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: 0.4963
+- Rewards/chosen: -1.3653
+- Rewards/rejected: -2.3944
+- Rewards/accuracies: 0.7639
+- Rewards/margins: 1.0292
+- Logps/rejected: -499.6516
+- Logps/chosen: -418.4965
+- Logits/rejected: 1.1202
+- Logits/chosen: 0.2514
+
+## 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 | Logits/chosen | Logits/rejected | Logps/chosen | Logps/rejected | Validation Loss | Rewards/accuracies | Rewards/chosen | Rewards/margins | Rewards/rejected |
+|:-------------:|:------:|:----:|:-------------:|:---------------:|:------------:|:--------------:|:---------------:|:------------------:|:--------------:|:---------------:|:----------------:|
+| 0.6267        | 0.1047 | 100  | -2.4885       | -2.4228         | -298.8784    | -298.6605      | 0.6219          | 0.7123             | -0.1691        | 0.2154          | -0.3845          |
+| 0.5618        | 0.2093 | 200  | -1.1643       | -0.9025         | -368.7673    | -409.5043      | 0.5559          | 0.75               | -0.8680        | 0.6250          | -1.4930          |
+| 0.5298        | 0.3140 | 300  | 0.1099        | 0.5848          | -387.0007    | -442.8027      | 0.5283          | 0.7679             | -1.0503        | 0.7756          | -1.8260          |
+| 0.5585        | 0.4186 | 400  | -1.2260       | -0.6461         | -368.9012    | -419.9413      | 0.5200          | 0.7639             | -0.8693        | 0.7280          | -1.5973          |
+| 0.5074        | 0.5233 | 500  | 0.1651        | 0.9655          | -402.5899    | -472.5687      | 0.5043          | 0.7698             | -1.2062        | 0.9174          | -2.1236          |
+| 0.4678        | 0.6279 | 600  | -0.1276       | 0.7528          | -405.3750    | -480.0774      | 0.4995          | 0.7698             | -1.2341        | 0.9646          | -2.1987          |
+| 0.4767        | 0.7326 | 700  | 0.4974        | -1.3438         | -2.3581      | 0.7619         | 1.0142          | -496.0146          | -416.3518      | 1.1957          | 0.3567           |
+| 0.475         | 0.8373 | 800  | 0.4971        | -1.3985         | -2.3889      | 0.7639         | 0.9904          | -499.1005          | -421.8184      | 1.0101          | 0.2102           |
+| 0.4828        | 0.9419 | 900  | 0.4963        | -1.3653         | -2.3944      | 0.7639         | 1.0292          | -499.6516          | -418.4965      | 1.1202          | 0.2514           |
+
+
+### Framework versions
+
+- Transformers 4.41.2
+- Pytorch 2.3.1+cu121
+- Datasets 2.19.2
+- Tokenizers 0.19.1

all_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 0.9994767137624281,
+    "total_flos": 0.0,
+    "train_loss": 0.1815465917137905,
+    "train_runtime": 7518.522,
+    "train_samples": 61134,
+    "train_samples_per_second": 8.131,
+    "train_steps_per_second": 0.127
+}

generation_config.json

@@ -0,0 +1,6 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 1,
+  "eos_token_id": 2,
+  "transformers_version": "4.41.2"
+}

model-00001-of-00003.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:22c5e94da6ada3519cd5d1f834fcf420403473b4b99f38dec5c822db29af2b36
+size 4943162336

model-00002-of-00003.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8ff73718a399ec2f26d9663d06c6608683708a40e88be87d66aba6183253207a
+size 4999819336

model-00003-of-00003.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bce061382a19ec3620790833c4539cd14996cd7be38a7aa711aceb75fda37311
+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/Jun21_05-38-58_ishant-gpu/events.out.tfevents.1718949441.ishant-gpu.19428.0

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c8349c5c1283e61bd595a0c38ad1bea50c5396298c7e8edd89fcf704bfea48a5
-size 28775
+oid sha256:d9baad43e6ad7dd8b15c7852775e04d5a6f91fca048fe1d75db9997366b7bb9c
+size 32569

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 0.9994767137624281,
+    "total_flos": 0.0,
+    "train_loss": 0.1815465917137905,
+    "train_runtime": 7518.522,
+    "train_samples": 61134,
+    "train_samples_per_second": 8.131,
+    "train_steps_per_second": 0.127
+}

trainer_state.json

@@ -0,0 +1,1611 @@
+{
+  "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": 0.010465724751439037,
+      "grad_norm": 10.83441036661972,
+      "learning_rate": 5.208333333333333e-08,
+      "logits/chosen": -2.6288294792175293,
+      "logits/rejected": -2.5920276641845703,
+      "logps/chosen": -269.02484130859375,
+      "logps/rejected": -247.2714385986328,
+      "loss": 0.6932,
+      "rewards/accuracies": 0.4625000059604645,
+      "rewards/chosen": 0.00044025605893693864,
+      "rewards/margins": 0.000744122255127877,
+      "rewards/rejected": -0.0003038661670871079,
+      "step": 10
+    },
+    {
+      "epoch": 0.020931449502878074,
+      "grad_norm": 12.596112829643161,
+      "learning_rate": 1.0416666666666667e-07,
+      "logits/chosen": -2.613401412963867,
+      "logits/rejected": -2.5758602619171143,
+      "logps/chosen": -283.1111755371094,
+      "logps/rejected": -282.2903747558594,
+      "loss": 0.6929,
+      "rewards/accuracies": 0.48750001192092896,
+      "rewards/chosen": -0.00027975160628557205,
+      "rewards/margins": -0.0001261378638446331,
+      "rewards/rejected": -0.00015361374244093895,
+      "step": 20
+    },
+    {
+      "epoch": 0.03139717425431711,
+      "grad_norm": 12.23952141074508,
+      "learning_rate": 1.5624999999999999e-07,
+      "logits/chosen": -2.6915743350982666,
+      "logits/rejected": -2.667309045791626,
+      "logps/chosen": -270.18572998046875,
+      "logps/rejected": -276.73675537109375,
+      "loss": 0.6921,
+      "rewards/accuracies": 0.5687500238418579,
+      "rewards/chosen": 0.0009639340569265187,
+      "rewards/margins": 0.0015827405732125044,
+      "rewards/rejected": -0.0006188066909089684,
+      "step": 30
+    },
+    {
+      "epoch": 0.04186289900575615,
+      "grad_norm": 11.873929193021679,
+      "learning_rate": 2.0833333333333333e-07,
+      "logits/chosen": -2.6651158332824707,
+      "logits/rejected": -2.5901150703430176,
+      "logps/chosen": -290.4391784667969,
+      "logps/rejected": -282.10711669921875,
+      "loss": 0.69,
+      "rewards/accuracies": 0.643750011920929,
+      "rewards/chosen": 0.005596032831817865,
+      "rewards/margins": 0.006025379989296198,
+      "rewards/rejected": -0.0004293472447898239,
+      "step": 40
+    },
+    {
+      "epoch": 0.052328623757195186,
+      "grad_norm": 13.181894776310466,
+      "learning_rate": 2.604166666666667e-07,
+      "logits/chosen": -2.671447277069092,
+      "logits/rejected": -2.5876405239105225,
+      "logps/chosen": -266.19891357421875,
+      "logps/rejected": -236.5002899169922,
+      "loss": 0.686,
+      "rewards/accuracies": 0.643750011920929,
+      "rewards/chosen": 0.013971127569675446,
+      "rewards/margins": 0.01311213057488203,
+      "rewards/rejected": 0.0008589973440393806,
+      "step": 50
+    },
+    {
+      "epoch": 0.06279434850863422,
+      "grad_norm": 11.864289091919504,
+      "learning_rate": 3.1249999999999997e-07,
+      "logits/chosen": -2.6260297298431396,
+      "logits/rejected": -2.5904088020324707,
+      "logps/chosen": -299.61639404296875,
+      "logps/rejected": -274.44915771484375,
+      "loss": 0.679,
+      "rewards/accuracies": 0.7250000238418579,
+      "rewards/chosen": 0.04407883808016777,
+      "rewards/margins": 0.03104178048670292,
+      "rewards/rejected": 0.013037058524787426,
+      "step": 60
+    },
+    {
+      "epoch": 0.07326007326007326,
+      "grad_norm": 11.92324421228381,
+      "learning_rate": 3.645833333333333e-07,
+      "logits/chosen": -2.5302624702453613,
+      "logits/rejected": -2.5223822593688965,
+      "logps/chosen": -257.5154113769531,
+      "logps/rejected": -262.79608154296875,
+      "loss": 0.6701,
+      "rewards/accuracies": 0.6937500238418579,
+      "rewards/chosen": 0.021234791725873947,
+      "rewards/margins": 0.06698025017976761,
+      "rewards/rejected": -0.04574545472860336,
+      "step": 70
+    },
+    {
+      "epoch": 0.0837257980115123,
+      "grad_norm": 14.017084052960483,
+      "learning_rate": 4.1666666666666667e-07,
+      "logits/chosen": -2.550006151199341,
+      "logits/rejected": -2.477769613265991,
+      "logps/chosen": -274.51092529296875,
+      "logps/rejected": -261.02435302734375,
+      "loss": 0.6505,
+      "rewards/accuracies": 0.706250011920929,
+      "rewards/chosen": 0.015317601151764393,
+      "rewards/margins": 0.11243382841348648,
+      "rewards/rejected": -0.09711623191833496,
+      "step": 80
+    },
+    {
+      "epoch": 0.09419152276295134,
+      "grad_norm": 15.68510838988432,
+      "learning_rate": 4.6874999999999996e-07,
+      "logits/chosen": -2.498664140701294,
+      "logits/rejected": -2.4632701873779297,
+      "logps/chosen": -261.152587890625,
+      "logps/rejected": -277.7818603515625,
+      "loss": 0.6428,
+      "rewards/accuracies": 0.625,
+      "rewards/chosen": -0.05554679036140442,
+      "rewards/margins": 0.07454034686088562,
+      "rewards/rejected": -0.13008712232112885,
+      "step": 90
+    },
+    {
+      "epoch": 0.10465724751439037,
+      "grad_norm": 13.973992370085414,
+      "learning_rate": 4.999732492681437e-07,
+      "logits/chosen": -2.4807047843933105,
+      "logits/rejected": -2.405395030975342,
+      "logps/chosen": -315.41790771484375,
+      "logps/rejected": -313.3694763183594,
+      "loss": 0.6267,
+      "rewards/accuracies": 0.7124999761581421,
+      "rewards/chosen": -0.15593726933002472,
+      "rewards/margins": 0.17772409319877625,
+      "rewards/rejected": -0.33366137742996216,
+      "step": 100
+    },
+    {
+      "epoch": 0.10465724751439037,
+      "eval_logits/chosen": -2.488471031188965,
+      "eval_logits/rejected": -2.4228012561798096,
+      "eval_logps/chosen": -298.8784484863281,
+      "eval_logps/rejected": -298.6604919433594,
+      "eval_loss": 0.6219407916069031,
+      "eval_rewards/accuracies": 0.7123016119003296,
+      "eval_rewards/chosen": -0.16910475492477417,
+      "eval_rewards/margins": 0.2154317945241928,
+      "eval_rewards/rejected": -0.3845365047454834,
+      "eval_runtime": 187.9022,
+      "eval_samples_per_second": 10.644,
+      "eval_steps_per_second": 0.335,
+      "step": 100
+    },
+    {
+      "epoch": 0.1151229722658294,
+      "grad_norm": 15.96260202543978,
+      "learning_rate": 4.996723692767926e-07,
+      "logits/chosen": -2.491821765899658,
+      "logits/rejected": -2.4371533393859863,
+      "logps/chosen": -274.3268127441406,
+      "logps/rejected": -287.5362854003906,
+      "loss": 0.6362,
+      "rewards/accuracies": 0.7124999761581421,
+      "rewards/chosen": -0.32872217893600464,
+      "rewards/margins": 0.17391590774059296,
+      "rewards/rejected": -0.5026381015777588,
+      "step": 110
+    },
+    {
+      "epoch": 0.12558869701726844,
+      "grad_norm": 17.158979848789446,
+      "learning_rate": 4.990375746213598e-07,
+      "logits/chosen": -2.429931879043579,
+      "logits/rejected": -2.371150493621826,
+      "logps/chosen": -276.56256103515625,
+      "logps/rejected": -327.0640563964844,
+      "loss": 0.6013,
+      "rewards/accuracies": 0.7250000238418579,
+      "rewards/chosen": -0.13984180986881256,
+      "rewards/margins": 0.24633511900901794,
+      "rewards/rejected": -0.3861769139766693,
+      "step": 120
+    },
+    {
+      "epoch": 0.1360544217687075,
+      "grad_norm": 22.882706219149167,
+      "learning_rate": 4.980697142834314e-07,
+      "logits/chosen": -2.407017230987549,
+      "logits/rejected": -2.352510452270508,
+      "logps/chosen": -307.400390625,
+      "logps/rejected": -317.0360107421875,
+      "loss": 0.5918,
+      "rewards/accuracies": 0.731249988079071,
+      "rewards/chosen": -0.26127806305885315,
+      "rewards/margins": 0.364439994096756,
+      "rewards/rejected": -0.6257181167602539,
+      "step": 130
+    },
+    {
+      "epoch": 0.14652014652014653,
+      "grad_norm": 17.994424262965605,
+      "learning_rate": 4.967700826904229e-07,
+      "logits/chosen": -2.3387370109558105,
+      "logits/rejected": -2.2937257289886475,
+      "logps/chosen": -318.0648498535156,
+      "logps/rejected": -351.2364807128906,
+      "loss": 0.5806,
+      "rewards/accuracies": 0.737500011920929,
+      "rewards/chosen": -0.2697739601135254,
+      "rewards/margins": 0.3528417944908142,
+      "rewards/rejected": -0.6226157546043396,
+      "step": 140
+    },
+    {
+      "epoch": 0.15698587127158556,
+      "grad_norm": 23.796525050931013,
+      "learning_rate": 4.951404179843962e-07,
+      "logits/chosen": -2.4819083213806152,
+      "logits/rejected": -2.419260025024414,
+      "logps/chosen": -314.9351501464844,
+      "logps/rejected": -308.2325439453125,
+      "loss": 0.5866,
+      "rewards/accuracies": 0.668749988079071,
+      "rewards/chosen": -0.2703876793384552,
+      "rewards/margins": 0.3419578969478607,
+      "rewards/rejected": -0.6123455762863159,
+      "step": 150
+    },
+    {
+      "epoch": 0.1674515960230246,
+      "grad_norm": 22.081896847350087,
+      "learning_rate": 4.931828996974498e-07,
+      "logits/chosen": -2.057835817337036,
+      "logits/rejected": -1.8778709173202515,
+      "logps/chosen": -326.67999267578125,
+      "logps/rejected": -331.5979919433594,
+      "loss": 0.5645,
+      "rewards/accuracies": 0.7250000238418579,
+      "rewards/chosen": -0.30867281556129456,
+      "rewards/margins": 0.44759130477905273,
+      "rewards/rejected": -0.7562640905380249,
+      "step": 160
+    },
+    {
+      "epoch": 0.17791732077446362,
+      "grad_norm": 29.732483466418604,
+      "learning_rate": 4.909001458367866e-07,
+      "logits/chosen": -1.0668941736221313,
+      "logits/rejected": -0.83452308177948,
+      "logps/chosen": -359.36517333984375,
+      "logps/rejected": -380.9996032714844,
+      "loss": 0.5579,
+      "rewards/accuracies": 0.6812499761581421,
+      "rewards/chosen": -0.8089600801467896,
+      "rewards/margins": 0.5925653576850891,
+      "rewards/rejected": -1.4015254974365234,
+      "step": 170
+    },
+    {
+      "epoch": 0.18838304552590268,
+      "grad_norm": 32.71624465909897,
+      "learning_rate": 4.882952093833627e-07,
+      "logits/chosen": -0.9632415771484375,
+      "logits/rejected": -0.6905760765075684,
+      "logps/chosen": -316.61285400390625,
+      "logps/rejected": -393.2853698730469,
+      "loss": 0.5255,
+      "rewards/accuracies": 0.7562500238418579,
+      "rewards/chosen": -0.7666488289833069,
+      "rewards/margins": 0.698399007320404,
+      "rewards/rejected": -1.465047836303711,
+      "step": 180
+    },
+    {
+      "epoch": 0.1988487702773417,
+      "grad_norm": 40.766965114697804,
+      "learning_rate": 4.853715742087946e-07,
+      "logits/chosen": -0.8538354635238647,
+      "logits/rejected": -0.7357890009880066,
+      "logps/chosen": -324.13482666015625,
+      "logps/rejected": -404.38104248046875,
+      "loss": 0.5416,
+      "rewards/accuracies": 0.7250000238418579,
+      "rewards/chosen": -0.8815537691116333,
+      "rewards/margins": 0.6634948253631592,
+      "rewards/rejected": -1.5450485944747925,
+      "step": 190
+    },
+    {
+      "epoch": 0.20931449502878074,
+      "grad_norm": 29.235739332049928,
+      "learning_rate": 4.821331504159906e-07,
+      "logits/chosen": -1.136308193206787,
+      "logits/rejected": -0.8372336626052856,
+      "logps/chosen": -400.89337158203125,
+      "logps/rejected": -403.34661865234375,
+      "loss": 0.5618,
+      "rewards/accuracies": 0.737500011920929,
+      "rewards/chosen": -0.9023653268814087,
+      "rewards/margins": 0.5599400997161865,
+      "rewards/rejected": -1.4623053073883057,
+      "step": 200
+    },
+    {
+      "epoch": 0.20931449502878074,
+      "eval_logits/chosen": -1.164330244064331,
+      "eval_logits/rejected": -0.902479887008667,
+      "eval_logps/chosen": -368.7673034667969,
+      "eval_logps/rejected": -409.5042724609375,
+      "eval_loss": 0.5558701753616333,
+      "eval_rewards/accuracies": 0.75,
+      "eval_rewards/chosen": -0.8679934144020081,
+      "eval_rewards/margins": 0.6249809265136719,
+      "eval_rewards/rejected": -1.4929742813110352,
+      "eval_runtime": 187.9171,
+      "eval_samples_per_second": 10.643,
+      "eval_steps_per_second": 0.335,
+      "step": 200
+    },
+    {
+      "epoch": 0.21978021978021978,
+      "grad_norm": 31.58922442348176,
+      "learning_rate": 4.785842691097342e-07,
+      "logits/chosen": -1.2115342617034912,
+      "logits/rejected": -0.8818065524101257,
+      "logps/chosen": -376.1970520019531,
+      "logps/rejected": -423.3211364746094,
+      "loss": 0.5707,
+      "rewards/accuracies": 0.675000011920929,
+      "rewards/chosen": -0.9338651895523071,
+      "rewards/margins": 0.5767671465873718,
+      "rewards/rejected": -1.5106322765350342,
+      "step": 210
+    },
+    {
+      "epoch": 0.2302459445316588,
+      "grad_norm": 32.894317779850255,
+      "learning_rate": 4.7472967660421603e-07,
+      "logits/chosen": -1.2553176879882812,
+      "logits/rejected": -0.8748141527175903,
+      "logps/chosen": -373.59405517578125,
+      "logps/rejected": -387.2649841308594,
+      "loss": 0.5356,
+      "rewards/accuracies": 0.737500011920929,
+      "rewards/chosen": -0.8638602495193481,
+      "rewards/margins": 0.6470237970352173,
+      "rewards/rejected": -1.5108838081359863,
+      "step": 220
+    },
+    {
+      "epoch": 0.24071166928309787,
+      "grad_norm": 28.08326614419892,
+      "learning_rate": 4.705745280752585e-07,
+      "logits/chosen": -0.6662174463272095,
+      "logits/rejected": -0.4035705029964447,
+      "logps/chosen": -383.9234313964844,
+      "logps/rejected": -422.6239318847656,
+      "loss": 0.5459,
+      "rewards/accuracies": 0.731249988079071,
+      "rewards/chosen": -0.9519649744033813,
+      "rewards/margins": 0.5889177918434143,
+      "rewards/rejected": -1.5408827066421509,
+      "step": 230
+    },
+    {
+      "epoch": 0.25117739403453687,
+      "grad_norm": 39.48203486696788,
+      "learning_rate": 4.6612438066572555e-07,
+      "logits/chosen": -0.6064160466194153,
+      "logits/rejected": -0.3868633806705475,
+      "logps/chosen": -335.01751708984375,
+      "logps/rejected": -390.9185791015625,
+      "loss": 0.5273,
+      "rewards/accuracies": 0.7437499761581421,
+      "rewards/chosen": -0.674433708190918,
+      "rewards/margins": 0.6930244565010071,
+      "rewards/rejected": -1.3674581050872803,
+      "step": 240
+    },
+    {
+      "epoch": 0.2616431187859759,
+      "grad_norm": 34.237893602613276,
+      "learning_rate": 4.6138518605333664e-07,
+      "logits/chosen": -0.47975221276283264,
+      "logits/rejected": -0.14331252872943878,
+      "logps/chosen": -361.3214111328125,
+      "logps/rejected": -415.11041259765625,
+      "loss": 0.5078,
+      "rewards/accuracies": 0.7437499761581421,
+      "rewards/chosen": -0.8353713750839233,
+      "rewards/margins": 0.7399575114250183,
+      "rewards/rejected": -1.575329065322876,
+      "step": 250
+    },
+    {
+      "epoch": 0.272108843537415,
+      "grad_norm": 32.902973510279786,
+      "learning_rate": 4.5636328249082514e-07,
+      "logits/chosen": -0.4277305603027344,
+      "logits/rejected": 0.006294989492744207,
+      "logps/chosen": -380.130615234375,
+      "logps/rejected": -431.3246154785156,
+      "loss": 0.5335,
+      "rewards/accuracies": 0.6812499761581421,
+      "rewards/chosen": -0.9116414189338684,
+      "rewards/margins": 0.5790916681289673,
+      "rewards/rejected": -1.490733027458191,
+      "step": 260
+    },
+    {
+      "epoch": 0.282574568288854,
+      "grad_norm": 32.87398996289449,
+      "learning_rate": 4.510653863290871e-07,
+      "logits/chosen": -0.31390756368637085,
+      "logits/rejected": 0.07414981722831726,
+      "logps/chosen": -358.8498229980469,
+      "logps/rejected": -385.8482360839844,
+      "loss": 0.5059,
+      "rewards/accuracies": 0.7562500238418579,
+      "rewards/chosen": -0.5919984579086304,
+      "rewards/margins": 0.7309141159057617,
+      "rewards/rejected": -1.322912573814392,
+      "step": 270
+    },
+    {
+      "epoch": 0.29304029304029305,
+      "grad_norm": 29.98028421270958,
+      "learning_rate": 4.4549858303465737e-07,
+      "logits/chosen": 0.725542426109314,
+      "logits/rejected": 1.1469298601150513,
+      "logps/chosen": -435.123046875,
+      "logps/rejected": -490.77728271484375,
+      "loss": 0.5528,
+      "rewards/accuracies": 0.731249988079071,
+      "rewards/chosen": -1.388789415359497,
+      "rewards/margins": 0.686213493347168,
+      "rewards/rejected": -2.075002670288086,
+      "step": 280
+    },
+    {
+      "epoch": 0.3035060177917321,
+      "grad_norm": 58.327111920842306,
+      "learning_rate": 4.396703177135261e-07,
+      "logits/chosen": 0.6794191598892212,
+      "logits/rejected": 1.2331650257110596,
+      "logps/chosen": -412.5553283691406,
+      "logps/rejected": -441.51885986328125,
+      "loss": 0.5419,
+      "rewards/accuracies": 0.731249988079071,
+      "rewards/chosen": -1.3977099657058716,
+      "rewards/margins": 0.6473307609558105,
+      "rewards/rejected": -2.0450408458709717,
+      "step": 290
+    },
+    {
+      "epoch": 0.3139717425431711,
+      "grad_norm": 32.108741797348614,
+      "learning_rate": 4.335883851539693e-07,
+      "logits/chosen": 0.3234609365463257,
+      "logits/rejected": 0.596695065498352,
+      "logps/chosen": -386.80084228515625,
+      "logps/rejected": -449.28826904296875,
+      "loss": 0.5298,
+      "rewards/accuracies": 0.71875,
+      "rewards/chosen": -1.177787184715271,
+      "rewards/margins": 0.6378301382064819,
+      "rewards/rejected": -1.815617322921753,
+      "step": 300
+    },
+    {
+      "epoch": 0.3139717425431711,
+      "eval_logits/chosen": 0.10990963876247406,
+      "eval_logits/rejected": 0.5848421454429626,
+      "eval_logps/chosen": -387.000732421875,
+      "eval_logps/rejected": -442.802734375,
+      "eval_loss": 0.5283085703849792,
+      "eval_rewards/accuracies": 0.7678571343421936,
+      "eval_rewards/chosen": -1.0503278970718384,
+      "eval_rewards/margins": 0.7756310105323792,
+      "eval_rewards/rejected": -1.8259588479995728,
+      "eval_runtime": 188.0257,
+      "eval_samples_per_second": 10.637,
+      "eval_steps_per_second": 0.335,
+      "step": 300
+    },
+    {
+      "epoch": 0.32443746729461015,
+      "grad_norm": 36.14740359571429,
+      "learning_rate": 4.272609194017105e-07,
+      "logits/chosen": -0.09506132453680038,
+      "logits/rejected": 0.9149357080459595,
+      "logps/chosen": -386.4206237792969,
+      "logps/rejected": -420.0350646972656,
+      "loss": 0.5151,
+      "rewards/accuracies": 0.75,
+      "rewards/chosen": -1.026588797569275,
+      "rewards/margins": 0.846285343170166,
+      "rewards/rejected": -1.8728742599487305,
+      "step": 310
+    },
+    {
+      "epoch": 0.3349031920460492,
+      "grad_norm": 81.36672999217505,
+      "learning_rate": 4.2069638288135547e-07,
+      "logits/chosen": 0.712912380695343,
+      "logits/rejected": 1.3047354221343994,
+      "logps/chosen": -368.98211669921875,
+      "logps/rejected": -457.4791564941406,
+      "loss": 0.539,
+      "rewards/accuracies": 0.699999988079071,
+      "rewards/chosen": -1.1596534252166748,
+      "rewards/margins": 0.9091463088989258,
+      "rewards/rejected": -2.0687994956970215,
+      "step": 320
+    },
+    {
+      "epoch": 0.3453689167974882,
+      "grad_norm": 37.1171085131941,
+      "learning_rate": 4.139035550786494e-07,
+      "logits/chosen": 0.6312053203582764,
+      "logits/rejected": 1.20461106300354,
+      "logps/chosen": -362.6842041015625,
+      "logps/rejected": -418.4046936035156,
+      "loss": 0.5228,
+      "rewards/accuracies": 0.737500011920929,
+      "rewards/chosen": -1.0634013414382935,
+      "rewards/margins": 0.7995752096176147,
+      "rewards/rejected": -1.8629766702651978,
+      "step": 330
+    },
+    {
+      "epoch": 0.35583464154892724,
+      "grad_norm": 37.627503772606644,
+      "learning_rate": 4.0689152079869306e-07,
+      "logits/chosen": 0.016367901116609573,
+      "logits/rejected": 0.5447463989257812,
+      "logps/chosen": -359.12408447265625,
+      "logps/rejected": -400.4968566894531,
+      "loss": 0.5483,
+      "rewards/accuracies": 0.7250000238418579,
+      "rewards/chosen": -0.8963689804077148,
+      "rewards/margins": 0.7486964464187622,
+      "rewards/rejected": -1.6450653076171875,
+      "step": 340
+    },
+    {
+      "epoch": 0.3663003663003663,
+      "grad_norm": 33.01557559410967,
+      "learning_rate": 3.99669658015821e-07,
+      "logits/chosen": -0.16072605550289154,
+      "logits/rejected": 0.38478875160217285,
+      "logps/chosen": -390.49713134765625,
+      "logps/rejected": -430.6659240722656,
+      "loss": 0.5069,
+      "rewards/accuracies": 0.737500011920929,
+      "rewards/chosen": -0.992621123790741,
+      "rewards/margins": 0.7531676292419434,
+      "rewards/rejected": -1.745788812637329,
+      "step": 350
+    },
+    {
+      "epoch": 0.37676609105180536,
+      "grad_norm": 44.68187947620173,
+      "learning_rate": 3.92247625331392e-07,
+      "logits/chosen": -0.1652616709470749,
+      "logits/rejected": 0.44031864404678345,
+      "logps/chosen": -382.75506591796875,
+      "logps/rejected": -438.5421447753906,
+      "loss": 0.4836,
+      "rewards/accuracies": 0.800000011920929,
+      "rewards/chosen": -0.9687082171440125,
+      "rewards/margins": 0.7890796065330505,
+      "rewards/rejected": -1.7577879428863525,
+      "step": 360
+    },
+    {
+      "epoch": 0.3872318158032444,
+      "grad_norm": 33.273260026493126,
+      "learning_rate": 3.846353490562664e-07,
+      "logits/chosen": 0.03936319425702095,
+      "logits/rejected": 0.5893961191177368,
+      "logps/chosen": -384.0589294433594,
+      "logps/rejected": -427.94195556640625,
+      "loss": 0.5164,
+      "rewards/accuracies": 0.71875,
+      "rewards/chosen": -1.2078273296356201,
+      "rewards/margins": 0.715100109577179,
+      "rewards/rejected": -1.9229274988174438,
+      "step": 370
+    },
+    {
+      "epoch": 0.3976975405546834,
+      "grad_norm": 34.01539660056953,
+      "learning_rate": 3.768430099352445e-07,
+      "logits/chosen": -0.5641449689865112,
+      "logits/rejected": 0.1418592482805252,
+      "logps/chosen": -361.6178894042969,
+      "logps/rejected": -431.11676025390625,
+      "loss": 0.5157,
+      "rewards/accuracies": 0.75,
+      "rewards/chosen": -1.0697083473205566,
+      "rewards/margins": 0.7781006693840027,
+      "rewards/rejected": -1.847809076309204,
+      "step": 380
+    },
+    {
+      "epoch": 0.40816326530612246,
+      "grad_norm": 47.794540434199334,
+      "learning_rate": 3.6888102953122304e-07,
+      "logits/chosen": -0.6706582307815552,
+      "logits/rejected": -0.07904417812824249,
+      "logps/chosen": -381.03082275390625,
+      "logps/rejected": -446.5023498535156,
+      "loss": 0.527,
+      "rewards/accuracies": 0.71875,
+      "rewards/chosen": -0.9941921234130859,
+      "rewards/margins": 0.7871489524841309,
+      "rewards/rejected": -1.7813411951065063,
+      "step": 390
+    },
+    {
+      "epoch": 0.4186289900575615,
+      "grad_norm": 105.50541790205155,
+      "learning_rate": 3.607600562872785e-07,
+      "logits/chosen": -0.7714785933494568,
+      "logits/rejected": -0.36160725355148315,
+      "logps/chosen": -372.08929443359375,
+      "logps/rejected": -427.34033203125,
+      "loss": 0.5585,
+      "rewards/accuracies": 0.6812499761581421,
+      "rewards/chosen": -1.0209808349609375,
+      "rewards/margins": 0.5754439830780029,
+      "rewards/rejected": -1.5964248180389404,
+      "step": 400
+    },
+    {
+      "epoch": 0.4186289900575615,
+      "eval_logits/chosen": -1.2259753942489624,
+      "eval_logits/rejected": -0.6461271643638611,
+      "eval_logps/chosen": -368.9012145996094,
+      "eval_logps/rejected": -419.9412841796875,
+      "eval_loss": 0.5200169682502747,
+      "eval_rewards/accuracies": 0.7638888955116272,
+      "eval_rewards/chosen": -0.8693323731422424,
+      "eval_rewards/margins": 0.7280117869377136,
+      "eval_rewards/rejected": -1.5973442792892456,
+      "eval_runtime": 188.0525,
+      "eval_samples_per_second": 10.635,
+      "eval_steps_per_second": 0.335,
+      "step": 400
+    },
+    {
+      "epoch": 0.4290947148090005,
+      "grad_norm": 26.241716083541785,
+      "learning_rate": 3.5249095128531856e-07,
+      "logits/chosen": -0.8230039477348328,
+      "logits/rejected": -0.25722962617874146,
+      "logps/chosen": -365.2613830566406,
+      "logps/rejected": -401.42303466796875,
+      "loss": 0.533,
+      "rewards/accuracies": 0.737500011920929,
+      "rewards/chosen": -0.935257613658905,
+      "rewards/margins": 0.6104739904403687,
+      "rewards/rejected": -1.545731782913208,
+      "step": 410
+    },
+    {
+      "epoch": 0.43956043956043955,
+      "grad_norm": 39.03130812586807,
+      "learning_rate": 3.4408477372034736e-07,
+      "logits/chosen": -0.37245553731918335,
+      "logits/rejected": 0.7487915754318237,
+      "logps/chosen": -372.9959716796875,
+      "logps/rejected": -427.50543212890625,
+      "loss": 0.5211,
+      "rewards/accuracies": 0.768750011920929,
+      "rewards/chosen": -0.9082446098327637,
+      "rewards/margins": 0.8466314077377319,
+      "rewards/rejected": -1.7548761367797852,
+      "step": 420
+    },
+    {
+      "epoch": 0.4500261643118786,
+      "grad_norm": 36.69584086608421,
+      "learning_rate": 3.3555276610977276e-07,
+      "logits/chosen": 0.6589815020561218,
+      "logits/rejected": 1.4469521045684814,
+      "logps/chosen": -358.0375671386719,
+      "logps/rejected": -443.81805419921875,
+      "loss": 0.5084,
+      "rewards/accuracies": 0.75,
+      "rewards/chosen": -1.042842149734497,
+      "rewards/margins": 0.8935502767562866,
+      "rewards/rejected": -1.9363921880722046,
+      "step": 430
+    },
+    {
+      "epoch": 0.4604918890633176,
+      "grad_norm": 51.30602318273095,
+      "learning_rate": 3.269063392575352e-07,
+      "logits/chosen": 0.5997369885444641,
+      "logits/rejected": 1.3968632221221924,
+      "logps/chosen": -401.80572509765625,
+      "logps/rejected": -433.399169921875,
+      "loss": 0.5186,
+      "rewards/accuracies": 0.7124999761581421,
+      "rewards/chosen": -1.0663107633590698,
+      "rewards/margins": 0.7133516073226929,
+      "rewards/rejected": -1.7796624898910522,
+      "step": 440
+    },
+    {
+      "epoch": 0.47095761381475665,
+      "grad_norm": 33.49322912748958,
+      "learning_rate": 3.1815705699316964e-07,
+      "logits/chosen": 0.1546940952539444,
+      "logits/rejected": 0.9945381879806519,
+      "logps/chosen": -377.54681396484375,
+      "logps/rejected": -482.41571044921875,
+      "loss": 0.5122,
+      "rewards/accuracies": 0.7875000238418579,
+      "rewards/chosen": -0.9515895843505859,
+      "rewards/margins": 1.089389443397522,
+      "rewards/rejected": -2.0409789085388184,
+      "step": 450
+    },
+    {
+      "epoch": 0.48142333856619574,
+      "grad_norm": 30.664524114008092,
+      "learning_rate": 3.0931662070620794e-07,
+      "logits/chosen": -0.1982288807630539,
+      "logits/rejected": 0.6370102167129517,
+      "logps/chosen": -359.25714111328125,
+      "logps/rejected": -424.4283142089844,
+      "loss": 0.5352,
+      "rewards/accuracies": 0.71875,
+      "rewards/chosen": -1.0179884433746338,
+      "rewards/margins": 0.7760909795761108,
+      "rewards/rejected": -1.7940794229507446,
+      "step": 460
+    },
+    {
+      "epoch": 0.49188906331763477,
+      "grad_norm": 46.179707835042336,
+      "learning_rate": 3.003968536966078e-07,
+      "logits/chosen": -0.19681891798973083,
+      "logits/rejected": 0.48038846254348755,
+      "logps/chosen": -375.6156311035156,
+      "logps/rejected": -413.18896484375,
+      "loss": 0.4836,
+      "rewards/accuracies": 0.706250011920929,
+      "rewards/chosen": -0.9182689785957336,
+      "rewards/margins": 0.7142398357391357,
+      "rewards/rejected": -1.6325088739395142,
+      "step": 470
+    },
+    {
+      "epoch": 0.5023547880690737,
+      "grad_norm": 39.61595548058563,
+      "learning_rate": 2.9140968536213693e-07,
+      "logits/chosen": 0.666913628578186,
+      "logits/rejected": 1.8080024719238281,
+      "logps/chosen": -383.5103454589844,
+      "logps/rejected": -473.51617431640625,
+      "loss": 0.508,
+      "rewards/accuracies": 0.8062499761581421,
+      "rewards/chosen": -1.1629759073257446,
+      "rewards/margins": 1.0592198371887207,
+      "rewards/rejected": -2.2221953868865967,
+      "step": 480
+    },
+    {
+      "epoch": 0.5128205128205128,
+      "grad_norm": 39.54220015635658,
+      "learning_rate": 2.823671352438608e-07,
+      "logits/chosen": 0.3436744213104248,
+      "logits/rejected": 1.8912662267684937,
+      "logps/chosen": -413.87078857421875,
+      "logps/rejected": -469.492431640625,
+      "loss": 0.4843,
+      "rewards/accuracies": 0.8500000238418579,
+      "rewards/chosen": -1.0193939208984375,
+      "rewards/margins": 1.1113532781600952,
+      "rewards/rejected": -2.1307473182678223,
+      "step": 490
+    },
+    {
+      "epoch": 0.5232862375719518,
+      "grad_norm": 33.25619580338524,
+      "learning_rate": 2.73281296951072e-07,
+      "logits/chosen": 0.5848696827888489,
+      "logits/rejected": 1.3165467977523804,
+      "logps/chosen": -379.54254150390625,
+      "logps/rejected": -430.6253967285156,
+      "loss": 0.5074,
+      "rewards/accuracies": 0.7749999761581421,
+      "rewards/chosen": -1.2089840173721313,
+      "rewards/margins": 0.8229929804801941,
+      "rewards/rejected": -2.0319769382476807,
+      "step": 500
+    },
+    {
+      "epoch": 0.5232862375719518,
+      "eval_logits/chosen": 0.16514113545417786,
+      "eval_logits/rejected": 0.9655360579490662,
+      "eval_logps/chosen": -402.58990478515625,
+      "eval_logps/rejected": -472.5686950683594,
+      "eval_loss": 0.5042572021484375,
+      "eval_rewards/accuracies": 0.7698412537574768,
+      "eval_rewards/chosen": -1.2062196731567383,
+      "eval_rewards/margins": 0.9173988699913025,
+      "eval_rewards/rejected": -2.1236186027526855,
+      "eval_runtime": 188.128,
+      "eval_samples_per_second": 10.631,
+      "eval_steps_per_second": 0.335,
+      "step": 500
+    },
+    {
+      "epoch": 0.533751962323391,
+      "grad_norm": 37.202500300291135,
+      "learning_rate": 2.641643219871597e-07,
+      "logits/chosen": -0.01917283609509468,
+      "logits/rejected": 0.7179991006851196,
+      "logps/chosen": -378.31353759765625,
+      "logps/rejected": -470.5415954589844,
+      "loss": 0.4773,
+      "rewards/accuracies": 0.793749988079071,
+      "rewards/chosen": -0.9437530636787415,
+      "rewards/margins": 1.1235778331756592,
+      "rewards/rejected": -2.067330837249756,
+      "step": 510
+    },
+    {
+      "epoch": 0.54421768707483,
+      "grad_norm": 35.63609405402596,
+      "learning_rate": 2.550284034980507e-07,
+      "logits/chosen": 0.1290065050125122,
+      "logits/rejected": 0.8995389938354492,
+      "logps/chosen": -391.45245361328125,
+      "logps/rejected": -450.935302734375,
+      "loss": 0.5197,
+      "rewards/accuracies": 0.699999988079071,
+      "rewards/chosen": -1.1938081979751587,
+      "rewards/margins": 0.8160873651504517,
+      "rewards/rejected": -2.0098955631256104,
+      "step": 520
+    },
+    {
+      "epoch": 0.554683411826269,
+      "grad_norm": 37.764443273543485,
+      "learning_rate": 2.4588575996495794e-07,
+      "logits/chosen": -0.4371569752693176,
+      "logits/rejected": 0.2857760488986969,
+      "logps/chosen": -350.2786560058594,
+      "logps/rejected": -428.732666015625,
+      "loss": 0.505,
+      "rewards/accuracies": 0.7562500238418579,
+      "rewards/chosen": -0.8153169751167297,
+      "rewards/margins": 0.9217718839645386,
+      "rewards/rejected": -1.7370887994766235,
+      "step": 530
+    },
+    {
+      "epoch": 0.565149136577708,
+      "grad_norm": 34.380579779251335,
+      "learning_rate": 2.367486188632446e-07,
+      "logits/chosen": -0.2316119372844696,
+      "logits/rejected": 0.5598622560501099,
+      "logps/chosen": -385.16131591796875,
+      "logps/rejected": -442.238525390625,
+      "loss": 0.5165,
+      "rewards/accuracies": 0.706250011920929,
+      "rewards/chosen": -0.9481117129325867,
+      "rewards/margins": 0.8683555722236633,
+      "rewards/rejected": -1.816467523574829,
+      "step": 540
+    },
+    {
+      "epoch": 0.5756148613291471,
+      "grad_norm": 44.258921212465225,
+      "learning_rate": 2.276292003092593e-07,
+      "logits/chosen": 0.2254778891801834,
+      "logits/rejected": 1.186446189880371,
+      "logps/chosen": -401.5592346191406,
+      "logps/rejected": -446.5083923339844,
+      "loss": 0.5346,
+      "rewards/accuracies": 0.737500011920929,
+      "rewards/chosen": -1.3028686046600342,
+      "rewards/margins": 0.8693955540657043,
+      "rewards/rejected": -2.172264575958252,
+      "step": 550
+    },
+    {
+      "epoch": 0.5860805860805861,
+      "grad_norm": 58.17960656157151,
+      "learning_rate": 2.185397007170141e-07,
+      "logits/chosen": 0.4589771628379822,
+      "logits/rejected": 1.423572301864624,
+      "logps/chosen": -409.08038330078125,
+      "logps/rejected": -472.00299072265625,
+      "loss": 0.5248,
+      "rewards/accuracies": 0.7562500238418579,
+      "rewards/chosen": -1.262632131576538,
+      "rewards/margins": 0.9083721041679382,
+      "rewards/rejected": -2.171004056930542,
+      "step": 560
+    },
+    {
+      "epoch": 0.5965463108320251,
+      "grad_norm": 37.58413657059095,
+      "learning_rate": 2.094922764865619e-07,
+      "logits/chosen": 0.26942554116249084,
+      "logits/rejected": 1.2135722637176514,
+      "logps/chosen": -423.8333435058594,
+      "logps/rejected": -494.722412109375,
+      "loss": 0.5078,
+      "rewards/accuracies": 0.762499988079071,
+      "rewards/chosen": -1.44338858127594,
+      "rewards/margins": 0.8417580723762512,
+      "rewards/rejected": -2.285146474838257,
+      "step": 570
+    },
+    {
+      "epoch": 0.6070120355834642,
+      "grad_norm": 37.085175613726506,
+      "learning_rate": 2.0049902774588797e-07,
+      "logits/chosen": 0.0792941227555275,
+      "logits/rejected": 0.9457789659500122,
+      "logps/chosen": -383.8653259277344,
+      "logps/rejected": -429.4107971191406,
+      "loss": 0.5141,
+      "rewards/accuracies": 0.768750011920929,
+      "rewards/chosen": -1.3052608966827393,
+      "rewards/margins": 0.8605607748031616,
+      "rewards/rejected": -2.1658215522766113,
+      "step": 580
+    },
+    {
+      "epoch": 0.6174777603349032,
+      "grad_norm": 36.23072166325168,
+      "learning_rate": 1.9157198216806238e-07,
+      "logits/chosen": -0.4192202687263489,
+      "logits/rejected": 0.5567474961280823,
+      "logps/chosen": -407.47918701171875,
+      "logps/rejected": -483.21771240234375,
+      "loss": 0.4922,
+      "rewards/accuracies": 0.6875,
+      "rewards/chosen": -1.2952667474746704,
+      "rewards/margins": 0.8493944406509399,
+      "rewards/rejected": -2.1446611881256104,
+      "step": 590
+    },
+    {
+      "epoch": 0.6279434850863422,
+      "grad_norm": 41.952323384763105,
+      "learning_rate": 1.8272307888529274e-07,
+      "logits/chosen": 0.15778926014900208,
+      "logits/rejected": 0.9704158902168274,
+      "logps/chosen": -409.7381286621094,
+      "logps/rejected": -479.462158203125,
+      "loss": 0.4678,
+      "rewards/accuracies": 0.7749999761581421,
+      "rewards/chosen": -1.2721447944641113,
+      "rewards/margins": 0.9216303825378418,
+      "rewards/rejected": -2.193774938583374,
+      "step": 600
+    },
+    {
+      "epoch": 0.6279434850863422,
+      "eval_logits/chosen": -0.1275799423456192,
+      "eval_logits/rejected": 0.7528029680252075,
+      "eval_logps/chosen": -405.3749694824219,
+      "eval_logps/rejected": -480.077392578125,
+      "eval_loss": 0.4995136857032776,
+      "eval_rewards/accuracies": 0.7698412537574768,
+      "eval_rewards/chosen": -1.234070062637329,
+      "eval_rewards/margins": 0.9646352529525757,
+      "eval_rewards/rejected": -2.1987051963806152,
+      "eval_runtime": 188.2618,
+      "eval_samples_per_second": 10.624,
+      "eval_steps_per_second": 0.335,
+      "step": 600
+    },
+    {
+      "epoch": 0.6384092098377813,
+      "grad_norm": 36.8980365038612,
+      "learning_rate": 1.7396415252139288e-07,
+      "logits/chosen": -0.17294003069400787,
+      "logits/rejected": 1.0471775531768799,
+      "logps/chosen": -429.52301025390625,
+      "logps/rejected": -468.57666015625,
+      "loss": 0.504,
+      "rewards/accuracies": 0.768750011920929,
+      "rewards/chosen": -1.2395267486572266,
+      "rewards/margins": 0.9991617202758789,
+      "rewards/rejected": -2.2386884689331055,
+      "step": 610
+    },
+    {
+      "epoch": 0.6488749345892203,
+      "grad_norm": 37.2641783147775,
+      "learning_rate": 1.6530691736402316e-07,
+      "logits/chosen": 0.1285095065832138,
+      "logits/rejected": 0.9715364575386047,
+      "logps/chosen": -385.4737243652344,
+      "logps/rejected": -448.80560302734375,
+      "loss": 0.4733,
+      "rewards/accuracies": 0.7749999761581421,
+      "rewards/chosen": -1.1445059776306152,
+      "rewards/margins": 0.9794843792915344,
+      "rewards/rejected": -2.123990297317505,
+      "step": 620
+    },
+    {
+      "epoch": 0.6593406593406593,
+      "grad_norm": 60.096239521040786,
+      "learning_rate": 1.5676295169786864e-07,
+      "logits/chosen": 0.053101230412721634,
+      "logits/rejected": 1.1824990510940552,
+      "logps/chosen": -383.7541809082031,
+      "logps/rejected": -444.8204650878906,
+      "loss": 0.4758,
+      "rewards/accuracies": 0.800000011920929,
+      "rewards/chosen": -1.0906901359558105,
+      "rewards/margins": 0.9977191686630249,
+      "rewards/rejected": -2.088409423828125,
+      "step": 630
+    },
+    {
+      "epoch": 0.6698063840920984,
+      "grad_norm": 46.96867253083772,
+      "learning_rate": 1.483436823197092e-07,
+      "logits/chosen": 0.7413724064826965,
+      "logits/rejected": 1.8815301656723022,
+      "logps/chosen": -407.17962646484375,
+      "logps/rejected": -458.6732482910156,
+      "loss": 0.4872,
+      "rewards/accuracies": 0.7875000238418579,
+      "rewards/chosen": -1.401586651802063,
+      "rewards/margins": 0.8984085321426392,
+      "rewards/rejected": -2.299994945526123,
+      "step": 640
+    },
+    {
+      "epoch": 0.6802721088435374,
+      "grad_norm": 42.88319624739259,
+      "learning_rate": 1.4006036925609243e-07,
+      "logits/chosen": 0.10141663253307343,
+      "logits/rejected": 0.8694413900375366,
+      "logps/chosen": -378.79437255859375,
+      "logps/rejected": -466.408935546875,
+      "loss": 0.4933,
+      "rewards/accuracies": 0.7124999761581421,
+      "rewards/chosen": -1.1111141443252563,
+      "rewards/margins": 0.8358948826789856,
+      "rewards/rejected": -1.9470090866088867,
+      "step": 650
+    },
+    {
+      "epoch": 0.6907378335949764,
+      "grad_norm": 51.88763965072434,
+      "learning_rate": 1.319240907040458e-07,
+      "logits/chosen": 0.04605213552713394,
+      "logits/rejected": 1.151568055152893,
+      "logps/chosen": -388.32391357421875,
+      "logps/rejected": -469.0081481933594,
+      "loss": 0.4886,
+      "rewards/accuracies": 0.7562500238418579,
+      "rewards/chosen": -1.061607003211975,
+      "rewards/margins": 1.0539276599884033,
+      "rewards/rejected": -2.115534782409668,
+      "step": 660
+    },
+    {
+      "epoch": 0.7012035583464155,
+      "grad_norm": 55.401157791020196,
+      "learning_rate": 1.239457282149695e-07,
+      "logits/chosen": -0.006102992687374353,
+      "logits/rejected": 0.8945139050483704,
+      "logps/chosen": -399.7294616699219,
+      "logps/rejected": -500.3643493652344,
+      "loss": 0.4712,
+      "rewards/accuracies": 0.768750011920929,
+      "rewards/chosen": -1.1118131875991821,
+      "rewards/margins": 1.06088387966156,
+      "rewards/rejected": -2.172697067260742,
+      "step": 670
+    },
+    {
+      "epoch": 0.7116692830978545,
+      "grad_norm": 40.22762583896388,
+      "learning_rate": 1.1613595214152711e-07,
+      "logits/chosen": 0.2633221745491028,
+      "logits/rejected": 1.1040217876434326,
+      "logps/chosen": -423.1806640625,
+      "logps/rejected": -509.38189697265625,
+      "loss": 0.5052,
+      "rewards/accuracies": 0.768750011920929,
+      "rewards/chosen": -1.1152422428131104,
+      "rewards/margins": 1.0996170043945312,
+      "rewards/rejected": -2.2148590087890625,
+      "step": 680
+    },
+    {
+      "epoch": 0.7221350078492935,
+      "grad_norm": 30.575855567467922,
+      "learning_rate": 1.0850520736699362e-07,
+      "logits/chosen": 0.04858311265707016,
+      "logits/rejected": 1.2780332565307617,
+      "logps/chosen": -390.74359130859375,
+      "logps/rejected": -458.5567932128906,
+      "loss": 0.4558,
+      "rewards/accuracies": 0.793749988079071,
+      "rewards/chosen": -1.2008898258209229,
+      "rewards/margins": 1.0576915740966797,
+      "rewards/rejected": -2.2585813999176025,
+      "step": 690
+    },
+    {
+      "epoch": 0.7326007326007326,
+      "grad_norm": 34.53526618452247,
+      "learning_rate": 1.0106369933615042e-07,
+      "logits/chosen": 0.4477645456790924,
+      "logits/rejected": 1.1958597898483276,
+      "logps/chosen": -393.6428527832031,
+      "logps/rejected": -463.24969482421875,
+      "loss": 0.4767,
+      "rewards/accuracies": 0.7562500238418579,
+      "rewards/chosen": -1.3954741954803467,
+      "rewards/margins": 0.8515383005142212,
+      "rewards/rejected": -2.2470126152038574,
+      "step": 700
+    },
+    {
+      "epoch": 0.7326007326007326,
+      "eval_logits/chosen": 0.3566991090774536,
+      "eval_logits/rejected": 1.195650577545166,
+      "eval_logps/chosen": -416.3518371582031,
+      "eval_logps/rejected": -496.01458740234375,
+      "eval_loss": 0.49741485714912415,
+      "eval_rewards/accuracies": 0.761904776096344,
+      "eval_rewards/chosen": -1.3438389301300049,
+      "eval_rewards/margins": 1.014238715171814,
+      "eval_rewards/rejected": -2.3580777645111084,
+      "eval_runtime": 187.7233,
+      "eval_samples_per_second": 10.654,
+      "eval_steps_per_second": 0.336,
+      "step": 700
+    },
+    {
+      "epoch": 0.7430664573521716,
+      "grad_norm": 45.7676677868217,
+      "learning_rate": 9.382138040640714e-08,
+      "logits/chosen": 0.588351845741272,
+      "logits/rejected": 1.5306826829910278,
+      "logps/chosen": -412.3795471191406,
+      "logps/rejected": -485.49188232421875,
+      "loss": 0.5374,
+      "rewards/accuracies": 0.7437499761581421,
+      "rewards/chosen": -1.5390466451644897,
+      "rewards/margins": 0.8864930272102356,
+      "rewards/rejected": -2.42553973197937,
+      "step": 710
+    },
+    {
+      "epoch": 0.7535321821036107,
+      "grad_norm": 36.760991249638835,
+      "learning_rate": 8.678793653740632e-08,
+      "logits/chosen": 0.5254176259040833,
+      "logits/rejected": 1.342352032661438,
+      "logps/chosen": -417.776611328125,
+      "logps/rejected": -491.72088623046875,
+      "loss": 0.4792,
+      "rewards/accuracies": 0.768750011920929,
+      "rewards/chosen": -1.340393304824829,
+      "rewards/margins": 1.1020629405975342,
+      "rewards/rejected": -2.442456007003784,
+      "step": 720
+    },
+    {
+      "epoch": 0.7639979068550498,
+      "grad_norm": 39.73391988291906,
+      "learning_rate": 7.997277433690983e-08,
+      "logits/chosen": 0.5563046932220459,
+      "logits/rejected": 1.2839845418930054,
+      "logps/chosen": -424.78240966796875,
+      "logps/rejected": -473.6419372558594,
+      "loss": 0.4991,
+      "rewards/accuracies": 0.737500011920929,
+      "rewards/chosen": -1.5942018032073975,
+      "rewards/margins": 0.7723394632339478,
+      "rewards/rejected": -2.3665413856506348,
+      "step": 730
+    },
+    {
+      "epoch": 0.7744636316064888,
+      "grad_norm": 51.17374712345499,
+      "learning_rate": 7.338500848029602e-08,
+      "logits/chosen": 0.4739949107170105,
+      "logits/rejected": 1.098067045211792,
+      "logps/chosen": -395.0653076171875,
+      "logps/rejected": -454.8160705566406,
+      "loss": 0.5093,
+      "rewards/accuracies": 0.7124999761581421,
+      "rewards/chosen": -1.4219887256622314,
+      "rewards/margins": 0.8037675619125366,
+      "rewards/rejected": -2.2257564067840576,
+      "step": 740
+    },
+    {
+      "epoch": 0.7849293563579278,
+      "grad_norm": 35.40234763834154,
+      "learning_rate": 6.70334495204884e-08,
+      "logits/chosen": 0.38877877593040466,
+      "logits/rejected": 1.059231162071228,
+      "logps/chosen": -398.44549560546875,
+      "logps/rejected": -491.61737060546875,
+      "loss": 0.4763,
+      "rewards/accuracies": 0.7124999761581421,
+      "rewards/chosen": -1.5158002376556396,
+      "rewards/margins": 0.9638466835021973,
+      "rewards/rejected": -2.479646921157837,
+      "step": 750
+    },
+    {
+      "epoch": 0.7953950811093669,
+      "grad_norm": 32.10738245200245,
+      "learning_rate": 6.092659210462231e-08,
+      "logits/chosen": 0.5303295850753784,
+      "logits/rejected": 1.1202110052108765,
+      "logps/chosen": -394.3402404785156,
+      "logps/rejected": -486.60028076171875,
+      "loss": 0.5102,
+      "rewards/accuracies": 0.7437499761581421,
+      "rewards/chosen": -1.5299334526062012,
+      "rewards/margins": 0.9059597849845886,
+      "rewards/rejected": -2.4358930587768555,
+      "step": 760
+    },
+    {
+      "epoch": 0.8058608058608059,
+      "grad_norm": 27.247225736152675,
+      "learning_rate": 5.507260361320737e-08,
+      "logits/chosen": 0.18302218616008759,
+      "logits/rejected": 0.5413929224014282,
+      "logps/chosen": -445.942138671875,
+      "logps/rejected": -551.6734008789062,
+      "loss": 0.464,
+      "rewards/accuracies": 0.7562500238418579,
+      "rewards/chosen": -1.4259153604507446,
+      "rewards/margins": 0.8710676431655884,
+      "rewards/rejected": -2.296983242034912,
+      "step": 770
+    },
+    {
+      "epoch": 0.8163265306122449,
+      "grad_norm": 45.84319205475222,
+      "learning_rate": 4.947931323697982e-08,
+      "logits/chosen": 0.2840246260166168,
+      "logits/rejected": 0.9958732724189758,
+      "logps/chosen": -458.2361755371094,
+      "logps/rejected": -497.19073486328125,
+      "loss": 0.5094,
+      "rewards/accuracies": 0.731249988079071,
+      "rewards/chosen": -1.3927518129348755,
+      "rewards/margins": 0.9035183787345886,
+      "rewards/rejected": -2.2962703704833984,
+      "step": 780
+    },
+    {
+      "epoch": 0.826792255363684,
+      "grad_norm": 44.089647753755735,
+      "learning_rate": 4.415420150605398e-08,
+      "logits/chosen": 0.5449953079223633,
+      "logits/rejected": 1.7322685718536377,
+      "logps/chosen": -432.45721435546875,
+      "logps/rejected": -500.68536376953125,
+      "loss": 0.5157,
+      "rewards/accuracies": 0.762499988079071,
+      "rewards/chosen": -1.5388188362121582,
+      "rewards/margins": 1.0153836011886597,
+      "rewards/rejected": -2.5542023181915283,
+      "step": 790
+    },
+    {
+      "epoch": 0.837257980115123,
+      "grad_norm": 34.70483028107009,
+      "learning_rate": 3.9104390285376374e-08,
+      "logits/chosen": 0.6709474325180054,
+      "logits/rejected": 1.429652452468872,
+      "logps/chosen": -424.94525146484375,
+      "logps/rejected": -514.93359375,
+      "loss": 0.475,
+      "rewards/accuracies": 0.768750011920929,
+      "rewards/chosen": -1.588392972946167,
+      "rewards/margins": 0.9277172088623047,
+      "rewards/rejected": -2.5161101818084717,
+      "step": 800
+    },
+    {
+      "epoch": 0.837257980115123,
+      "eval_logits/chosen": 0.2101772576570511,
+      "eval_logits/rejected": 1.0101326704025269,
+      "eval_logps/chosen": -421.8183898925781,
+      "eval_logps/rejected": -499.10052490234375,
+      "eval_loss": 0.49712884426116943,
+      "eval_rewards/accuracies": 0.7638888955116272,
+      "eval_rewards/chosen": -1.3985040187835693,
+      "eval_rewards/margins": 0.9904327392578125,
+      "eval_rewards/rejected": -2.388936758041382,
+      "eval_runtime": 187.4518,
+      "eval_samples_per_second": 10.669,
+      "eval_steps_per_second": 0.336,
+      "step": 800
+    },
+    {
+      "epoch": 0.847723704866562,
+      "grad_norm": 177.69409876949015,
+      "learning_rate": 3.433663324986208e-08,
+      "logits/chosen": 0.23488643765449524,
+      "logits/rejected": 1.2444018125534058,
+      "logps/chosen": -428.610595703125,
+      "logps/rejected": -474.4007263183594,
+      "loss": 0.5007,
+      "rewards/accuracies": 0.7749999761581421,
+      "rewards/chosen": -1.373721718788147,
+      "rewards/margins": 0.9057692289352417,
+      "rewards/rejected": -2.2794909477233887,
+      "step": 810
+    },
+    {
+      "epoch": 0.858189429618001,
+      "grad_norm": 34.88776442194179,
+      "learning_rate": 2.9857306851953897e-08,
+      "logits/chosen": 0.18579676747322083,
+      "logits/rejected": 1.0101568698883057,
+      "logps/chosen": -445.45135498046875,
+      "logps/rejected": -517.8641357421875,
+      "loss": 0.4978,
+      "rewards/accuracies": 0.7437499761581421,
+      "rewards/chosen": -1.4067695140838623,
+      "rewards/margins": 0.9408528208732605,
+      "rewards/rejected": -2.3476223945617676,
+      "step": 820
+    },
+    {
+      "epoch": 0.8686551543694401,
+      "grad_norm": 38.51135078969842,
+      "learning_rate": 2.567240179368185e-08,
+      "logits/chosen": 0.36057838797569275,
+      "logits/rejected": 1.117890477180481,
+      "logps/chosen": -402.0654602050781,
+      "logps/rejected": -492.4375915527344,
+      "loss": 0.5064,
+      "rewards/accuracies": 0.7124999761581421,
+      "rewards/chosen": -1.439205288887024,
+      "rewards/margins": 0.9872032999992371,
+      "rewards/rejected": -2.4264087677001953,
+      "step": 830
+    },
+    {
+      "epoch": 0.8791208791208791,
+      "grad_norm": 36.67834755885448,
+      "learning_rate": 2.1787515014630357e-08,
+      "logits/chosen": 0.33748364448547363,
+      "logits/rejected": 1.2341878414154053,
+      "logps/chosen": -429.6908264160156,
+      "logps/rejected": -508.3697204589844,
+      "loss": 0.4633,
+      "rewards/accuracies": 0.7437499761581421,
+      "rewards/chosen": -1.4741687774658203,
+      "rewards/margins": 0.9488083124160767,
+      "rewards/rejected": -2.4229774475097656,
+      "step": 840
+    },
+    {
+      "epoch": 0.8895866038723181,
+      "grad_norm": 43.50042104660291,
+      "learning_rate": 1.820784220652766e-08,
+      "logits/chosen": 0.6195130944252014,
+      "logits/rejected": 1.6208534240722656,
+      "logps/chosen": -424.727294921875,
+      "logps/rejected": -481.82470703125,
+      "loss": 0.4617,
+      "rewards/accuracies": 0.768750011920929,
+      "rewards/chosen": -1.4389230012893677,
+      "rewards/margins": 1.0447529554367065,
+      "rewards/rejected": -2.483675956726074,
+      "step": 850
+    },
+    {
+      "epoch": 0.9000523286237572,
+      "grad_norm": 39.02433201892746,
+      "learning_rate": 1.4938170864468636e-08,
+      "logits/chosen": 0.5546199679374695,
+      "logits/rejected": 1.5712369680404663,
+      "logps/chosen": -434.83526611328125,
+      "logps/rejected": -505.6844787597656,
+      "loss": 0.4839,
+      "rewards/accuracies": 0.7562500238418579,
+      "rewards/chosen": -1.4654333591461182,
+      "rewards/margins": 0.9778777360916138,
+      "rewards/rejected": -2.4433112144470215,
+      "step": 860
+    },
+    {
+      "epoch": 0.9105180533751962,
+      "grad_norm": 59.667427453114435,
+      "learning_rate": 1.1982873884064465e-08,
+      "logits/chosen": 0.2385740727186203,
+      "logits/rejected": 1.2312796115875244,
+      "logps/chosen": -410.1748046875,
+      "logps/rejected": -481.6756286621094,
+      "loss": 0.4943,
+      "rewards/accuracies": 0.737500011920929,
+      "rewards/chosen": -1.3840436935424805,
+      "rewards/margins": 0.967445969581604,
+      "rewards/rejected": -2.351489543914795,
+      "step": 870
+    },
+    {
+      "epoch": 0.9209837781266352,
+      "grad_norm": 34.34973812097979,
+      "learning_rate": 9.345903713082304e-09,
+      "logits/chosen": 0.22545938193798065,
+      "logits/rejected": 1.0729210376739502,
+      "logps/chosen": -431.2491149902344,
+      "logps/rejected": -526.138427734375,
+      "loss": 0.4804,
+      "rewards/accuracies": 0.768750011920929,
+      "rewards/chosen": -1.2711529731750488,
+      "rewards/margins": 1.043867826461792,
+      "rewards/rejected": -2.3150203227996826,
+      "step": 880
+    },
+    {
+      "epoch": 0.9314495028780743,
+      "grad_norm": 31.959454214655377,
+      "learning_rate": 7.030787065396865e-09,
+      "logits/chosen": 0.2903442978858948,
+      "logits/rejected": 1.0575335025787354,
+      "logps/chosen": -420.6573181152344,
+      "logps/rejected": -513.2061767578125,
+      "loss": 0.4855,
+      "rewards/accuracies": 0.737500011920929,
+      "rewards/chosen": -1.3298420906066895,
+      "rewards/margins": 1.0002529621124268,
+      "rewards/rejected": -2.330094814300537,
+      "step": 890
+    },
+    {
+      "epoch": 0.9419152276295133,
+      "grad_norm": 47.23727995191594,
+      "learning_rate": 5.04062020432286e-09,
+      "logits/chosen": 0.19465875625610352,
+      "logits/rejected": 0.8790876269340515,
+      "logps/chosen": -413.4163513183594,
+      "logps/rejected": -506.9383850097656,
+      "loss": 0.4828,
+      "rewards/accuracies": 0.762499988079071,
+      "rewards/chosen": -1.3302490711212158,
+      "rewards/margins": 0.9251013994216919,
+      "rewards/rejected": -2.2553505897521973,
+      "step": 900
+    },
+    {
+      "epoch": 0.9419152276295133,
+      "eval_logits/chosen": 0.251442551612854,
+      "eval_logits/rejected": 1.1202069520950317,
+      "eval_logps/chosen": -418.4964599609375,
+      "eval_logps/rejected": -499.6516418457031,
+      "eval_loss": 0.4963185489177704,
+      "eval_rewards/accuracies": 0.7638888955116272,
+      "eval_rewards/chosen": -1.3652852773666382,
+      "eval_rewards/margins": 1.0291632413864136,
+      "eval_rewards/rejected": -2.394448757171631,
+      "eval_runtime": 187.5429,
+      "eval_samples_per_second": 10.664,
+      "eval_steps_per_second": 0.336,
+      "step": 900
+    },
+    {
+      "epoch": 0.9523809523809523,
+      "grad_norm": 57.54106835407393,
+      "learning_rate": 3.3780648016376866e-09,
+      "logits/chosen": 0.4812156558036804,
+      "logits/rejected": 1.2290699481964111,
+      "logps/chosen": -363.21429443359375,
+      "logps/rejected": -442.1210021972656,
+      "loss": 0.4842,
+      "rewards/accuracies": 0.7749999761581421,
+      "rewards/chosen": -1.3979285955429077,
+      "rewards/margins": 0.894494891166687,
+      "rewards/rejected": -2.2924234867095947,
+      "step": 910
+    },
+    {
+      "epoch": 0.9628466771323915,
+      "grad_norm": 46.08332660505321,
+      "learning_rate": 2.0453443778310766e-09,
+      "logits/chosen": 0.28524526953697205,
+      "logits/rejected": 1.2063854932785034,
+      "logps/chosen": -419.14453125,
+      "logps/rejected": -487.4471740722656,
+      "loss": 0.4991,
+      "rewards/accuracies": 0.7562500238418579,
+      "rewards/chosen": -1.3250755071640015,
+      "rewards/margins": 1.0333658456802368,
+      "rewards/rejected": -2.358441114425659,
+      "step": 920
+    },
+    {
+      "epoch": 0.9733124018838305,
+      "grad_norm": 43.54975464877693,
+      "learning_rate": 1.0442413283435758e-09,
+      "logits/chosen": 0.056365132331848145,
+      "logits/rejected": 1.318570852279663,
+      "logps/chosen": -439.2395935058594,
+      "logps/rejected": -490.8741760253906,
+      "loss": 0.488,
+      "rewards/accuracies": 0.7749999761581421,
+      "rewards/chosen": -1.2440545558929443,
+      "rewards/margins": 1.0225216150283813,
+      "rewards/rejected": -2.2665762901306152,
+      "step": 930
+    },
+    {
+      "epoch": 0.9837781266352695,
+      "grad_norm": 42.097145146923,
+      "learning_rate": 3.760945397705828e-10,
+      "logits/chosen": 0.23059086501598358,
+      "logits/rejected": 1.1359683275222778,
+      "logps/chosen": -389.34002685546875,
+      "logps/rejected": -485.71234130859375,
+      "loss": 0.4755,
+      "rewards/accuracies": 0.731249988079071,
+      "rewards/chosen": -1.2615879774093628,
+      "rewards/margins": 0.9983237981796265,
+      "rewards/rejected": -2.2599120140075684,
+      "step": 940
+    },
+    {
+      "epoch": 0.9942438513867086,
+      "grad_norm": 38.985040795872834,
+      "learning_rate": 4.17975992204056e-11,
+      "logits/chosen": 0.29284530878067017,
+      "logits/rejected": 0.8657740354537964,
+      "logps/chosen": -441.49786376953125,
+      "logps/rejected": -506.05804443359375,
+      "loss": 0.4913,
+      "rewards/accuracies": 0.706250011920929,
+      "rewards/chosen": -1.386546015739441,
+      "rewards/margins": 0.8242311477661133,
+      "rewards/rejected": -2.2107772827148438,
+      "step": 950
+    },
+    {
+      "epoch": 0.9994767137624281,
+      "step": 955,
+      "total_flos": 0.0,
+      "train_loss": 0.1815465917137905,
+      "train_runtime": 7518.522,
+      "train_samples_per_second": 8.131,
+      "train_steps_per_second": 0.127
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 955,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 1,
+  "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
+}