      "rewards/accuracies": 0.762499988079071,
+      "rewards/chosen": -1.8073606491088867,
+      "rewards/margins": 0.9799903631210327,
+      "rewards/rejected": -2.787351131439209,
+      "step": 640
+    },
+    {
+      "epoch": 0.6802721088435374,
+      "grad_norm": 58.635870372435406,
+      "learning_rate": 1.4006036925609243e-07,
+      "logits/chosen": -0.6169118881225586,
+      "logits/rejected": 0.4020913243293762,
+      "logps/chosen": -419.74786376953125,
+      "logps/rejected": -509.29840087890625,
+      "loss": 0.4833,
+      "rewards/accuracies": 0.71875,
+      "rewards/chosen": -1.5206490755081177,
+      "rewards/margins": 0.8552547693252563,
+      "rewards/rejected": -2.375903606414795,
+      "step": 650
+    },
+    {
+      "epoch": 0.6907378335949764,
+      "grad_norm": 38.14798973487795,
+      "learning_rate": 1.319240907040458e-07,
+      "logits/chosen": -0.6595967411994934,
+      "logits/rejected": 0.701393723487854,
+      "logps/chosen": -430.74365234375,
+      "logps/rejected": -514.9932250976562,
+      "loss": 0.4825,
+      "rewards/accuracies": 0.75,
+      "rewards/chosen": -1.485804557800293,
+      "rewards/margins": 1.0895816087722778,
+      "rewards/rejected": -2.5753860473632812,
+      "step": 660
+    },
+    {
+      "epoch": 0.7012035583464155,
+      "grad_norm": 45.96703369486118,
+      "learning_rate": 1.239457282149695e-07,
+      "logits/chosen": -0.6679301261901855,
+      "logits/rejected": 0.42252644896507263,
+      "logps/chosen": -438.337158203125,
+      "logps/rejected": -540.0033569335938,
+      "loss": 0.472,
+      "rewards/accuracies": 0.793749988079071,
+      "rewards/chosen": -1.4978899955749512,
+      "rewards/margins": 1.0711971521377563,
+      "rewards/rejected": -2.569087028503418,
+      "step": 670
+    },
+    {
+      "epoch": 0.7116692830978545,
+      "grad_norm": 36.89505960468928,
+      "learning_rate": 1.1613595214152711e-07,
+      "logits/chosen": -0.2877568006515503,
+      "logits/rejected": 0.7360206246376038,
+      "logps/chosen": -468.3255920410156,
+      "logps/rejected": -552.3143310546875,
+      "loss": 0.5059,
+      "rewards/accuracies": 0.762499988079071,
+      "rewards/chosen": -1.5666911602020264,
+      "rewards/margins": 1.0774924755096436,
+      "rewards/rejected": -2.64418363571167,
+      "step": 680
+    },
+    {
+      "epoch": 0.7221350078492935,
+      "grad_norm": 26.931344238941527,
+      "learning_rate": 1.0850520736699362e-07,
+      "logits/chosen": -0.49702662229537964,
+      "logits/rejected": 0.9417774081230164,
+      "logps/chosen": -430.6412658691406,
+      "logps/rejected": -496.180419921875,
+      "loss": 0.4527,
+      "rewards/accuracies": 0.8062499761581421,
+      "rewards/chosen": -1.5998663902282715,
+      "rewards/margins": 1.0349512100219727,
+      "rewards/rejected": -2.634817361831665,
+      "step": 690
+    },
+    {
+      "epoch": 0.7326007326007326,
+      "grad_norm": 34.042068496483864,
+      "learning_rate": 1.0106369933615042e-07,
+      "logits/chosen": -0.22161564230918884,
+      "logits/rejected": 0.7004293203353882,
+      "logps/chosen": -416.830322265625,
+      "logps/rejected": -488.65106201171875,
+      "loss": 0.4716,
+      "rewards/accuracies": 0.7562500238418579,
+      "rewards/chosen": -1.6273488998413086,
+      "rewards/margins": 0.8736777305603027,
+      "rewards/rejected": -2.5010266304016113,
+      "step": 700
+    },
+    {
+      "epoch": 0.7326007326007326,
+      "eval_logits/chosen": -0.3817674517631531,
+      "eval_logits/rejected": 0.7091061472892761,
+      "eval_logps/chosen": -438.7652893066406,
+      "eval_logps/rejected": -520.2406005859375,
+      "eval_loss": 0.49287474155426025,
+      "eval_rewards/accuracies": 0.7777777910232544,
+      "eval_rewards/chosen": -1.5679733753204346,
+      "eval_rewards/margins": 1.032364010810852,
+      "eval_rewards/rejected": -2.600337505340576,
+      "eval_runtime": 175.3398,
+      "eval_samples_per_second": 11.406,
+      "eval_steps_per_second": 0.359,
+      "step": 700
+    },
+    {
+      "epoch": 0.7430664573521716,
+      "grad_norm": 34.12335024104434,
+      "learning_rate": 9.382138040640714e-08,
+      "logits/chosen": -0.200923353433609,
+      "logits/rejected": 1.0540679693222046,
+      "logps/chosen": -429.7672424316406,
+      "logps/rejected": -506.0777282714844,
+      "loss": 0.5254,
+      "rewards/accuracies": 0.7250000238418579,
+      "rewards/chosen": -1.7129242420196533,
+      "rewards/margins": 0.9184734225273132,
+      "rewards/rejected": -2.6313977241516113,
+      "step": 710
+    },
+    {
+      "epoch": 0.7535321821036107,
+      "grad_norm": 30.02495734311258,
+      "learning_rate": 8.678793653740632e-08,
+      "logits/chosen": -0.33009666204452515,
+      "logits/rejected": 0.9047529101371765,
+      "logps/chosen": -432.42010498046875,
+      "logps/rejected": -512.3065185546875,
+      "loss": 0.4648,
+      "rewards/accuracies": 0.793749988079071,
+      "rewards/chosen": -1.4868285655975342,
+      "rewards/margins": 1.1614835262298584,
+      "rewards/rejected": -2.6483120918273926,
+      "step": 720
+    },
+    {
+      "epoch": 0.7639979068550498,
+      "grad_norm": 38.12612104973478,
+      "learning_rate": 7.997277433690983e-08,
+      "logits/chosen": -0.24449904263019562,
+      "logits/rejected": 0.697944164276123,
+      "logps/chosen": -442.3524475097656,
+      "logps/rejected": -493.40228271484375,
+      "loss": 0.4935,
+      "rewards/accuracies": 0.762499988079071,
+      "rewards/chosen": -1.7699024677276611,
+      "rewards/margins": 0.7942432761192322,
+      "rewards/rejected": -2.564145565032959,
+      "step": 730
+    },
+    {
+      "epoch": 0.7744636316064888,
+      "grad_norm": 34.63386515718176,
+      "learning_rate": 7.338500848029602e-08,
+      "logits/chosen": -0.09529142081737518,
+      "logits/rejected": 0.7665776610374451,
+      "logps/chosen": -415.37127685546875,
+      "logps/rejected": -480.74981689453125,
+      "loss": 0.5094,
+      "rewards/accuracies": 0.731249988079071,
+      "rewards/chosen": -1.6250483989715576,
+      "rewards/margins": 0.8600455522537231,
+      "rewards/rejected": -2.485093832015991,
+      "step": 740
+    },
+    {
+      "epoch": 0.7849293563579278,
+      "grad_norm": 38.170157629518556,
+      "learning_rate": 6.70334495204884e-08,
+      "logits/chosen": -0.16666540503501892,
+      "logits/rejected": 0.6466744542121887,
+      "logps/chosen": -414.00482177734375,
+      "logps/rejected": -509.15478515625,
+      "loss": 0.4725,
+      "rewards/accuracies": 0.7250000238418579,
+      "rewards/chosen": -1.6713943481445312,
+      "rewards/margins": 0.9836255311965942,
+      "rewards/rejected": -2.655019760131836,
+      "step": 750
+    },
+    {
+      "epoch": 0.7953950811093669,
+      "grad_norm": 32.154322102412,
+      "learning_rate": 6.092659210462231e-08,
+      "logits/chosen": -0.10375770181417465,
+      "logits/rejected": 0.7752261757850647,
+      "logps/chosen": -410.65399169921875,
+      "logps/rejected": -507.827880859375,
+      "loss": 0.5116,
+      "rewards/accuracies": 0.7562500238418579,
+      "rewards/chosen": -1.6930710077285767,
+      "rewards/margins": 0.9550973773002625,
+      "rewards/rejected": -2.6481685638427734,
+      "step": 760
+    },
+    {
+      "epoch": 0.8058608058608059,
+      "grad_norm": 29.7627819143098,
+      "learning_rate": 5.507260361320737e-08,
+      "logits/chosen": -0.4120585322380066,
+      "logits/rejected": 0.24446937441825867,
+      "logps/chosen": -466.15069580078125,
+      "logps/rejected": -568.6295776367188,
+      "loss": 0.4641,
+      "rewards/accuracies": 0.7250000238418579,
+      "rewards/chosen": -1.6279999017715454,
+      "rewards/margins": 0.8385451436042786,
+      "rewards/rejected": -2.4665448665618896,
+      "step": 770
+    },
+    {
+      "epoch": 0.8163265306122449,
+      "grad_norm": 40.436476118941954,
+      "learning_rate": 4.947931323697982e-08,
+      "logits/chosen": -0.4582279324531555,
+      "logits/rejected": 0.3044646382331848,
+      "logps/chosen": -473.2518005371094,
+      "logps/rejected": -511.49334716796875,
+      "loss": 0.5022,
+      "rewards/accuracies": 0.737500011920929,
+      "rewards/chosen": -1.542907953262329,
+      "rewards/margins": 0.8963887095451355,
+      "rewards/rejected": -2.4392964839935303,
+      "step": 780
+    },
+    {
+      "epoch": 0.826792255363684,
+      "grad_norm": 38.14021936834171,
+      "learning_rate": 4.415420150605398e-08,
+      "logits/chosen": -0.21675653755664825,
+      "logits/rejected": 1.2191559076309204,
+      "logps/chosen": -446.891357421875,
+      "logps/rejected": -514.4319458007812,
+      "loss": 0.5074,
+      "rewards/accuracies": 0.793749988079071,
+      "rewards/chosen": -1.6831605434417725,
+      "rewards/margins": 1.0085071325302124,
+      "rewards/rejected": -2.6916680335998535,
+      "step": 790
+    },
+    {
+      "epoch": 0.837257980115123,
+      "grad_norm": 33.872578527467745,
+      "learning_rate": 3.9104390285376374e-08,
+      "logits/chosen": 0.09026177227497101,
+      "logits/rejected": 0.9493114352226257,
+      "logps/chosen": -439.46429443359375,
+      "logps/rejected": -529.3217163085938,
+      "loss": 0.4741,
+      "rewards/accuracies": 0.7562500238418579,
+      "rewards/chosen": -1.7335838079452515,
+      "rewards/margins": 0.9264078140258789,
+      "rewards/rejected": -2.65999174118042,
+      "step": 800
+    },
+    {
+      "epoch": 0.837257980115123,
+      "eval_logits/chosen": -0.3940890431404114,
+      "eval_logits/rejected": 0.6599191427230835,
+      "eval_logps/chosen": -437.01190185546875,
+      "eval_logps/rejected": -517.6414794921875,
+      "eval_loss": 0.49065762758255005,
+      "eval_rewards/accuracies": 0.7777777910232544,
+      "eval_rewards/chosen": -1.5504390001296997,
+      "eval_rewards/margins": 1.02390718460083,
+      "eval_rewards/rejected": -2.5743465423583984,
+      "eval_runtime": 176.8304,
+      "eval_samples_per_second": 11.31,
+      "eval_steps_per_second": 0.356,
+      "step": 800
+    },
+    {
+      "epoch": 0.847723704866562,
+      "grad_norm": 32.1124717145771,
+      "learning_rate": 3.433663324986208e-08,
+      "logits/chosen": -0.3709171712398529,
+      "logits/rejected": 0.708153486251831,
+      "logps/chosen": -451.8567810058594,
+      "logps/rejected": -501.13153076171875,
+      "loss": 0.4866,
+      "rewards/accuracies": 0.762499988079071,
+      "rewards/chosen": -1.606183409690857,
+      "rewards/margins": 0.9406149983406067,
+      "rewards/rejected": -2.5467987060546875,
+      "step": 810
+    },
+    {
+      "epoch": 0.858189429618001,
+      "grad_norm": 36.42213321353283,
+      "learning_rate": 2.9857306851953897e-08,
+      "logits/chosen": -0.5197598934173584,
+      "logits/rejected": 0.44594240188598633,
+      "logps/chosen": -462.8509216308594,
+      "logps/rejected": -537.5787963867188,
+      "loss": 0.498,
+      "rewards/accuracies": 0.737500011920929,
+      "rewards/chosen": -1.5807650089263916,
+      "rewards/margins": 0.9640032052993774,
+      "rewards/rejected": -2.5447685718536377,
+      "step": 820
+    },
+    {
+      "epoch": 0.8686551543694401,
+      "grad_norm": 33.241447274140704,
+      "learning_rate": 2.567240179368185e-08,
+      "logits/chosen": -0.35766562819480896,
+      "logits/rejected": 0.4714294373989105,
+      "logps/chosen": -418.21746826171875,
+      "logps/rejected": -512.0104370117188,
+      "loss": 0.5054,
+      "rewards/accuracies": 0.731249988079071,
+      "rewards/chosen": -1.600725531578064,
+      "rewards/margins": 1.0214117765426636,
+      "rewards/rejected": -2.6221375465393066,
+      "step": 830
+    },
+    {
+      "epoch": 0.8791208791208791,
+      "grad_norm": 30.190104504480274,
+      "learning_rate": 2.1787515014630357e-08,
+      "logits/chosen": -0.35540395975112915,
+      "logits/rejected": 0.8115441203117371,
+      "logps/chosen": -448.0243225097656,
+      "logps/rejected": -531.029541015625,
+      "loss": 0.4616,
+      "rewards/accuracies": 0.75,
+      "rewards/chosen": -1.6575037240982056,
+      "rewards/margins": 0.9920711517333984,
+      "rewards/rejected": -2.6495752334594727,
+      "step": 840
+    },
+    {
+      "epoch": 0.8895866038723181,
+      "grad_norm": 41.82041936139999,
+      "learning_rate": 1.820784220652766e-08,
+      "logits/chosen": -0.12738969922065735,
+      "logits/rejected": 1.10179603099823,
+      "logps/chosen": -434.4571228027344,
+      "logps/rejected": -497.9303283691406,
+      "loss": 0.455,
+      "rewards/accuracies": 0.793749988079071,
+      "rewards/chosen": -1.5362211465835571,
+      "rewards/margins": 1.1085121631622314,
+      "rewards/rejected": -2.64473295211792,
+      "step": 850
+    },
+    {
+      "epoch": 0.9000523286237572,
+      "grad_norm": 34.694500021885084,
+      "learning_rate": 1.4938170864468636e-08,
+      "logits/chosen": -0.25795820355415344,
+      "logits/rejected": 0.9191703796386719,
+      "logps/chosen": -451.52752685546875,
+      "logps/rejected": -523.0709838867188,
+      "loss": 0.4787,
+      "rewards/accuracies": 0.7749999761581421,
+      "rewards/chosen": -1.6323562860488892,
+      "rewards/margins": 0.984820544719696,
+      "rewards/rejected": -2.6171765327453613,
+      "step": 860
+    },
+    {
+      "epoch": 0.9105180533751962,
+      "grad_norm": 34.514941887456985,
+      "learning_rate": 1.1982873884064465e-08,
+      "logits/chosen": -0.48437461256980896,
+      "logits/rejected": 0.7289548516273499,
+      "logps/chosen": -428.81390380859375,
+      "logps/rejected": -501.2752990722656,
+      "loss": 0.489,
+      "rewards/accuracies": 0.762499988079071,
+      "rewards/chosen": -1.5704351663589478,
+      "rewards/margins": 0.9770514369010925,
+      "rewards/rejected": -2.5474865436553955,
+      "step": 870
+    },
+    {
+      "epoch": 0.9209837781266352,
+      "grad_norm": 33.54810196280602,
+      "learning_rate": 9.345903713082304e-09,
+      "logits/chosen": -0.3731919825077057,
+      "logits/rejected": 0.4349101185798645,
+      "logps/chosen": -459.6202697753906,
+      "logps/rejected": -546.8380126953125,
+      "loss": 0.4823,
+      "rewards/accuracies": 0.731249988079071,
+      "rewards/chosen": -1.5548641681671143,
+      "rewards/margins": 0.9671527743339539,
+      "rewards/rejected": -2.522017002105713,
+      "step": 880
+    },
+    {
+      "epoch": 0.9314495028780743,
+      "grad_norm": 32.531647089433314,
+      "learning_rate": 7.030787065396865e-09,
+      "logits/chosen": -0.3866179585456848,
+      "logits/rejected": 0.5604432225227356,
+      "logps/chosen": -444.09967041015625,
+      "logps/rejected": -537.5520629882812,
+      "loss": 0.4825,
+      "rewards/accuracies": 0.7437499761581421,
+      "rewards/chosen": -1.5642650127410889,
+      "rewards/margins": 1.0092883110046387,
+      "rewards/rejected": -2.5735533237457275,
+      "step": 890
+    },
+    {
+      "epoch": 0.9419152276295133,
+      "grad_norm": 41.37080153124819,
+      "learning_rate": 5.04062020432286e-09,
+      "logits/chosen": -0.411233514547348,
+      "logits/rejected": 0.5156325101852417,
+      "logps/chosen": -434.09906005859375,
+      "logps/rejected": -530.5025024414062,
+      "loss": 0.4741,
+      "rewards/accuracies": 0.793749988079071,
+      "rewards/chosen": -1.5370757579803467,
+      "rewards/margins": 0.953914999961853,
+      "rewards/rejected": -2.49099063873291,
+      "step": 900
+    },
+    {
+      "epoch": 0.9419152276295133,
+      "eval_logits/chosen": -0.3568094074726105,
+      "eval_logits/rejected": 0.7152466773986816,
+      "eval_logps/chosen": -440.10498046875,
+      "eval_logps/rejected": -523.0108032226562,
+      "eval_loss": 0.49055662751197815,
+      "eval_rewards/accuracies": 0.7698412537574768,
+      "eval_rewards/chosen": -1.5813697576522827,
+      "eval_rewards/margins": 1.0466694831848145,
+      "eval_rewards/rejected": -2.6280391216278076,
+      "eval_runtime": 175.5396,
+      "eval_samples_per_second": 11.393,
+      "eval_steps_per_second": 0.359,
+      "step": 900
+    },
+    {
+      "epoch": 0.9523809523809523,
+      "grad_norm": 47.043240406771055,
+      "learning_rate": 3.3780648016376866e-09,
+      "logits/chosen": -0.13907718658447266,
+      "logits/rejected": 0.8174427151679993,
+      "logps/chosen": -380.21087646484375,
+      "logps/rejected": -457.331298828125,
+      "loss": 0.4873,
+      "rewards/accuracies": 0.75,
+      "rewards/chosen": -1.5678941011428833,
+      "rewards/margins": 0.8766323924064636,
+      "rewards/rejected": -2.444526433944702,
+      "step": 910
+    },
+    {
+      "epoch": 0.9628466771323915,
+      "grad_norm": 41.04154104780731,
+      "learning_rate": 2.0453443778310766e-09,
+      "logits/chosen": -0.3518763482570648,
+      "logits/rejected": 0.7049046754837036,
+      "logps/chosen": -444.53271484375,
+      "logps/rejected": -510.38812255859375,
+      "loss": 0.4989,
+      "rewards/accuracies": 0.768750011920929,
+      "rewards/chosen": -1.5789575576782227,
+      "rewards/margins": 1.008893370628357,
+      "rewards/rejected": -2.587851047515869,
+      "step": 920
+    },
+    {
+      "epoch": 0.9733124018838305,
+      "grad_norm": 31.21202125429197,
+      "learning_rate": 1.0442413283435758e-09,
+      "logits/chosen": -0.5221754312515259,
+      "logits/rejected": 1.012073040008545,
+      "logps/chosen": -460.8104553222656,
+      "logps/rejected": -517.5181884765625,
+      "loss": 0.4808,
+      "rewards/accuracies": 0.762499988079071,
+      "rewards/chosen": -1.4597631692886353,
+      "rewards/margins": 1.07325279712677,
+      "rewards/rejected": -2.5330162048339844,
+      "step": 930
+    },
+    {
+      "epoch": 0.9837781266352695,
+      "grad_norm": 31.068531683647514,
+      "learning_rate": 3.760945397705828e-10,
+      "logits/chosen": -0.38587886095046997,
+      "logits/rejected": 0.7057468891143799,
+      "logps/chosen": -410.48834228515625,
+      "logps/rejected": -507.0341796875,
+      "loss": 0.4773,
+      "rewards/accuracies": 0.71875,
+      "rewards/chosen": -1.4730709791183472,
+      "rewards/margins": 1.0000593662261963,
+      "rewards/rejected": -2.473130464553833,
+      "step": 940
+    },
+    {
+      "epoch": 0.9942438513867086,
+      "grad_norm": 32.463686030627095,
+      "learning_rate": 4.17975992204056e-11,
+      "logits/chosen": -0.2789713740348816,
+      "logits/rejected": 0.4086712896823883,
+      "logps/chosen": -464.32086181640625,
+      "logps/rejected": -530.4679565429688,
+      "loss": 0.4828,
+      "rewards/accuracies": 0.731249988079071,
+      "rewards/chosen": -1.6147760152816772,
+      "rewards/margins": 0.8401002883911133,
+      "rewards/rejected": -2.45487642288208,
+      "step": 950
+    },
+    {
+      "epoch": 0.9994767137624281,
+      "step": 955,
+      "total_flos": 0.0,
+      "train_loss": 0.5247097397349891,
+      "train_runtime": 18219.8315,
+      "train_samples_per_second": 3.355,
+      "train_steps_per_second": 0.052
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 955,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 1,
+  "save_steps": 100,
+  "total_flos": 0.0,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}