Model save

README.md

@@ -0,0 +1,79 @@
+---
+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 was trained from scratch on an unknown dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.5090
+- Rewards/chosen: -1.1007
+- Rewards/rejected: -2.0002
+- Rewards/accuracies: 0.7738
+- Rewards/margins: 0.8995
+- Logps/rejected: -466.1724
+- Logps/chosen: -401.8018
+- Logits/rejected: 3.6229
+- Logits/chosen: 2.8669
+
+## 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 |
+|:-------------:|:------:|:----:|:---------------:|:--------------:|:----------------:|:------------------:|:---------------:|:--------------:|:------------:|:---------------:|:-------------:|
+| 0.6512        | 0.1047 | 100  | 0.6511          | -0.0190        | -0.1266          | 0.6964             | 0.1076          | -278.8090      | -293.6257    | -2.3851         | -2.4490       |
+| 0.5992        | 0.2093 | 200  | 0.5944          | -0.2668        | -0.6535          | 0.7103             | 0.3866          | -331.5005      | -318.4129    | -1.7454         | -1.8605       |
+| 0.5469        | 0.3140 | 300  | 0.5530          | -0.6557        | -1.3199          | 0.7520             | 0.6642          | -398.1460      | -357.2993    | -0.7401         | -0.9693       |
+| 0.5491        | 0.4186 | 400  | 0.5448          | -1.0399        | -1.6860          | 0.7282             | 0.6462          | -434.7570      | -395.7156    | 1.3254          | 0.9052        |
+| 0.5351        | 0.5233 | 500  | 0.5296          | -0.8199        | -1.6144          | 0.7679             | 0.7945          | -427.5919      | -373.7142    | 2.7946          | 2.2107        |
+| 0.4879        | 0.6279 | 600  | 0.5152          | -1.1813        | -2.0574          | 0.7619             | 0.8761          | -471.8891      | -409.8589    | 3.3049          | 2.6265        |
+| 0.4963        | 0.7326 | 700  | 0.5121          | -1.1447        | -2.0602          | 0.7679             | 0.9156          | -472.1772      | -406.1937    | 3.7355          | 2.9642        |
+| 0.5009        | 0.8373 | 800  | 0.5099          | -1.1326        | -2.0244          | 0.7679             | 0.8919          | -468.5970      | -404.9855    | 3.6202          | 2.8807        |
+| 0.4926        | 0.9419 | 900  | 0.5090          | -1.1007        | -2.0002          | 0.7738             | 0.8995          | -466.1724      | -401.8018    | 3.6229          | 2.8669        |
+
+
+### Framework versions
+
+- Transformers 4.40.2
+- Pytorch 2.1.2
+- Datasets 2.19.1
+- Tokenizers 0.19.1

all_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 0.9994767137624281,
+    "total_flos": 0.0,
+    "train_loss": 0.5469793729133007,
+    "train_runtime": 17738.8932,
+    "train_samples": 61134,
+    "train_samples_per_second": 3.446,
+    "train_steps_per_second": 0.054
+}

generation_config.json

@@ -0,0 +1,6 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 1,
+  "eos_token_id": 2,
+  "transformers_version": "4.40.2"
+}

model-00001-of-00003.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f344a6895cd48cefdd8b6b436c50090b9003c765f6ed8cf5fa615dea27ded625
+size 4943162336

model-00002-of-00003.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:42e7fef3168b4a6b125704181b773886f590af7bc16f17232c42fd4460d93abc
+size 4999819336

model-00003-of-00003.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1534592415e97c08d74927426d8633c1b1a1d5fd1ac6ff27384b37311a6e2379
+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/May13_03-18-43_ubuntu/events.out.tfevents.1715570910.ubuntu.179147.0

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3752ef11e7cb43e4f97f8fa43ece532582b6fcedae29d68b44f90b5e234811b1
-size 73964
+oid sha256:904f05d14f30103147882f791fecb8bec4965e47aef1967f29693a0591915d05
+size 77758

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 0.9994767137624281,
+    "total_flos": 0.0,
+    "train_loss": 0.5469793729133007,
+    "train_runtime": 17738.8932,
+    "train_samples": 61134,
+    "train_samples_per_second": 3.446,
+    "train_steps_per_second": 0.054
+}

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": 13.049610594455547,
+      "learning_rate": 5.208333333333333e-09,
+      "logits/chosen": -2.8506698608398438,
+      "logits/rejected": -2.73948335647583,
+      "logps/chosen": -393.0640563964844,
+      "logps/rejected": -363.6795654296875,
+      "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": 9.923171226328847,
+      "learning_rate": 5.208333333333333e-08,
+      "logits/chosen": -2.4738845825195312,
+      "logits/rejected": -2.4443697929382324,
+      "logps/chosen": -266.9654235839844,
+      "logps/rejected": -241.94708251953125,
+      "loss": 0.6931,
+      "rewards/accuracies": 0.4236111044883728,
+      "rewards/chosen": -0.0002560871362220496,
+      "rewards/margins": 0.00017374774324707687,
+      "rewards/rejected": -0.00042983482126146555,
+      "step": 10
+    },
+    {
+      "epoch": 0.020931449502878074,
+      "grad_norm": 11.841313057253167,
+      "learning_rate": 1.0416666666666667e-07,
+      "logits/chosen": -2.478590488433838,
+      "logits/rejected": -2.4328413009643555,
+      "logps/chosen": -290.9645690917969,
+      "logps/rejected": -289.4146728515625,
+      "loss": 0.6931,
+      "rewards/accuracies": 0.5562499761581421,
+      "rewards/chosen": 0.00037191130104474723,
+      "rewards/margins": 0.0003235845942981541,
+      "rewards/rejected": 4.832679405808449e-05,
+      "step": 20
+    },
+    {
+      "epoch": 0.03139717425431711,
+      "grad_norm": 11.403660759662708,
+      "learning_rate": 1.5624999999999999e-07,
+      "logits/chosen": -2.5795135498046875,
+      "logits/rejected": -2.5510220527648926,
+      "logps/chosen": -280.20513916015625,
+      "logps/rejected": -281.74273681640625,
+      "loss": 0.6928,
+      "rewards/accuracies": 0.59375,
+      "rewards/chosen": 0.0011973328655585647,
+      "rewards/margins": 0.0008153930539265275,
+      "rewards/rejected": 0.00038193975342437625,
+      "step": 30
+    },
+    {
+      "epoch": 0.04186289900575615,
+      "grad_norm": 12.012171719363277,
+      "learning_rate": 2.0833333333333333e-07,
+      "logits/chosen": -2.540154457092285,
+      "logits/rejected": -2.4493587017059326,
+      "logps/chosen": -298.2313232421875,
+      "logps/rejected": -286.40972900390625,
+      "loss": 0.6917,
+      "rewards/accuracies": 0.606249988079071,
+      "rewards/chosen": 0.0023375265300273895,
+      "rewards/margins": 0.002678005024790764,
+      "rewards/rejected": -0.0003404785820748657,
+      "step": 40
+    },
+    {
+      "epoch": 0.052328623757195186,
+      "grad_norm": 11.0036408527525,
+      "learning_rate": 2.604166666666667e-07,
+      "logits/chosen": -2.558659553527832,
+      "logits/rejected": -2.4637410640716553,
+      "logps/chosen": -274.10552978515625,
+      "logps/rejected": -240.5741424560547,
+      "loss": 0.6899,
+      "rewards/accuracies": 0.668749988079071,
+      "rewards/chosen": 0.0027955034747719765,
+      "rewards/margins": 0.005770463030785322,
+      "rewards/rejected": -0.0029749595560133457,
+      "step": 50
+    },
+    {
+      "epoch": 0.06279434850863422,
+      "grad_norm": 11.03790571829773,
+      "learning_rate": 3.1249999999999997e-07,
+      "logits/chosen": -2.5305769443511963,
+      "logits/rejected": -2.487318754196167,
+      "logps/chosen": -310.541748046875,
+      "logps/rejected": -283.3028869628906,
+      "loss": 0.6859,
+      "rewards/accuracies": 0.65625,
+      "rewards/chosen": 0.007199519779533148,
+      "rewards/margins": 0.012489348649978638,
+      "rewards/rejected": -0.005289827939122915,
+      "step": 60
+    },
+    {
+      "epoch": 0.07326007326007326,
+      "grad_norm": 11.084374871096724,
+      "learning_rate": 3.645833333333333e-07,
+      "logits/chosen": -2.411536693572998,
+      "logits/rejected": -2.412984848022461,
+      "logps/chosen": -272.4049987792969,
+      "logps/rejected": -267.6209716796875,
+      "loss": 0.6811,
+      "rewards/accuracies": 0.706250011920929,
+      "rewards/chosen": -0.010773601941764355,
+      "rewards/margins": 0.029750987887382507,
+      "rewards/rejected": -0.04052459076046944,
+      "step": 70
+    },
+    {
+      "epoch": 0.0837257980115123,
+      "grad_norm": 11.255442825991683,
+      "learning_rate": 4.1666666666666667e-07,
+      "logits/chosen": -2.482091188430786,
+      "logits/rejected": -2.3982691764831543,
+      "logps/chosen": -288.2399597167969,
+      "logps/rejected": -264.4327087402344,
+      "loss": 0.6713,
+      "rewards/accuracies": 0.699999988079071,
+      "rewards/chosen": -0.031397901475429535,
+      "rewards/margins": 0.04932558909058571,
+      "rewards/rejected": -0.08072347939014435,
+      "step": 80
+    },
+    {
+      "epoch": 0.09419152276295134,
+      "grad_norm": 11.171512775493825,
+      "learning_rate": 4.6874999999999996e-07,
+      "logits/chosen": -2.407710313796997,
+      "logits/rejected": -2.388087749481201,
+      "logps/chosen": -270.56170654296875,
+      "logps/rejected": -281.1300354003906,
+      "loss": 0.6686,
+      "rewards/accuracies": 0.6499999761581421,
+      "rewards/chosen": -0.050538141280412674,
+      "rewards/margins": 0.035463497042655945,
+      "rewards/rejected": -0.08600164204835892,
+      "step": 90
+    },
+    {
+      "epoch": 0.10465724751439037,
+      "grad_norm": 28.327342502442335,
+      "learning_rate": 4.999732492681437e-07,
+      "logits/chosen": -2.399855375289917,
+      "logits/rejected": -2.334043025970459,
+      "logps/chosen": -312.9099426269531,
+      "logps/rejected": -298.2613525390625,
+      "loss": 0.6512,
+      "rewards/accuracies": 0.668749988079071,
+      "rewards/chosen": -0.03542739897966385,
+      "rewards/margins": 0.09241531044244766,
+      "rewards/rejected": -0.1278427243232727,
+      "step": 100
+    },
+    {
+      "epoch": 0.10465724751439037,
+      "eval_logits/chosen": -2.4489662647247314,
+      "eval_logits/rejected": -2.385138988494873,
+      "eval_logps/chosen": -293.625732421875,
+      "eval_logps/rejected": -278.8089904785156,
+      "eval_loss": 0.651090681552887,
+      "eval_rewards/accuracies": 0.6964285969734192,
+      "eval_rewards/chosen": -0.018974944949150085,
+      "eval_rewards/margins": 0.10759145021438599,
+      "eval_rewards/rejected": -0.12656639516353607,
+      "eval_runtime": 207.7302,
+      "eval_samples_per_second": 9.628,
+      "eval_steps_per_second": 0.303,
+      "step": 100
+    },
+    {
+      "epoch": 0.1151229722658294,
+      "grad_norm": 14.73263116699472,
+      "learning_rate": 4.996723692767926e-07,
+      "logits/chosen": -2.439448356628418,
+      "logits/rejected": -2.3935670852661133,
+      "logps/chosen": -252.96243286132812,
+      "logps/rejected": -260.0435485839844,
+      "loss": 0.6453,
+      "rewards/accuracies": 0.6812499761581421,
+      "rewards/chosen": -0.0579073540866375,
+      "rewards/margins": 0.11410091072320938,
+      "rewards/rejected": -0.17200824618339539,
+      "step": 110
+    },
+    {
+      "epoch": 0.12558869701726844,
+      "grad_norm": 23.560665335618697,
+      "learning_rate": 4.990375746213598e-07,
+      "logits/chosen": -2.34342360496521,
+      "logits/rejected": -2.283945083618164,
+      "logps/chosen": -295.8249206542969,
+      "logps/rejected": -336.9806823730469,
+      "loss": 0.6295,
+      "rewards/accuracies": 0.71875,
+      "rewards/chosen": -0.24184003472328186,
+      "rewards/margins": 0.17421653866767883,
+      "rewards/rejected": -0.4160565733909607,
+      "step": 120
+    },
+    {
+      "epoch": 0.1360544217687075,
+      "grad_norm": 18.74262713927055,
+      "learning_rate": 4.980697142834314e-07,
+      "logits/chosen": -2.247887134552002,
+      "logits/rejected": -2.222834825515747,
+      "logps/chosen": -315.74041748046875,
+      "logps/rejected": -314.04107666015625,
+      "loss": 0.6158,
+      "rewards/accuracies": 0.737500011920929,
+      "rewards/chosen": -0.25312143564224243,
+      "rewards/margins": 0.2810151278972626,
+      "rewards/rejected": -0.5341365933418274,
+      "step": 130
+    },
+    {
+      "epoch": 0.14652014652014653,
+      "grad_norm": 22.087464893229022,
+      "learning_rate": 4.967700826904229e-07,
+      "logits/chosen": -2.151130437850952,
+      "logits/rejected": -2.1017792224884033,
+      "logps/chosen": -298.3894958496094,
+      "logps/rejected": -315.7734069824219,
+      "loss": 0.6132,
+      "rewards/accuracies": 0.75,
+      "rewards/chosen": -0.03686642646789551,
+      "rewards/margins": 0.23762297630310059,
+      "rewards/rejected": -0.2744894027709961,
+      "step": 140
+    },
+    {
+      "epoch": 0.15698587127158556,
+      "grad_norm": 18.914759113825603,
+      "learning_rate": 4.951404179843962e-07,
+      "logits/chosen": -2.191765069961548,
+      "logits/rejected": -2.1456985473632812,
+      "logps/chosen": -327.501953125,
+      "logps/rejected": -307.2294616699219,
+      "loss": 0.6076,
+      "rewards/accuracies": 0.643750011920929,
+      "rewards/chosen": -0.27950796484947205,
+      "rewards/margins": 0.26142698526382446,
+      "rewards/rejected": -0.5409349203109741,
+      "step": 150
+    },
+    {
+      "epoch": 0.1674515960230246,
+      "grad_norm": 17.603963759838965,
+      "learning_rate": 4.931828996974498e-07,
+      "logits/chosen": -2.0961146354675293,
+      "logits/rejected": -1.9957778453826904,
+      "logps/chosen": -339.2377014160156,
+      "logps/rejected": -330.255126953125,
+      "loss": 0.5809,
+      "rewards/accuracies": 0.6812499761581421,
+      "rewards/chosen": -0.3309095501899719,
+      "rewards/margins": 0.370383083820343,
+      "rewards/rejected": -0.7012926340103149,
+      "step": 160
+    },
+    {
+      "epoch": 0.17791732077446362,
+      "grad_norm": 19.22776362116877,
+      "learning_rate": 4.909001458367866e-07,
+      "logits/chosen": -2.040045976638794,
+      "logits/rejected": -1.9285751581192017,
+      "logps/chosen": -332.156982421875,
+      "logps/rejected": -335.94482421875,
+      "loss": 0.5714,
+      "rewards/accuracies": 0.706250011920929,
+      "rewards/chosen": -0.46435314416885376,
+      "rewards/margins": 0.4518101215362549,
+      "rewards/rejected": -0.9161633253097534,
+      "step": 170
+    },
+    {
+      "epoch": 0.18838304552590268,
+      "grad_norm": 26.83242675712374,
+      "learning_rate": 4.882952093833627e-07,
+      "logits/chosen": -1.919015645980835,
+      "logits/rejected": -1.8708810806274414,
+      "logps/chosen": -305.6362609863281,
+      "logps/rejected": -353.7890930175781,
+      "loss": 0.5699,
+      "rewards/accuracies": 0.7124999761581421,
+      "rewards/chosen": -0.5798056721687317,
+      "rewards/margins": 0.4349302649497986,
+      "rewards/rejected": -1.0147359371185303,
+      "step": 180
+    },
+    {
+      "epoch": 0.1988487702773417,
+      "grad_norm": 25.701066458320277,
+      "learning_rate": 4.853715742087946e-07,
+      "logits/chosen": -1.67587149143219,
+      "logits/rejected": -1.6145784854888916,
+      "logps/chosen": -309.1290588378906,
+      "logps/rejected": -364.4269714355469,
+      "loss": 0.5592,
+      "rewards/accuracies": 0.731249988079071,
+      "rewards/chosen": -0.629682183265686,
+      "rewards/margins": 0.446587473154068,
+      "rewards/rejected": -1.0762696266174316,
+      "step": 190
+    },
+    {
+      "epoch": 0.20931449502878074,
+      "grad_norm": 20.15827071199051,
+      "learning_rate": 4.821331504159906e-07,
+      "logits/chosen": -1.937814712524414,
+      "logits/rejected": -1.7745281457901,
+      "logps/chosen": -350.3457946777344,
+      "logps/rejected": -320.7272033691406,
+      "loss": 0.5992,
+      "rewards/accuracies": 0.65625,
+      "rewards/chosen": -0.26695579290390015,
+      "rewards/margins": 0.3060823976993561,
+      "rewards/rejected": -0.5730382204055786,
+      "step": 200
+    },
+    {
+      "epoch": 0.20931449502878074,
+      "eval_logits/chosen": -1.8605066537857056,
+      "eval_logits/rejected": -1.7453688383102417,
+      "eval_logps/chosen": -318.41290283203125,
+      "eval_logps/rejected": -331.50048828125,
+      "eval_loss": 0.5943721532821655,
+      "eval_rewards/accuracies": 0.7103174328804016,
+      "eval_rewards/chosen": -0.2668464183807373,
+      "eval_rewards/margins": 0.38663458824157715,
+      "eval_rewards/rejected": -0.6534810066223145,
+      "eval_runtime": 206.6088,
+      "eval_samples_per_second": 9.68,
+      "eval_steps_per_second": 0.305,
+      "step": 200
+    },
+    {
+      "epoch": 0.21978021978021978,
+      "grad_norm": 24.1164776604431,
+      "learning_rate": 4.785842691097342e-07,
+      "logits/chosen": -1.7344303131103516,
+      "logits/rejected": -1.580756664276123,
+      "logps/chosen": -343.45013427734375,
+      "logps/rejected": -374.0958251953125,
+      "loss": 0.5853,
+      "rewards/accuracies": 0.6812499761581421,
+      "rewards/chosen": -0.5093246102333069,
+      "rewards/margins": 0.4380338191986084,
+      "rewards/rejected": -0.9473584890365601,
+      "step": 210
+    },
+    {
+      "epoch": 0.2302459445316588,
+      "grad_norm": 21.015327641628687,
+      "learning_rate": 4.7472967660421603e-07,
+      "logits/chosen": -1.6451599597930908,
+      "logits/rejected": -1.429702639579773,
+      "logps/chosen": -370.35992431640625,
+      "logps/rejected": -355.7476806640625,
+      "loss": 0.5794,
+      "rewards/accuracies": 0.706250011920929,
+      "rewards/chosen": -0.7406880855560303,
+      "rewards/margins": 0.42992156744003296,
+      "rewards/rejected": -1.170609712600708,
+      "step": 220
+    },
+    {
+      "epoch": 0.24071166928309787,
+      "grad_norm": 28.888863997061613,
+      "learning_rate": 4.705745280752585e-07,
+      "logits/chosen": -1.4397763013839722,
+      "logits/rejected": -1.367865800857544,
+      "logps/chosen": -367.10125732421875,
+      "logps/rejected": -389.14215087890625,
+      "loss": 0.5629,
+      "rewards/accuracies": 0.71875,
+      "rewards/chosen": -0.707022488117218,
+      "rewards/margins": 0.4359462261199951,
+      "rewards/rejected": -1.142968773841858,
+      "step": 230
+    },
+    {
+      "epoch": 0.25117739403453687,
+      "grad_norm": 26.673481537344585,
+      "learning_rate": 4.6612438066572555e-07,
+      "logits/chosen": -1.3668618202209473,
+      "logits/rejected": -1.2908679246902466,
+      "logps/chosen": -337.0715026855469,
+      "logps/rejected": -376.19598388671875,
+      "loss": 0.5489,
+      "rewards/accuracies": 0.75,
+      "rewards/chosen": -0.5933712720870972,
+      "rewards/margins": 0.5500311255455017,
+      "rewards/rejected": -1.1434024572372437,
+      "step": 240
+    },
+    {
+      "epoch": 0.2616431187859759,
+      "grad_norm": 22.197028615951425,
+      "learning_rate": 4.6138518605333664e-07,
+      "logits/chosen": -1.5141481161117554,
+      "logits/rejected": -1.4509350061416626,
+      "logps/chosen": -338.15631103515625,
+      "logps/rejected": -368.1983337402344,
+      "loss": 0.5386,
+      "rewards/accuracies": 0.737500011920929,
+      "rewards/chosen": -0.4901779592037201,
+      "rewards/margins": 0.5138848423957825,
+      "rewards/rejected": -1.0040627717971802,
+      "step": 250
+    },
+    {
+      "epoch": 0.272108843537415,
+      "grad_norm": 25.465027636854067,
+      "learning_rate": 4.5636328249082514e-07,
+      "logits/chosen": -1.5978645086288452,
+      "logits/rejected": -1.446722388267517,
+      "logps/chosen": -372.3781433105469,
+      "logps/rejected": -411.69073486328125,
+      "loss": 0.5477,
+      "rewards/accuracies": 0.6875,
+      "rewards/chosen": -0.7267775535583496,
+      "rewards/margins": 0.4993767738342285,
+      "rewards/rejected": -1.2261545658111572,
+      "step": 260
+    },
+    {
+      "epoch": 0.282574568288854,
+      "grad_norm": 29.229696217954388,
+      "learning_rate": 4.510653863290871e-07,
+      "logits/chosen": -1.4106498956680298,
+      "logits/rejected": -1.2901843786239624,
+      "logps/chosen": -367.2438049316406,
+      "logps/rejected": -374.22210693359375,
+      "loss": 0.5385,
+      "rewards/accuracies": 0.7437499761581421,
+      "rewards/chosen": -0.5454630851745605,
+      "rewards/margins": 0.5804157853126526,
+      "rewards/rejected": -1.1258786916732788,
+      "step": 270
+    },
+    {
+      "epoch": 0.29304029304029305,
+      "grad_norm": 23.994602412834077,
+      "learning_rate": 4.4549858303465737e-07,
+      "logits/chosen": -1.4280427694320679,
+      "logits/rejected": -1.3019177913665771,
+      "logps/chosen": -372.3103332519531,
+      "logps/rejected": -410.4972229003906,
+      "loss": 0.5667,
+      "rewards/accuracies": 0.7562500238418579,
+      "rewards/chosen": -0.6570708751678467,
+      "rewards/margins": 0.584852397441864,
+      "rewards/rejected": -1.2419233322143555,
+      "step": 280
+    },
+    {
+      "epoch": 0.3035060177917321,
+      "grad_norm": 34.15033045239415,
+      "learning_rate": 4.396703177135261e-07,
+      "logits/chosen": -1.376426339149475,
+      "logits/rejected": -1.093380093574524,
+      "logps/chosen": -349.5055847167969,
+      "logps/rejected": -368.0146484375,
+      "loss": 0.5476,
+      "rewards/accuracies": 0.7562500238418579,
+      "rewards/chosen": -0.6797803640365601,
+      "rewards/margins": 0.5791158080101013,
+      "rewards/rejected": -1.2588961124420166,
+      "step": 290
+    },
+    {
+      "epoch": 0.3139717425431711,
+      "grad_norm": 31.440651841522758,
+      "learning_rate": 4.335883851539693e-07,
+      "logits/chosen": -1.177045226097107,
+      "logits/rejected": -1.063737392425537,
+      "logps/chosen": -349.52752685546875,
+      "logps/rejected": -404.7928161621094,
+      "loss": 0.5469,
+      "rewards/accuracies": 0.6937500238418579,
+      "rewards/chosen": -0.7268069982528687,
+      "rewards/margins": 0.5689330697059631,
+      "rewards/rejected": -1.295740008354187,
+      "step": 300
+    },
+    {
+      "epoch": 0.3139717425431711,
+      "eval_logits/chosen": -0.9692657589912415,
+      "eval_logits/rejected": -0.7400917410850525,
+      "eval_logps/chosen": -357.29931640625,
+      "eval_logps/rejected": -398.1459655761719,
+      "eval_loss": 0.5529553294181824,
+      "eval_rewards/accuracies": 0.7519841194152832,
+      "eval_rewards/chosen": -0.6557106971740723,
+      "eval_rewards/margins": 0.6642255187034607,
+      "eval_rewards/rejected": -1.3199362754821777,
+      "eval_runtime": 206.9211,
+      "eval_samples_per_second": 9.666,
+      "eval_steps_per_second": 0.304,
+      "step": 300
+    },
+    {
+      "epoch": 0.32443746729461015,
+      "grad_norm": 19.948138397899992,
+      "learning_rate": 4.272609194017105e-07,
+      "logits/chosen": -1.28805673122406,
+      "logits/rejected": -0.8344380259513855,
+      "logps/chosen": -348.4212341308594,
+      "logps/rejected": -353.79974365234375,
+      "loss": 0.5427,
+      "rewards/accuracies": 0.6937500238418579,
+      "rewards/chosen": -0.5610443949699402,
+      "rewards/margins": 0.6562029123306274,
+      "rewards/rejected": -1.2172472476959229,
+      "step": 310
+    },
+    {
+      "epoch": 0.3349031920460492,
+      "grad_norm": 25.497519916223954,
+      "learning_rate": 4.2069638288135547e-07,
+      "logits/chosen": -1.0992823839187622,
+      "logits/rejected": -0.9369346499443054,
+      "logps/chosen": -325.60382080078125,
+      "logps/rejected": -384.43951416015625,
+      "loss": 0.5569,
+      "rewards/accuracies": 0.6937500238418579,
+      "rewards/chosen": -0.6635289192199707,
+      "rewards/margins": 0.6613875031471252,
+      "rewards/rejected": -1.3249164819717407,
+      "step": 320
+    },
+    {
+      "epoch": 0.3453689167974882,
+      "grad_norm": 29.20532289580398,
+      "learning_rate": 4.139035550786494e-07,
+      "logits/chosen": -0.9095026850700378,
+      "logits/rejected": -0.7446672320365906,
+      "logps/chosen": -370.08892822265625,
+      "logps/rejected": -396.7486572265625,
+      "loss": 0.5563,
+      "rewards/accuracies": 0.699999988079071,
+      "rewards/chosen": -1.0744504928588867,
+      "rewards/margins": 0.5352171659469604,
+      "rewards/rejected": -1.6096675395965576,
+      "step": 330
+    },
+    {
+      "epoch": 0.35583464154892724,
+      "grad_norm": 25.275145205759078,
+      "learning_rate": 4.0689152079869306e-07,
+      "logits/chosen": -0.9068937301635742,
+      "logits/rejected": -0.7064866423606873,
+      "logps/chosen": -366.93756103515625,
+      "logps/rejected": -388.6629638671875,
+      "loss": 0.5583,
+      "rewards/accuracies": 0.675000011920929,
+      "rewards/chosen": -0.8770108222961426,
+      "rewards/margins": 0.5930118560791016,
+      "rewards/rejected": -1.4700227975845337,
+      "step": 340
+    },
+    {
+      "epoch": 0.3663003663003663,
+      "grad_norm": 27.076055018325327,
+      "learning_rate": 3.99669658015821e-07,
+      "logits/chosen": -1.016298770904541,
+      "logits/rejected": -0.820306122303009,
+      "logps/chosen": -382.3629455566406,
+      "logps/rejected": -405.1012878417969,
+      "loss": 0.5308,
+      "rewards/accuracies": 0.706250011920929,
+      "rewards/chosen": -0.8123846054077148,
+      "rewards/margins": 0.5460414290428162,
+      "rewards/rejected": -1.3584258556365967,
+      "step": 350
+    },
+    {
+      "epoch": 0.37676609105180536,
+      "grad_norm": 31.551108556319388,
+      "learning_rate": 3.92247625331392e-07,
+      "logits/chosen": -0.6309953927993774,
+      "logits/rejected": -0.4506828188896179,
+      "logps/chosen": -375.9930725097656,
+      "logps/rejected": -412.9039611816406,
+      "loss": 0.5097,
+      "rewards/accuracies": 0.7562500238418579,
+      "rewards/chosen": -0.8602720499038696,
+      "rewards/margins": 0.6197329759597778,
+      "rewards/rejected": -1.4800050258636475,
+      "step": 360
+    },
+    {
+      "epoch": 0.3872318158032444,
+      "grad_norm": 29.323564206187836,
+      "learning_rate": 3.846353490562664e-07,
+      "logits/chosen": -0.02009066380560398,
+      "logits/rejected": 0.17469237744808197,
+      "logps/chosen": -373.5651550292969,
+      "logps/rejected": -395.3067932128906,
+      "loss": 0.5408,
+      "rewards/accuracies": 0.699999988079071,
+      "rewards/chosen": -1.021385669708252,
+      "rewards/margins": 0.5213597416877747,
+      "rewards/rejected": -1.5427453517913818,
+      "step": 370
+    },
+    {
+      "epoch": 0.3976975405546834,
+      "grad_norm": 36.01925145688383,
+      "learning_rate": 3.768430099352445e-07,
+      "logits/chosen": 0.2435513287782669,
+      "logits/rejected": 0.49736109375953674,
+      "logps/chosen": -330.96917724609375,
+      "logps/rejected": -392.32318115234375,
+      "loss": 0.5246,
+      "rewards/accuracies": 0.7437499761581421,
+      "rewards/chosen": -0.7018967866897583,
+      "rewards/margins": 0.6938203573226929,
+      "rewards/rejected": -1.3957171440124512,
+      "step": 380
+    },
+    {
+      "epoch": 0.40816326530612246,
+      "grad_norm": 33.52534414233946,
+      "learning_rate": 3.6888102953122304e-07,
+      "logits/chosen": 0.8603371381759644,
+      "logits/rejected": 1.1312789916992188,
+      "logps/chosen": -370.71270751953125,
+      "logps/rejected": -422.418212890625,
+      "loss": 0.5388,
+      "rewards/accuracies": 0.7250000238418579,
+      "rewards/chosen": -0.8173589706420898,
+      "rewards/margins": 0.6764534115791321,
+      "rewards/rejected": -1.4938123226165771,
+      "step": 390
+    },
+    {
+      "epoch": 0.4186289900575615,
+      "grad_norm": 44.423215808026434,
+      "learning_rate": 3.607600562872785e-07,
+      "logits/chosen": 1.0831114053726196,
+      "logits/rejected": 1.3025102615356445,
+      "logps/chosen": -386.6149597167969,
+      "logps/rejected": -434.16961669921875,
+      "loss": 0.5491,
+      "rewards/accuracies": 0.6812499761581421,
+      "rewards/chosen": -1.0485079288482666,
+      "rewards/margins": 0.5725308656692505,
+      "rewards/rejected": -1.621038794517517,
+      "step": 400
+    },
+    {
+      "epoch": 0.4186289900575615,
+      "eval_logits/chosen": 0.9051657915115356,
+      "eval_logits/rejected": 1.325432300567627,
+      "eval_logps/chosen": -395.71563720703125,
+      "eval_logps/rejected": -434.7570495605469,
+      "eval_loss": 0.5448315143585205,
+      "eval_rewards/accuracies": 0.72817462682724,
+      "eval_rewards/chosen": -1.0398739576339722,
+      "eval_rewards/margins": 0.6461727023124695,
+      "eval_rewards/rejected": -1.686046838760376,
+      "eval_runtime": 206.8887,
+      "eval_samples_per_second": 9.667,
+      "eval_steps_per_second": 0.305,
+      "step": 400
+    },
+    {
+      "epoch": 0.4290947148090005,
+      "grad_norm": 25.253356523152696,
+      "learning_rate": 3.5249095128531856e-07,
+      "logits/chosen": 1.432478666305542,
+      "logits/rejected": 1.8890857696533203,
+      "logps/chosen": -385.14434814453125,
+      "logps/rejected": -410.26593017578125,
+      "loss": 0.5508,
+      "rewards/accuracies": 0.6812499761581421,
+      "rewards/chosen": -1.040719985961914,
+      "rewards/margins": 0.5529226064682007,
+      "rewards/rejected": -1.5936425924301147,
+      "step": 410
+    },
+    {
+      "epoch": 0.43956043956043955,
+      "grad_norm": 27.063513600375742,
+      "learning_rate": 3.4408477372034736e-07,
+      "logits/chosen": 1.2875887155532837,
+      "logits/rejected": 2.005409002304077,
+      "logps/chosen": -367.0966796875,
+      "logps/rejected": -398.1778259277344,
+      "loss": 0.5497,
+      "rewards/accuracies": 0.7562500238418579,
+      "rewards/chosen": -0.7143039107322693,
+      "rewards/margins": 0.6772809028625488,
+      "rewards/rejected": -1.3915849924087524,
+      "step": 420
+    },
+    {
+      "epoch": 0.4500261643118786,
+      "grad_norm": 28.5484383688476,
+      "learning_rate": 3.3555276610977276e-07,
+      "logits/chosen": 1.905025839805603,
+      "logits/rejected": 2.3225185871124268,
+      "logps/chosen": -334.4276123046875,
+      "logps/rejected": -393.8622131347656,
+      "loss": 0.5318,
+      "rewards/accuracies": 0.7250000238418579,
+      "rewards/chosen": -0.7228254079818726,
+      "rewards/margins": 0.649430513381958,
+      "rewards/rejected": -1.3722560405731201,
+      "step": 430
+    },
+    {
+      "epoch": 0.4604918890633176,
+      "grad_norm": 39.797028152379326,
+      "learning_rate": 3.269063392575352e-07,
+      "logits/chosen": 1.8605855703353882,
+      "logits/rejected": 2.4145779609680176,
+      "logps/chosen": -393.1373596191406,
+      "logps/rejected": -411.34051513671875,
+      "loss": 0.5353,
+      "rewards/accuracies": 0.6812499761581421,
+      "rewards/chosen": -0.939368724822998,
+      "rewards/margins": 0.6202534437179565,
+      "rewards/rejected": -1.559622049331665,
+      "step": 440
+    },
+    {
+      "epoch": 0.47095761381475665,
+      "grad_norm": 33.99719296521047,
+      "learning_rate": 3.1815705699316964e-07,
+      "logits/chosen": 1.7213594913482666,
+      "logits/rejected": 2.1949357986450195,
+      "logps/chosen": -384.70172119140625,
+      "logps/rejected": -470.4630432128906,
+      "loss": 0.5331,
+      "rewards/accuracies": 0.793749988079071,
+      "rewards/chosen": -0.9140175580978394,
+      "rewards/margins": 0.9557299613952637,
+      "rewards/rejected": -1.869747519493103,
+      "step": 450
+    },
+    {
+      "epoch": 0.48142333856619574,
+      "grad_norm": 27.24462293822628,
+      "learning_rate": 3.0931662070620794e-07,
+      "logits/chosen": 1.7857542037963867,
+      "logits/rejected": 2.25050687789917,
+      "logps/chosen": -364.74761962890625,
+      "logps/rejected": -416.922119140625,
+      "loss": 0.5533,
+      "rewards/accuracies": 0.699999988079071,
+      "rewards/chosen": -0.9933540225028992,
+      "rewards/margins": 0.6881140470504761,
+      "rewards/rejected": -1.6814680099487305,
+      "step": 460
+    },
+    {
+      "epoch": 0.49188906331763477,
+      "grad_norm": 31.148955390461847,
+      "learning_rate": 3.003968536966078e-07,
+      "logits/chosen": 1.6684377193450928,
+      "logits/rejected": 2.1535468101501465,
+      "logps/chosen": -366.65692138671875,
+      "logps/rejected": -396.07806396484375,
+      "loss": 0.5004,
+      "rewards/accuracies": 0.731249988079071,
+      "rewards/chosen": -0.7513037919998169,
+      "rewards/margins": 0.6742973923683167,
+      "rewards/rejected": -1.4256012439727783,
+      "step": 470
+    },
+    {
+      "epoch": 0.5023547880690737,
+      "grad_norm": 39.54954979892467,
+      "learning_rate": 2.9140968536213693e-07,
+      "logits/chosen": 2.398738384246826,
+      "logits/rejected": 3.106950283050537,
+      "logps/chosen": -374.4889221191406,
+      "logps/rejected": -451.7330017089844,
+      "loss": 0.5195,
+      "rewards/accuracies": 0.800000011920929,
+      "rewards/chosen": -0.9869669675827026,
+      "rewards/margins": 0.9735086560249329,
+      "rewards/rejected": -1.9604755640029907,
+      "step": 480
+    },
+    {
+      "epoch": 0.5128205128205128,
+      "grad_norm": 36.77376973228987,
+      "learning_rate": 2.823671352438608e-07,
+      "logits/chosen": 2.0673065185546875,
+      "logits/rejected": 3.221296787261963,
+      "logps/chosen": -407.5098876953125,
+      "logps/rejected": -466.18096923828125,
+      "loss": 0.515,
+      "rewards/accuracies": 0.8062499761581421,
+      "rewards/chosen": -0.8568946123123169,
+      "rewards/margins": 1.1183595657348633,
+      "rewards/rejected": -1.9752540588378906,
+      "step": 490
+    },
+    {
+      "epoch": 0.5232862375719518,
+      "grad_norm": 28.11028467508396,
+      "learning_rate": 2.73281296951072e-07,
+      "logits/chosen": 2.630121946334839,
+      "logits/rejected": 3.05456805229187,
+      "logps/chosen": -347.7069396972656,
+      "logps/rejected": -379.8544921875,
+      "loss": 0.5351,
+      "rewards/accuracies": 0.7562500238418579,
+      "rewards/chosen": -0.8024020195007324,
+      "rewards/margins": 0.6900447607040405,
+      "rewards/rejected": -1.4924466609954834,
+      "step": 500
+    },
+    {
+      "epoch": 0.5232862375719518,
+      "eval_logits/chosen": 2.2106759548187256,
+      "eval_logits/rejected": 2.794649600982666,
+      "eval_logps/chosen": -373.7142333984375,
+      "eval_logps/rejected": -427.5918884277344,
+      "eval_loss": 0.5295986533164978,
+      "eval_rewards/accuracies": 0.7678571343421936,
+      "eval_rewards/chosen": -0.8198599219322205,
+      "eval_rewards/margins": 0.794535219669342,
+      "eval_rewards/rejected": -1.6143951416015625,
+      "eval_runtime": 206.5716,
+      "eval_samples_per_second": 9.682,
+      "eval_steps_per_second": 0.305,
+      "step": 500
+    },
+    {
+      "epoch": 0.533751962323391,
+      "grad_norm": 27.989874290246416,
+      "learning_rate": 2.641643219871597e-07,
+      "logits/chosen": 2.2483530044555664,
+      "logits/rejected": 2.7509026527404785,
+      "logps/chosen": -374.50189208984375,
+      "logps/rejected": -446.3634338378906,
+      "loss": 0.5095,
+      "rewards/accuracies": 0.75,
+      "rewards/chosen": -0.8130585551261902,
+      "rewards/margins": 0.952698826789856,
+      "rewards/rejected": -1.7657575607299805,
+      "step": 510
+    },
+    {
+      "epoch": 0.54421768707483,
+      "grad_norm": 29.655837468334063,
+      "learning_rate": 2.550284034980507e-07,
+      "logits/chosen": 2.823676586151123,
+      "logits/rejected": 3.116036891937256,
+      "logps/chosen": -435.7244567871094,
+      "logps/rejected": -478.9291076660156,
+      "loss": 0.5372,
+      "rewards/accuracies": 0.6875,
+      "rewards/chosen": -1.5566352605819702,
+      "rewards/margins": 0.6389855146408081,
+      "rewards/rejected": -2.1956207752227783,
+      "step": 520
+    },
+    {
+      "epoch": 0.554683411826269,
+      "grad_norm": 42.30907391512789,
+      "learning_rate": 2.4588575996495794e-07,
+      "logits/chosen": 2.5769808292388916,
+      "logits/rejected": 2.9770617485046387,
+      "logps/chosen": -409.3755187988281,
+      "logps/rejected": -474.4515686035156,
+      "loss": 0.5268,
+      "rewards/accuracies": 0.731249988079071,
+      "rewards/chosen": -1.3210904598236084,
+      "rewards/margins": 0.8030015826225281,
+      "rewards/rejected": -2.1240925788879395,
+      "step": 530
+    },
+    {
+      "epoch": 0.565149136577708,
+      "grad_norm": 23.65599619341314,
+      "learning_rate": 2.367486188632446e-07,
+      "logits/chosen": 2.1307921409606934,
+      "logits/rejected": 2.942972421646118,
+      "logps/chosen": -400.3769836425781,
+      "logps/rejected": -444.0709533691406,
+      "loss": 0.5275,
+      "rewards/accuracies": 0.706250011920929,
+      "rewards/chosen": -1.0123635530471802,
+      "rewards/margins": 0.7693914771080017,
+      "rewards/rejected": -1.7817550897598267,
+      "step": 540
+    },
+    {
+      "epoch": 0.5756148613291471,
+      "grad_norm": 33.373008010435754,
+      "learning_rate": 2.276292003092593e-07,
+      "logits/chosen": 2.142712354660034,
+      "logits/rejected": 3.0578479766845703,
+      "logps/chosen": -376.69659423828125,
+      "logps/rejected": -413.7811584472656,
+      "loss": 0.5301,
+      "rewards/accuracies": 0.71875,
+      "rewards/chosen": -0.9706202745437622,
+      "rewards/margins": 0.8356066942214966,
+      "rewards/rejected": -1.8062269687652588,
+      "step": 550
+    },
+    {
+      "epoch": 0.5860805860805861,
+      "grad_norm": 30.978416389391672,
+      "learning_rate": 2.185397007170141e-07,
+      "logits/chosen": 2.390106678009033,
+      "logits/rejected": 3.2527661323547363,
+      "logps/chosen": -385.70123291015625,
+      "logps/rejected": -440.04498291015625,
+      "loss": 0.5433,
+      "rewards/accuracies": 0.768750011920929,
+      "rewards/chosen": -0.9431241154670715,
+      "rewards/margins": 0.8373795747756958,
+      "rewards/rejected": -1.7805036306381226,
+      "step": 560
+    },
+    {
+      "epoch": 0.5965463108320251,
+      "grad_norm": 29.694372417345516,
+      "learning_rate": 2.094922764865619e-07,
+      "logits/chosen": 2.408522844314575,
+      "logits/rejected": 2.9558959007263184,
+      "logps/chosen": -408.5779724121094,
+      "logps/rejected": -465.5567321777344,
+      "loss": 0.5209,
+      "rewards/accuracies": 0.7749999761581421,
+      "rewards/chosen": -1.2187942266464233,
+      "rewards/margins": 0.6892539262771606,
+      "rewards/rejected": -1.9080482721328735,
+      "step": 570
+    },
+    {
+      "epoch": 0.6070120355834642,
+      "grad_norm": 28.00809419141954,
+      "learning_rate": 2.0049902774588797e-07,
+      "logits/chosen": 2.3852643966674805,
+      "logits/rejected": 3.023937940597534,
+      "logps/chosen": -366.2859802246094,
+      "logps/rejected": -395.6317443847656,
+      "loss": 0.5325,
+      "rewards/accuracies": 0.75,
+      "rewards/chosen": -1.0701501369476318,
+      "rewards/margins": 0.7390429377555847,
+      "rewards/rejected": -1.8091930150985718,
+      "step": 580
+    },
+    {
+      "epoch": 0.6174777603349032,
+      "grad_norm": 24.289299698883042,
+      "learning_rate": 1.9157198216806238e-07,
+      "logits/chosen": 2.0921032428741455,
+      "logits/rejected": 2.947533130645752,
+      "logps/chosen": -406.106201171875,
+      "logps/rejected": -468.72412109375,
+      "loss": 0.5016,
+      "rewards/accuracies": 0.7250000238418579,
+      "rewards/chosen": -1.1557496786117554,
+      "rewards/margins": 0.7965903282165527,
+      "rewards/rejected": -1.9523401260375977,
+      "step": 590
+    },
+    {
+      "epoch": 0.6279434850863422,
+      "grad_norm": 25.85769354594119,
+      "learning_rate": 1.8272307888529274e-07,
+      "logits/chosen": 3.03981351852417,
+      "logits/rejected": 3.630103588104248,
+      "logps/chosen": -407.9981384277344,
+      "logps/rejected": -463.118408203125,
+      "loss": 0.4879,
+      "rewards/accuracies": 0.75,
+      "rewards/chosen": -1.185763955116272,
+      "rewards/margins": 0.8175700306892395,
+      "rewards/rejected": -2.0033340454101562,
+      "step": 600
+    },
+    {
+      "epoch": 0.6279434850863422,
+      "eval_logits/chosen": 2.6265408992767334,
+      "eval_logits/rejected": 3.3048717975616455,
+      "eval_logps/chosen": -409.8588562011719,
+      "eval_logps/rejected": -471.8890686035156,
+      "eval_loss": 0.5151771306991577,
+      "eval_rewards/accuracies": 0.761904776096344,
+      "eval_rewards/chosen": -1.1813055276870728,
+      "eval_rewards/margins": 0.8760614395141602,
+      "eval_rewards/rejected": -2.0573670864105225,
+      "eval_runtime": 207.0197,
+      "eval_samples_per_second": 9.661,
+      "eval_steps_per_second": 0.304,
+      "step": 600
+    },
+    {
+      "epoch": 0.6384092098377813,
+      "grad_norm": 29.03204002521062,
+      "learning_rate": 1.7396415252139288e-07,
+      "logits/chosen": 2.516125440597534,
+      "logits/rejected": 3.6158394813537598,
+      "logps/chosen": -436.82586669921875,
+      "logps/rejected": -467.8284606933594,
+      "loss": 0.5229,
+      "rewards/accuracies": 0.8125,
+      "rewards/chosen": -1.1957874298095703,
+      "rewards/margins": 0.9723994135856628,
+      "rewards/rejected": -2.168186902999878,
+      "step": 610
+    },
+    {
+      "epoch": 0.6488749345892203,
+      "grad_norm": 33.79893234613174,
+      "learning_rate": 1.6530691736402316e-07,
+      "logits/chosen": 2.626211643218994,
+      "logits/rejected": 3.2447261810302734,
+      "logps/chosen": -382.3304443359375,
+      "logps/rejected": -431.347412109375,
+      "loss": 0.4952,
+      "rewards/accuracies": 0.7562500238418579,
+      "rewards/chosen": -1.0164293050765991,
+      "rewards/margins": 0.860329806804657,
+      "rewards/rejected": -1.8767591714859009,
+      "step": 620
+    },
+    {
+      "epoch": 0.6593406593406593,
+      "grad_norm": 34.859918532528575,
+      "learning_rate": 1.5676295169786864e-07,
+      "logits/chosen": 2.4743380546569824,
+      "logits/rejected": 3.4562110900878906,
+      "logps/chosen": -378.14453125,
+      "logps/rejected": -421.9088439941406,
+      "loss": 0.489,
+      "rewards/accuracies": 0.800000011920929,
+      "rewards/chosen": -0.9536962509155273,
+      "rewards/margins": 0.8669854402542114,
+      "rewards/rejected": -1.8206815719604492,
+      "step": 630
+    },
+    {
+      "epoch": 0.6698063840920984,
+      "grad_norm": 30.5082473824507,
+      "learning_rate": 1.483436823197092e-07,
+      "logits/chosen": 3.266901731491089,
+      "logits/rejected": 4.045676231384277,
+      "logps/chosen": -389.3726501464844,
+      "logps/rejected": -436.15130615234375,
+      "loss": 0.491,
+      "rewards/accuracies": 0.800000011920929,
+      "rewards/chosen": -1.1519488096237183,
+      "rewards/margins": 0.8454155921936035,
+      "rewards/rejected": -1.9973644018173218,
+      "step": 640
+    },
+    {
+      "epoch": 0.6802721088435374,
+      "grad_norm": 31.79264424904545,
+      "learning_rate": 1.4006036925609243e-07,
+      "logits/chosen": 2.823011875152588,
+      "logits/rejected": 3.2023251056671143,
+      "logps/chosen": -369.93890380859375,
+      "logps/rejected": -441.5672912597656,
+      "loss": 0.5156,
+      "rewards/accuracies": 0.7250000238418579,
+      "rewards/chosen": -0.9513274431228638,
+      "rewards/margins": 0.6953738331794739,
+      "rewards/rejected": -1.6467012166976929,
+      "step": 650
+    },
+    {
+      "epoch": 0.6907378335949764,
+      "grad_norm": 35.011717804641286,
+      "learning_rate": 1.319240907040458e-07,
+      "logits/chosen": 2.821119785308838,
+      "logits/rejected": 3.6564712524414062,
+      "logps/chosen": -379.1351318359375,
+      "logps/rejected": -446.2752990722656,
+      "loss": 0.5005,
+      "rewards/accuracies": 0.7875000238418579,
+      "rewards/chosen": -0.8885371088981628,
+      "rewards/margins": 0.9357091188430786,
+      "rewards/rejected": -1.8242461681365967,
+      "step": 660
+    },
+    {
+      "epoch": 0.7012035583464155,
+      "grad_norm": 44.38851088783024,
+      "learning_rate": 1.239457282149695e-07,
+      "logits/chosen": 2.5138425827026367,
+      "logits/rejected": 3.129256010055542,
+      "logps/chosen": -400.1007080078125,
+      "logps/rejected": -483.2186584472656,
+      "loss": 0.4896,
+      "rewards/accuracies": 0.7437499761581421,
+      "rewards/chosen": -1.0179020166397095,
+      "rewards/margins": 0.9014749526977539,
+      "rewards/rejected": -1.9193767309188843,
+      "step": 670
+    },
+    {
+      "epoch": 0.7116692830978545,
+      "grad_norm": 36.04089864350046,
+      "learning_rate": 1.1613595214152711e-07,
+      "logits/chosen": 2.95906138420105,
+      "logits/rejected": 3.7457995414733887,
+      "logps/chosen": -422.68048095703125,
+      "logps/rejected": -492.7059020996094,
+      "loss": 0.5295,
+      "rewards/accuracies": 0.75,
+      "rewards/chosen": -1.0410895347595215,
+      "rewards/margins": 0.9487093687057495,
+      "rewards/rejected": -1.98979914188385,
+      "step": 680
+    },
+    {
+      "epoch": 0.7221350078492935,
+      "grad_norm": 27.994923582530255,
+      "learning_rate": 1.0850520736699362e-07,
+      "logits/chosen": 2.7905449867248535,
+      "logits/rejected": 3.9359989166259766,
+      "logps/chosen": -395.8002624511719,
+      "logps/rejected": -444.7784118652344,
+      "loss": 0.4713,
+      "rewards/accuracies": 0.7875000238418579,
+      "rewards/chosen": -1.1196714639663696,
+      "rewards/margins": 0.9401882290840149,
+      "rewards/rejected": -2.0598597526550293,
+      "step": 690
+    },
+    {
+      "epoch": 0.7326007326007326,
+      "grad_norm": 35.28859090328718,
+      "learning_rate": 1.0106369933615042e-07,
+      "logits/chosen": 3.153747081756592,
+      "logits/rejected": 3.7201926708221436,
+      "logps/chosen": -380.5829772949219,
+      "logps/rejected": -441.12725830078125,
+      "loss": 0.4963,
+      "rewards/accuracies": 0.768750011920929,
+      "rewards/chosen": -1.1806877851486206,
+      "rewards/margins": 0.8047922849655151,
+      "rewards/rejected": -1.9854803085327148,
+      "step": 700
+    },
+    {
+      "epoch": 0.7326007326007326,
+      "eval_logits/chosen": 2.964158535003662,
+      "eval_logits/rejected": 3.7354674339294434,
+      "eval_logps/chosen": -406.1936950683594,
+      "eval_logps/rejected": -472.17718505859375,
+      "eval_loss": 0.5121069550514221,
+      "eval_rewards/accuracies": 0.7678571343421936,
+      "eval_rewards/chosen": -1.1446545124053955,
+      "eval_rewards/margins": 0.9155932664871216,
+      "eval_rewards/rejected": -2.0602478981018066,
+      "eval_runtime": 206.3663,
+      "eval_samples_per_second": 9.692,
+      "eval_steps_per_second": 0.305,
+      "step": 700
+    },
+    {
+      "epoch": 0.7430664573521716,
+      "grad_norm": 38.837178916274965,
+      "learning_rate": 9.382138040640714e-08,
+      "logits/chosen": 3.2909464836120605,
+      "logits/rejected": 4.154778480529785,
+      "logps/chosen": -395.8663635253906,
+      "logps/rejected": -456.48876953125,
+      "loss": 0.5616,
+      "rewards/accuracies": 0.7437499761581421,
+      "rewards/chosen": -1.310706615447998,
+      "rewards/margins": 0.7455329298973083,
+      "rewards/rejected": -2.056239604949951,
+      "step": 710
+    },
+    {
+      "epoch": 0.7535321821036107,
+      "grad_norm": 30.704020790083344,
+      "learning_rate": 8.678793653740632e-08,
+      "logits/chosen": 2.996277093887329,
+      "logits/rejected": 3.8715827465057373,
+      "logps/chosen": -400.0126037597656,
+      "logps/rejected": -464.01190185546875,
+      "loss": 0.485,
+      "rewards/accuracies": 0.737500011920929,
+      "rewards/chosen": -1.0708668231964111,
+      "rewards/margins": 1.039184808731079,
+      "rewards/rejected": -2.1100516319274902,
+      "step": 720
+    },
+    {
+      "epoch": 0.7639979068550498,
+      "grad_norm": 35.131057152362374,
+      "learning_rate": 7.997277433690983e-08,
+      "logits/chosen": 3.1339640617370605,
+      "logits/rejected": 3.905971050262451,
+      "logps/chosen": -408.9098205566406,
+      "logps/rejected": -444.6366271972656,
+      "loss": 0.5214,
+      "rewards/accuracies": 0.6937500238418579,
+      "rewards/chosen": -1.3567955493927002,
+      "rewards/margins": 0.6699576377868652,
+      "rewards/rejected": -2.0267529487609863,
+      "step": 730
+    },
+    {
+      "epoch": 0.7744636316064888,
+      "grad_norm": 36.00931667563455,
+      "learning_rate": 7.338500848029602e-08,
+      "logits/chosen": 3.6650943756103516,
+      "logits/rejected": 4.101107120513916,
+      "logps/chosen": -385.85235595703125,
+      "logps/rejected": -435.08001708984375,
+      "loss": 0.511,
+      "rewards/accuracies": 0.706250011920929,
+      "rewards/chosen": -1.2581779956817627,
+      "rewards/margins": 0.7443889379501343,
+      "rewards/rejected": -2.0025668144226074,
+      "step": 740
+    },
+    {
+      "epoch": 0.7849293563579278,
+      "grad_norm": 37.253882347681056,
+      "learning_rate": 6.70334495204884e-08,
+      "logits/chosen": 3.2969279289245605,
+      "logits/rejected": 3.850818634033203,
+      "logps/chosen": -392.4854736328125,
+      "logps/rejected": -471.72735595703125,
+      "loss": 0.4954,
+      "rewards/accuracies": 0.75,
+      "rewards/chosen": -1.3654239177703857,
+      "rewards/margins": 0.8617910146713257,
+      "rewards/rejected": -2.227215051651001,
+      "step": 750
+    },
+    {
+      "epoch": 0.7953950811093669,
+      "grad_norm": 29.3295504287162,
+      "learning_rate": 6.092659210462231e-08,
+      "logits/chosen": 3.505380153656006,
+      "logits/rejected": 4.019203186035156,
+      "logps/chosen": -379.2978820800781,
+      "logps/rejected": -467.8504943847656,
+      "loss": 0.5216,
+      "rewards/accuracies": 0.762499988079071,
+      "rewards/chosen": -1.297946572303772,
+      "rewards/margins": 0.8633238077163696,
+      "rewards/rejected": -2.1612706184387207,
+      "step": 760
+    },
+    {
+      "epoch": 0.8058608058608059,
+      "grad_norm": 21.569914844701195,
+      "learning_rate": 5.507260361320737e-08,
+      "logits/chosen": 2.9461817741394043,
+      "logits/rejected": 3.025027275085449,
+      "logps/chosen": -437.7095642089844,
+      "logps/rejected": -529.31640625,
+      "loss": 0.4845,
+      "rewards/accuracies": 0.706250011920929,
+      "rewards/chosen": -1.2131383419036865,
+      "rewards/margins": 0.7450088262557983,
+      "rewards/rejected": -1.9581472873687744,
+      "step": 770
+    },
+    {
+      "epoch": 0.8163265306122449,
+      "grad_norm": 45.89392227655327,
+      "learning_rate": 4.947931323697982e-08,
+      "logits/chosen": 2.7798984050750732,
+      "logits/rejected": 3.5839569568634033,
+      "logps/chosen": -432.95355224609375,
+      "logps/rejected": -481.6676330566406,
+      "loss": 0.5095,
+      "rewards/accuracies": 0.768750011920929,
+      "rewards/chosen": -1.0683799982070923,
+      "rewards/margins": 0.9385541677474976,
+      "rewards/rejected": -2.0069339275360107,
+      "step": 780
+    },
+    {
+      "epoch": 0.826792255363684,
+      "grad_norm": 37.03603192236556,
+      "learning_rate": 4.415420150605398e-08,
+      "logits/chosen": 3.013467311859131,
+      "logits/rejected": 4.149591445922852,
+      "logps/chosen": -408.00384521484375,
+      "logps/rejected": -459.72137451171875,
+      "loss": 0.5355,
+      "rewards/accuracies": 0.737500011920929,
+      "rewards/chosen": -1.2126652002334595,
+      "rewards/margins": 0.8756219148635864,
+      "rewards/rejected": -2.088286876678467,
+      "step": 790
+    },
+    {
+      "epoch": 0.837257980115123,
+      "grad_norm": 36.63966888461611,
+      "learning_rate": 3.9104390285376374e-08,
+      "logits/chosen": 3.445955991744995,
+      "logits/rejected": 3.866286516189575,
+      "logps/chosen": -404.0279846191406,
+      "logps/rejected": -477.2325744628906,
+      "loss": 0.5009,
+      "rewards/accuracies": 0.7250000238418579,
+      "rewards/chosen": -1.2799769639968872,
+      "rewards/margins": 0.7887359857559204,
+      "rewards/rejected": -2.0687129497528076,
+      "step": 800
+    },
+    {
+      "epoch": 0.837257980115123,
+      "eval_logits/chosen": 2.880671501159668,
+      "eval_logits/rejected": 3.620241403579712,
+      "eval_logps/chosen": -404.9854736328125,
+      "eval_logps/rejected": -468.59698486328125,
+      "eval_loss": 0.5099306702613831,
+      "eval_rewards/accuracies": 0.7678571343421936,
+      "eval_rewards/chosen": -1.132572054862976,
+      "eval_rewards/margins": 0.8918740749359131,
+      "eval_rewards/rejected": -2.0244460105895996,
+      "eval_runtime": 207.1194,
+      "eval_samples_per_second": 9.656,
+      "eval_steps_per_second": 0.304,
+      "step": 800
+    },
+    {
+      "epoch": 0.847723704866562,
+      "grad_norm": 33.12574503192376,
+      "learning_rate": 3.433663324986208e-08,
+      "logits/chosen": 2.9985527992248535,
+      "logits/rejected": 4.114502906799316,
+      "logps/chosen": -416.48870849609375,
+      "logps/rejected": -452.5269470214844,
+      "loss": 0.5217,
+      "rewards/accuracies": 0.7250000238418579,
+      "rewards/chosen": -1.1603349447250366,
+      "rewards/margins": 0.857321560382843,
+      "rewards/rejected": -2.0176565647125244,
+      "step": 810
+    },
+    {
+      "epoch": 0.858189429618001,
+      "grad_norm": 33.66040812926622,
+      "learning_rate": 2.9857306851953897e-08,
+      "logits/chosen": 3.03739857673645,
+      "logits/rejected": 3.649869203567505,
+      "logps/chosen": -421.085693359375,
+      "logps/rejected": -482.296142578125,
+      "loss": 0.5146,
+      "rewards/accuracies": 0.71875,
+      "rewards/chosen": -1.0747066736221313,
+      "rewards/margins": 0.8519481420516968,
+      "rewards/rejected": -1.9266548156738281,
+      "step": 820
+    },
+    {
+      "epoch": 0.8686551543694401,
+      "grad_norm": 31.816093073286183,
+      "learning_rate": 2.567240179368185e-08,
+      "logits/chosen": 3.1092729568481445,
+      "logits/rejected": 3.5417799949645996,
+      "logps/chosen": -384.48370361328125,
+      "logps/rejected": -454.34490966796875,
+      "loss": 0.5214,
+      "rewards/accuracies": 0.699999988079071,
+      "rewards/chosen": -1.1859172582626343,
+      "rewards/margins": 0.811884880065918,
+      "rewards/rejected": -1.9978021383285522,
+      "step": 830
+    },
+    {
+      "epoch": 0.8791208791208791,
+      "grad_norm": 28.55090493864424,
+      "learning_rate": 2.1787515014630357e-08,
+      "logits/chosen": 2.9592807292938232,
+      "logits/rejected": 3.8776512145996094,
+      "logps/chosen": -400.7276306152344,
+      "logps/rejected": -469.5052185058594,
+      "loss": 0.4815,
+      "rewards/accuracies": 0.762499988079071,
+      "rewards/chosen": -1.125767469406128,
+      "rewards/margins": 0.8751488924026489,
+      "rewards/rejected": -2.0009164810180664,
+      "step": 840
+    },
+    {
+      "epoch": 0.8895866038723181,
+      "grad_norm": 31.22183892445856,
+      "learning_rate": 1.820784220652766e-08,
+      "logits/chosen": 3.456631898880005,
+      "logits/rejected": 4.290311336517334,
+      "logps/chosen": -399.3476257324219,
+      "logps/rejected": -439.2876892089844,
+      "loss": 0.4787,
+      "rewards/accuracies": 0.737500011920929,
+      "rewards/chosen": -1.0915892124176025,
+      "rewards/margins": 0.9232375025749207,
+      "rewards/rejected": -2.014826774597168,
+      "step": 850
+    },
+    {
+      "epoch": 0.9000523286237572,
+      "grad_norm": 35.15771241025392,
+      "learning_rate": 1.4938170864468636e-08,
+      "logits/chosen": 3.2210631370544434,
+      "logits/rejected": 4.138030052185059,
+      "logps/chosen": -412.39312744140625,
+      "logps/rejected": -465.10247802734375,
+      "loss": 0.4928,
+      "rewards/accuracies": 0.768750011920929,
+      "rewards/chosen": -1.153668761253357,
+      "rewards/margins": 0.8063527941703796,
+      "rewards/rejected": -1.960021734237671,
+      "step": 860
+    },
+    {
+      "epoch": 0.9105180533751962,
+      "grad_norm": 33.04453272484295,
+      "learning_rate": 1.1982873884064465e-08,
+      "logits/chosen": 2.7921483516693115,
+      "logits/rejected": 3.708995819091797,
+      "logps/chosen": -389.3924865722656,
+      "logps/rejected": -452.8601989746094,
+      "loss": 0.5058,
+      "rewards/accuracies": 0.768750011920929,
+      "rewards/chosen": -1.1078681945800781,
+      "rewards/margins": 0.8689074516296387,
+      "rewards/rejected": -1.9767755270004272,
+      "step": 870
+    },
+    {
+      "epoch": 0.9209837781266352,
+      "grad_norm": 35.629216258806984,
+      "learning_rate": 9.345903713082304e-09,
+      "logits/chosen": 2.9039440155029297,
+      "logits/rejected": 3.495666980743408,
+      "logps/chosen": -428.51153564453125,
+      "logps/rejected": -496.33441162109375,
+      "loss": 0.5042,
+      "rewards/accuracies": 0.7250000238418579,
+      "rewards/chosen": -1.1306082010269165,
+      "rewards/margins": 0.8036009669303894,
+      "rewards/rejected": -1.9342091083526611,
+      "step": 880
+    },
+    {
+      "epoch": 0.9314495028780743,
+      "grad_norm": 106.56418629412597,
+      "learning_rate": 7.030787065396865e-09,
+      "logits/chosen": 2.817387819290161,
+      "logits/rejected": 3.4940922260284424,
+      "logps/chosen": -410.35577392578125,
+      "logps/rejected": -489.2994079589844,
+      "loss": 0.5174,
+      "rewards/accuracies": 0.706250011920929,
+      "rewards/chosen": -1.1004455089569092,
+      "rewards/margins": 0.8370639681816101,
+      "rewards/rejected": -1.937509536743164,
+      "step": 890
+    },
+    {
+      "epoch": 0.9419152276295133,
+      "grad_norm": 36.70600586918964,
+      "learning_rate": 5.04062020432286e-09,
+      "logits/chosen": 2.8297171592712402,
+      "logits/rejected": 3.260392665863037,
+      "logps/chosen": -402.04913330078125,
+      "logps/rejected": -482.6382751464844,
+      "loss": 0.4926,
+      "rewards/accuracies": 0.75,
+      "rewards/chosen": -1.1460270881652832,
+      "rewards/margins": 0.7792080640792847,
+      "rewards/rejected": -1.9252351522445679,
+      "step": 900
+    },
+    {
+      "epoch": 0.9419152276295133,
+      "eval_logits/chosen": 2.8668997287750244,
+      "eval_logits/rejected": 3.6229140758514404,
+      "eval_logps/chosen": -401.8017883300781,
+      "eval_logps/rejected": -466.17242431640625,
+      "eval_loss": 0.5090468525886536,
+      "eval_rewards/accuracies": 0.773809552192688,
+      "eval_rewards/chosen": -1.1007351875305176,
+      "eval_rewards/margins": 0.8994654417037964,
+      "eval_rewards/rejected": -2.0002005100250244,
+      "eval_runtime": 206.5006,
+      "eval_samples_per_second": 9.685,
+      "eval_steps_per_second": 0.305,
+      "step": 900
+    },
+    {
+      "epoch": 0.9523809523809523,
+      "grad_norm": 37.09203561694287,
+      "learning_rate": 3.3780648016376866e-09,
+      "logits/chosen": 3.3864834308624268,
+      "logits/rejected": 3.882901430130005,
+      "logps/chosen": -339.8305358886719,
+      "logps/rejected": -407.73040771484375,
+      "loss": 0.5015,
+      "rewards/accuracies": 0.762499988079071,
+      "rewards/chosen": -1.1011087894439697,
+      "rewards/margins": 0.8280231356620789,
+      "rewards/rejected": -1.9291318655014038,
+      "step": 910
+    },
+    {
+      "epoch": 0.9628466771323915,
+      "grad_norm": 28.19193668627002,
+      "learning_rate": 2.0453443778310766e-09,
+      "logits/chosen": 2.986339807510376,
+      "logits/rejected": 3.869495391845703,
+      "logps/chosen": -399.61761474609375,
+      "logps/rejected": -450.4912109375,
+      "loss": 0.505,
+      "rewards/accuracies": 0.768750011920929,
+      "rewards/chosen": -1.0854606628417969,
+      "rewards/margins": 0.8986684679985046,
+      "rewards/rejected": -1.9841293096542358,
+      "step": 920
+    },
+    {
+      "epoch": 0.9733124018838305,
+      "grad_norm": 32.3326199545507,
+      "learning_rate": 1.0442413283435758e-09,
+      "logits/chosen": 2.8337090015411377,
+      "logits/rejected": 3.929595470428467,
+      "logps/chosen": -426.2201232910156,
+      "logps/rejected": -462.2978515625,
+      "loss": 0.5018,
+      "rewards/accuracies": 0.737500011920929,
+      "rewards/chosen": -1.014497995376587,
+      "rewards/margins": 0.8778727650642395,
+      "rewards/rejected": -1.8923708200454712,
+      "step": 930
+    },
+    {
+      "epoch": 0.9837781266352695,
+      "grad_norm": 28.69530353443634,
+      "learning_rate": 3.760945397705828e-10,
+      "logits/chosen": 2.8209691047668457,
+      "logits/rejected": 3.537851333618164,
+      "logps/chosen": -376.01727294921875,
+      "logps/rejected": -450.14923095703125,
+      "loss": 0.5011,
+      "rewards/accuracies": 0.7250000238418579,
+      "rewards/chosen": -1.0734994411468506,
+      "rewards/margins": 0.799116313457489,
+      "rewards/rejected": -1.8726158142089844,
+      "step": 940
+    },
+    {
+      "epoch": 0.9942438513867086,
+      "grad_norm": 33.327725431244126,
+      "learning_rate": 4.17975992204056e-11,
+      "logits/chosen": 2.93715500831604,
+      "logits/rejected": 3.396860122680664,
+      "logps/chosen": -426.2603454589844,
+      "logps/rejected": -476.61175537109375,
+      "loss": 0.4972,
+      "rewards/accuracies": 0.737500011920929,
+      "rewards/chosen": -1.1491888761520386,
+      "rewards/margins": 0.7069063186645508,
+      "rewards/rejected": -1.8560950756072998,
+      "step": 950
+    },
+    {
+      "epoch": 0.9994767137624281,
+      "step": 955,
+      "total_flos": 0.0,
+      "train_loss": 0.5469793729133007,
+      "train_runtime": 17738.8932,
+      "train_samples_per_second": 3.446,
+      "train_steps_per_second": 0.054
+    }
+  ],
+  "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
+}