Add files using upload-large-folder tool

healed/calib/glean_keep25_s1224/args.json

@@ -0,0 +1,62 @@
+{
+  "student": "outputs/pruned/glean-0125inst-general-keep25",
+  "teacher": "allenai/OLMoE-1B-7B-0125-Instruct",
+  "training_mode": "off-policy",
+  "kl_direction": "forward",
+  "dataset": "allenai/RLVR-MATH",
+  "dataset_sources": null,
+  "max_difficulty": null,
+  "trajectories": "outputs/teacher_trajectories/dolci_math_curated.jsonl",
+  "trajectory_dataset": "allenai/Dolci-Instruct-RL",
+  "off_policy_frames": "chat",
+  "off_policy_max_seq_len": 2048,
+  "topk_targets": "outputs/teacher_trajectories/dolci_combined_top128",
+  "max_loss_tokens": null,
+  "loss_tokens_per_step": 120000,
+  "teacher_device": "cuda:0",
+  "student_device": "cuda:0",
+  "lr": 3e-05,
+  "optimizer": "adamw8bit",
+  "weight_decay": 0.1,
+  "epochs": 9,
+  "prompts_per_step": 256,
+  "group_size": 1,
+  "rollout_batch": 64,
+  "micro_batch": 3,
+  "max_new_tokens": 256,
+  "max_prompt_len": 1024,
+  "warmup_steps": 10,
+  "max_grad_norm": 1.0,
+  "eval_every": 10,
+  "gsm8k_every": 0,
+  "gsm8k_n": 256,
+  "gsm8k_batch": 16,
+  "gsm8k_max_new_tokens": 512,
+  "gsm8k_frames": "chat",
+  "save_every": 400,
+  "out_dir": "outputs/healed/calib/glean_keep25_s1224",
+  "sweep": 400,
+  "wandb": true,
+  "wandb_project": "glean-calib",
+  "wandb_run_name": "calib_keep25_400",
+  "wandb_run_id": null,
+  "wandb_resume": null,
+  "wandb_mode": "online",
+  "no_wandb_sync": false,
+  "debug": false,
+  "resume_from": "outputs/healed/calib/glean_keep25_s1224/step0300",
+  "start_step": 0,
+  "no_grad_checkpointing": false,
+  "seed": 1224,
+  "no_teacher_overlap": false,
+  "sync_checkpoints": false,
+  "rollout_engine": "hf",
+  "vllm_gpu": null,
+  "vllm_port": 8377,
+  "vllm_refresh_every": 5,
+  "vllm_serve_bin": "vllm-plugin/.venv/bin/python",
+  "vllm_gpu_mem_util": 0.85,
+  "vllm_refresh_mode": "reload",
+  "vllm_live_dir": null,
+  "resolved_kl_direction": "forward"
+}

healed/calib/glean_keep50_s1224/step0300/chat_template.jinja

@@ -0,0 +1,9 @@
+{{ bos_token }}{% for message in messages %}{% if message['role'] == 'system' %}{{ '<|system|>
+' + message['content'] + '
+' }}{% elif message['role'] == 'user' %}{{ '<|user|>
+' + message['content'] + '
+' }}{% elif message['role'] == 'assistant' %}{% if not loop.last %}{{ '<|assistant|>
+'  + message['content'] + eos_token + '
+' }}{% else %}{{ '<|assistant|>
+'  + message['content'] + eos_token }}{% endif %}{% endif %}{% if loop.last and add_generation_prompt %}{{ '<|assistant|>
+' }}{% endif %}{% endfor %}

healed/calib/glean_keep50_s1224/step0300/config.json

@@ -0,0 +1,918 @@
+{
+  "architectures": [
+    "PrunedOlmoeForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "auto_map": {
+    "AutoConfig": "configuration_pruned_olmoe.PrunedOlmoeConfig",
+    "AutoModelForCausalLM": "modeling_pruned_olmoe.PrunedOlmoeForCausalLM"
+  },
+  "clip_qkv": null,
+  "dtype": "bfloat16",
+  "eos_token_id": 50279,
+  "expert_widths": [
+    [
+      768,
+      384,
+      384,
+      384,
+      896,
+      896,
+      256,
+      1024,
+      256,
+      384,
+      1024,
+      384,
+      128,
+      512,
+      896,
+      896,
+      128,
+      768,
+      768,
+      512,
+      384,
+      512,
+      640,
+      896,
+      512,
+      512,
+      512,
+      128,
+      896,
+      384,
+      896,
+      512,
+      512,
+      768,
+      896,
+      1024,
+      256,
+      512,
+      256,
+      256,
+      256,
+      896,
+      512,
+      1024,
+      1024,
+      256,
+      896,
+      512,
+      640,
+      1024,
+      896,
+      768,
+      768,
+      256,
+      128
+    ],
+    [
+      896,
+      896,
+      1024,
+      256,
+      896,
+      384,
+      640,
+      512,
+      1024,
+      256,
+      256,
+      640,
+      896,
+      896,
+      896,
+      1024,
+      640,
+      128,
+      1024,
+      384,
+      768,
+      128,
+      256,
+      640,
+      384,
+      896,
+      1024,
+      1024,
+      896,
+      1024,
+      1024,
+      768,
+      768,
+      256,
+      1024,
+      640,
+      768,
+      640,
+      384,
+      1024,
+      256,
+      768,
+      256,
+      1024,
+      1024,
+      512,
+      512,
+      256,
+      256
+    ],
+    [
+      640,
+      1024,
+      384,
+      896,
+      1024,
+      512,
+      256,
+      1024,
+      768,
+      512,
+      128,
+      1024,
+      1024,
+      1024,
+      512,
+      1024,
+      896,
+      384,
+      768,
+      640,
+      1024,
+      768,
+      896,
+      256,
+      896,
+      1024,
+      640,
+      512,
+      1024,
+      128,
+      768,
+      640,
+      256,
+      640,
+      512,
+      256,
+      896,
+      640,
+      256,
+      896,
+      768,
+      512,
+      1024,
+      896,
+      896,
+      768,
+      384,
+      128
+    ],
+    [
+      768,
+      1024,
+      1024,
+      512,
+      512,
+      1024,
+      640,
+      256,
+      640,
+      256,
+      1024,
+      384,
+      896,
+      128,
+      896,
+      768,
+      256,
+      512,
+      384,
+      1024,
+      768,
+      768,
+      1024,
+      896,
+      1024,
+      768,
+      384,
+      256,
+      512,
+      768,
+      1024,
+      768,
+      896,
+      128,
+      384,
+      896,
+      896,
+      896,
+      896,
+      768,
+      896,
+      256,
+      768,
+      1024,
+      896,
+      1024,
+      256
+    ],
+    [
+      896,
+      512,
+      256,
+      256,
+      640,
+      768,
+      640,
+      896,
+      640,
+      640,
+      896,
+      896,
+      640,
+      896,
+      768,
+      1024,
+      512,
+      896,
+      512,
+      256,
+      384,
+      256,
+      768,
+      512,
+      1024,
+      384,
+      768,
+      768,
+      512,
+      768,
+      896,
+      640,
+      896,
+      256,
+      384,
+      512,
+      640,
+      640,
+      512,
+      384,
+      384,
+      384,
+      768,
+      128,
+      256,
+      896,
+      640,
+      1024,
+      512,
+      1024,
+      384,
+      1024
+    ],
+    [
+      1024,
+      384,
+      256,
+      384,
+      896,
+      768,
+      512,
+      896,
+      640,
+      896,
+      896,
+      256,
+      896,
+      512,
+      768,
+      640,
+      896,
+      512,
+      384,
+      768,
+      640,
+      640,
+      512,
+      256,
+      896,
+      640,
+      384,
+      256,
+      512,
+      768,
+      768,
+      384,
+      1024,
+      384,
+      384,
+      256,
+      896,
+      256,
+      768,
+      768,
+      640,
+      768,
+      896,
+      512,
+      896,
+      256,
+      896,
+      1024,
+      896,
+      640,
+      384,
+      384
+    ],
+    [
+      640,
+      256,
+      768,
+      512,
+      1024,
+      896,
+      896,
+      768,
+      512,
+      128,
+      896,
+      512,
+      384,
+      256,
+      640,
+      896,
+      1024,
+      896,
+      256,
+      640,
+      896,
+      256,
+      896,
+      640,
+      640,
+      384,
+      512,
+      1024,
+      768,
+      640,
+      768,
+      256,
+      384,
+      640,
+      512,
+      128,
+      256,
+      512,
+      768,
+      896,
+      256,
+      1024,
+      128,
+      128,
+      896,
+      896,
+      1024,
+      384,
+      768,
+      896,
+      768,
+      1024
+    ],
+    [
+      896,
+      256,
+      1024,
+      1024,
+      512,
+      256,
+      256,
+      384,
+      640,
+      768,
+      768,
+      768,
+      1024,
+      384,
+      768,
+      512,
+      256,
+      512,
+      768,
+      640,
+      256,
+      896,
+      896,
+      768,
+      640,
+      512,
+      768,
+      640,
+      640,
+      768,
+      768,
+      512,
+      640,
+      512,
+      512,
+      256,
+      896,
+      128,
+      768,
+      384,
+      128,
+      768,
+      1024,
+      384,
+      384,
+      640,
+      768,
+      768,
+      768,
+      512,
+      1024,
+      640,
+      384
+    ],
+    [
+      512,
+      640,
+      256,
+      512,
+      640,
+      128,
+      512,
+      256,
+      768,
+      512,
+      640,
+      640,
+      768,
+      128,
+      384,
+      896,
+      896,
+      384,
+      256,
+      896,
+      512,
+      128,
+      768,
+      768,
+      256,
+      768,
+      896,
+      384,
+      768,
+      640,
+      640,
+      256,
+      512,
+      512,
+      1024,
+      896,
+      640,
+      256,
+      768,
+      512,
+      256,
+      256,
+      896,
+      640,
+      256,
+      512,
+      384,
+      512,
+      384,
+      640,
+      256,
+      1024,
+      1024,
+      640,
+      256,
+      896,
+      512,
+      896
+    ],
+    [
+      640,
+      640,
+      896,
+      512,
+      256,
+      1024,
+      384,
+      1024,
+      640,
+      384,
+      640,
+      384,
+      512,
+      896,
+      512,
+      256,
+      896,
+      640,
+      896,
+      640,
+      256,
+      256,
+      896,
+      384,
+      640,
+      768,
+      256,
+      640,
+      896,
+      512,
+      768,
+      896,
+      128,
+      128,
+      640,
+      768,
+      768,
+      384,
+      896,
+      640,
+      768,
+      384,
+      512,
+      1024,
+      768,
+      896,
+      896,
+      512,
+      640,
+      768,
+      512,
+      640,
+      256
+    ],
+    [
+      640,
+      512,
+      896,
+      768,
+      256,
+      384,
+      768,
+      384,
+      640,
+      640,
+      896,
+      512,
+      384,
+      896,
+      256,
+      640,
+      1024,
+      896,
+      896,
+      256,
+      512,
+      896,
+      896,
+      128,
+      256,
+      896,
+      512,
+      896,
+      384,
+      384,
+      384,
+      640,
+      896,
+      640,
+      640,
+      128,
+      128,
+      896,
+      896,
+      384,
+      896,
+      384,
+      384,
+      640,
+      896,
+      1024,
+      512,
+      1024,
+      256,
+      256,
+      1024,
+      384,
+      768,
+      384
+    ],
+    [
+      768,
+      512,
+      640,
+      768,
+      128,
+      384,
+      256,
+      896,
+      768,
+      768,
+      640,
+      640,
+      384,
+      640,
+      768,
+      768,
+      384,
+      640,
+      896,
+      896,
+      768,
+      1024,
+      896,
+      896,
+      384,
+      896,
+      640,
+      384,
+      768,
+      384,
+      512,
+      256,
+      640,
+      512,
+      1024,
+      1024,
+      512,
+      768,
+      512,
+      384,
+      512,
+      896,
+      128,
+      640,
+      640,
+      1024,
+      384,
+      768,
+      1024,
+      512,
+      512
+    ],
+    [
+      768,
+      384,
+      512,
+      768,
+      256,
+      768,
+      1024,
+      896,
+      640,
+      896,
+      512,
+      768,
+      640,
+      768,
+      896,
+      384,
+      768,
+      896,
+      384,
+      640,
+      128,
+      256,
+      384,
+      256,
+      384,
+      640,
+      1024,
+      768,
+      512,
+      512,
+      768,
+      512,
+      896,
+      896,
+      896,
+      384,
+      768,
+      512,
+      768,
+      1024,
+      896,
+      896,
+      256,
+      384,
+      512,
+      384,
+      1024,
+      512,
+      640,
+      768,
+      640
+    ],
+    [
+      896,
+      896,
+      1024,
+      768,
+      768,
+      384,
+      256,
+      768,
+      256,
+      896,
+      384,
+      896,
+      768,
+      640,
+      512,
+      768,
+      512,
+      512,
+      384,
+      896,
+      256,
+      640,
+      256,
+      640,
+      640,
+      768,
+      512,
+      640,
+      640,
+      768,
+      896,
+      512,
+      896,
+      512,
+      640,
+      768,
+      896,
+      512,
+      128,
+      128,
+      128,
+      768,
+      896,
+      640,
+      384,
+      640,
+      256,
+      896,
+      1024,
+      896,
+      384,
+      512,
+      512
+    ],
+    [
+      640,
+      384,
+      384,
+      512,
+      256,
+      768,
+      1024,
+      384,
+      1024,
+      1024,
+      640,
+      640,
+      384,
+      896,
+      640,
+      256,
+      640,
+      1024,
+      1024,
+      384,
+      512,
+      640,
+      512,
+      512,
+      1024,
+      896,
+      256,
+      896,
+      384,
+      256,
+      128,
+      256,
+      768,
+      768,
+      768,
+      640,
+      768,
+      896,
+      896,
+      768,
+      512,
+      896,
+      128,
+      768,
+      640,
+      256,
+      512,
+      768,
+      384,
+      384,
+      384,
+      640,
+      640,
+      384
+    ],
+    [
+      896,
+      1024,
+      640,
+      768,
+      896,
+      256,
+      128,
+      640,
+      384,
+      128,
+      768,
+      512,
+      768,
+      512,
+      768,
+      896,
+      256,
+      512,
+      640,
+      128,
+      256,
+      384,
+      768,
+      512,
+      128,
+      640,
+      896,
+      640,
+      1024,
+      384,
+      1024,
+      768,
+      512,
+      256,
+      512,
+      256,
+      1024,
+      896,
+      256,
+      768,
+      640,
+      256,
+      384,
+      128,
+      640,
+      1024,
+      768,
+      512,
+      768,
+      1024,
+      768,
+      128,
+      512,
+      640,
+      640,
+      512
+    ]
+  ],
+  "glean_metadata": {
+    "base_model": "allenai/OLMoE-1B-7B-0125-Instruct",
+    "block_size": 128,
+    "criterion": "reap",
+    "dead_experts": 186,
+    "keep_fraction": 0.5,
+    "min_width": 128,
+    "params": 3697555456,
+    "scores": "outputs/scores_general_mix/scores.pt"
+  },
+  "hidden_act": "silu",
+  "hidden_size": 2048,
+  "initializer_range": 0.02,
+  "intermediate_size": 1024,
+  "max_position_embeddings": 4096,
+  "model_type": "pruned_olmoe",
+  "norm_topk_prob": false,
+  "num_attention_heads": 16,
+  "num_experts": 64,
+  "num_experts_per_tok": 8,
+  "num_hidden_layers": 16,
+  "num_key_value_heads": 16,
+  "output_router_logits": false,
+  "pad_token_id": 1,
+  "rms_norm_eps": 1e-05,
+  "rope_scaling": null,
+  "rope_theta": 10000.0,
+  "router_aux_loss_coef": 0.01,
+  "tie_word_embeddings": false,
+  "transformers_version": "4.57.6",
+  "use_cache": false,
+  "vocab_size": 50304
+}

healed/calib/glean_keep50_s1224/step0300/configuration_pruned_olmoe.py

@@ -0,0 +1,27 @@
+"""Configuration for GLEAN-pruned OLMoE: variable-width, variable-count experts.
+"""
+
+from transformers.models.olmoe.configuration_olmoe import OlmoeConfig
+
+
+class PrunedOlmoeConfig(OlmoeConfig):
+    """OlmoeConfig plus a per-(layer, expert) width table.
+
+    ``expert_widths[l]`` lists the SwiGLU intermediate width of each surviving
+    expert in decoder layer ``l``, in expert order. Lists are ragged: layers
+    may keep different numbers of experts (deleted experts simply don't
+    appear — the router in layer ``l`` has ``len(expert_widths[l])`` rows),
+    and each width may differ (multiples of the GEMM block size, 128, for
+    variable-MegaBlocks execution). ``None`` means an unpruned model
+    (uniform ``num_experts`` × ``intermediate_size``).
+
+    The inherited ``num_experts`` / ``intermediate_size`` keep their ORIGINAL
+    (pre-pruning) values for provenance; the width table is authoritative for
+    the built architecture.
+    """
+
+    model_type = "pruned_olmoe"
+
+    def __init__(self, expert_widths: list[list[int]] | None = None, **kwargs):
+        super().__init__(**kwargs)
+        self.expert_widths = expert_widths

healed/calib/glean_keep50_s1224/step0300/generation_config.json

@@ -0,0 +1,6 @@
+{
+  "_from_model_config": true,
+  "eos_token_id": 50279,
+  "pad_token_id": 1,
+  "transformers_version": "4.57.6"
+}

healed/calib/glean_keep50_s1224/step0300/modeling_pruned_olmoe.py

@@ -0,0 +1,66 @@
+"""GLEAN-pruned OLMoE: HF-loadable model with ragged (variable-width) experts.
+
+Pattern follows hbfreed/variable-flex-olmo's PrunedFlexOlmoForCausalLM
+(docs/recon/prior-work-hbfreed.md), generalized from one scalar width to a
+per-(layer, expert) width table: ``super().__init__`` builds the uniform
+architecture from the config, then every MoE block is rebuilt to its pruned
+shape — surviving experts only, each at its own width, router sliced to
+match — so the state dict aligns exactly with what
+``glean.prune.prune_channels_global`` leaves behind.
+
+Caveat: ``output_router_logits=True`` (the load-balancing aux loss) assumes a
+uniform ``config.num_experts`` and is unsupported on ragged models.
+"""
+
+import torch.nn as nn
+from transformers.activations import ACT2FN
+from transformers.models.olmoe.modeling_olmoe import OlmoeForCausalLM
+
+from .configuration_pruned_olmoe import PrunedOlmoeConfig
+
+
+class RaggedOlmoeMLP(nn.Module):
+    """OlmoeMLP with an explicit intermediate width (SwiGLU, no biases)."""
+
+    def __init__(self, hidden_size: int, intermediate_size: int, hidden_act: str):
+        super().__init__()
+        self.hidden_size = hidden_size
+        self.intermediate_size = intermediate_size
+        self.gate_proj = nn.Linear(hidden_size, intermediate_size, bias=False)
+        self.up_proj = nn.Linear(hidden_size, intermediate_size, bias=False)
+        self.down_proj = nn.Linear(intermediate_size, hidden_size, bias=False)
+        self.act_fn = ACT2FN[hidden_act]
+
+    def forward(self, x):
+        return self.down_proj(self.act_fn(self.gate_proj(x)) * self.up_proj(x))
+
+
+class PrunedOlmoeForCausalLM(OlmoeForCausalLM):
+    """OLMoE with per-layer surviving-expert lists at per-expert widths."""
+
+    config_class = PrunedOlmoeConfig
+
+    def __init__(self, config: PrunedOlmoeConfig):
+        super().__init__(config)
+        widths_table = getattr(config, "expert_widths", None)
+        if widths_table is None:
+            return  # unpruned: plain OLMoE
+        if len(widths_table) != len(self.model.layers):
+            raise ValueError(
+                f"expert_widths has {len(widths_table)} rows but the model has "
+                f"{len(self.model.layers)} decoder layers"
+            )
+        for layer, widths in zip(self.model.layers, widths_table):
+            if any(w <= 0 for w in widths):
+                raise ValueError("expert_widths must list surviving experts only (>0)")
+            block = layer.mlp
+            if len(widths) < block.top_k:
+                raise ValueError(
+                    f"a layer keeps {len(widths)} experts < top_k={block.top_k}"
+                )
+            block.num_experts = len(widths)
+            block.gate = nn.Linear(config.hidden_size, len(widths), bias=False)
+            block.experts = nn.ModuleList(
+                RaggedOlmoeMLP(config.hidden_size, w, config.hidden_act)
+                for w in widths
+            )

healed/calib/glean_keep50_s1224/step0300/special_tokens_map.json

@@ -0,0 +1,23 @@
+{
+  "bos_token": {
+    "content": "|||IP_ADDRESS|||",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "|||IP_ADDRESS|||",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<pad>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

healed/calib/glean_keep50_s1224/step0300/tokenizer_config.json

@@ -0,0 +1,247 @@
+{
+  "add_bos_token": false,
+  "add_eos_token": false,
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<|padding|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "50254": {
+      "content": "                        ",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50255": {
+      "content": "                       ",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50256": {
+      "content": "                      ",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50257": {
+      "content": "                     ",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50258": {
+      "content": "                    ",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50259": {
+      "content": "                   ",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50260": {
+      "content": "                  ",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50261": {
+      "content": "                 ",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50262": {
+      "content": "                ",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50263": {
+      "content": "               ",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50264": {
+      "content": "              ",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50265": {
+      "content": "             ",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50266": {
+      "content": "            ",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50267": {
+      "content": "           ",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50268": {
+      "content": "          ",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50269": {
+      "content": "         ",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50270": {
+      "content": "        ",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50271": {
+      "content": "       ",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50272": {
+      "content": "      ",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50273": {
+      "content": "     ",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50274": {
+      "content": "    ",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50275": {
+      "content": "   ",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50276": {
+      "content": "  ",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50277": {
+      "content": "|||EMAIL_ADDRESS|||",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50278": {
+      "content": "|||PHONE_NUMBER|||",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50279": {
+      "content": "|||IP_ADDRESS|||",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "50280": {
+      "content": "<pad>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "|||IP_ADDRESS|||",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "|||IP_ADDRESS|||",
+  "extra_special_tokens": {},
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<pad>",
+  "tokenizer_class": "GPTNeoXTokenizer",
+  "unk_token": null
+}

healed/calib/glean_keep50_s1224/wandb/debug-internal.log

@@ -0,0 +1,222 @@
+{"time":"2026-07-18T00:11:44.793714915-07:00","level":"INFO","msg":"wandb-core"}
+{"time":"2026-07-18T00:11:44.794199111-07:00","level":"INFO","msg":"stream: starting","core version":"0.28.0"}
+{"time":"2026-07-18T00:11:45.013894413-07:00","level":"INFO","msg":"stream: created new stream","id":"yv1sfljs"}
+{"time":"2026-07-18T00:11:45.013970395-07:00","level":"INFO","msg":"handler: started"}
+{"time":"2026-07-18T00:11:45.01412458-07:00","level":"INFO","msg":"stream: started"}
+{"time":"2026-07-18T00:11:45.014196703-07:00","level":"INFO","msg":"sender: started"}
+{"time":"2026-07-18T00:11:45.014196073-07:00","level":"INFO","msg":"writer: started","stream_id":"yv1sfljs"}
+{"time":"2026-07-18T00:11:46.563594303-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"console_offset":0,"console_lines":1}
+{"time":"2026-07-18T00:11:46.95944868-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:12:01.563964302-07:00","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":0,"events_lines":2,"console_offset":0,"console_lines":4,"uploaded_len":2}
+{"time":"2026-07-18T00:12:01.922626494-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:12:16.563902467-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":2,"events_lines":2}
+{"time":"2026-07-18T00:12:17.130910326-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:12:31.564307007-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":4,"events_lines":2}
+{"time":"2026-07-18T00:12:32.103022534-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:12:46.56386523-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":6,"events_lines":2}
+{"time":"2026-07-18T00:12:46.892188012-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:13:01.564664143-07:00","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":0,"history_lines":1,"events_offset":8,"events_lines":2,"console_offset":4,"console_lines":1}
+{"time":"2026-07-18T00:13:01.909293123-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:13:16.563941047-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":10,"events_lines":2}
+{"time":"2026-07-18T00:13:16.895601089-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:13:31.564257044-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":12,"events_lines":2}
+{"time":"2026-07-18T00:13:32.18181939-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:13:46.564284573-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":14,"events_lines":2}
+{"time":"2026-07-18T00:13:46.936002326-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:14:01.564443676-07:00","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":1,"history_lines":1,"events_offset":16,"events_lines":2,"console_offset":5,"console_lines":1}
+{"time":"2026-07-18T00:14:01.963116786-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:14:16.564455264-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":18,"events_lines":2}
+{"time":"2026-07-18T00:14:16.822305163-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:14:31.564400649-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":20,"events_lines":2}
+{"time":"2026-07-18T00:14:31.890525458-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:14:46.563951971-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":22,"events_lines":2}
+{"time":"2026-07-18T00:14:46.929560412-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:15:01.564177895-07:00","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":2,"history_lines":1,"events_offset":24,"events_lines":2,"console_offset":6,"console_lines":1}
+{"time":"2026-07-18T00:15:01.862915982-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:15:16.56379052-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":26,"events_lines":2}
+{"time":"2026-07-18T00:15:16.988090405-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:15:31.564213511-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":28,"events_lines":2}
+{"time":"2026-07-18T00:15:31.847182598-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:15:46.564526319-07:00","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":3,"history_lines":1,"events_offset":30,"events_lines":2,"console_offset":7,"console_lines":1}
+{"time":"2026-07-18T00:15:46.840427383-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:16:01.564267347-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":32,"events_lines":2}
+{"time":"2026-07-18T00:16:01.820678469-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:16:16.564557264-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":34,"events_lines":2}
+{"time":"2026-07-18T00:16:16.851818852-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:16:31.564619773-07:00","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":4,"history_lines":1,"events_offset":36,"events_lines":2,"console_offset":8,"console_lines":1}
+{"time":"2026-07-18T00:16:31.895043034-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:16:46.56432257-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":38,"events_lines":2}
+{"time":"2026-07-18T00:16:46.891202585-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:17:01.564442971-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":40,"events_lines":2}
+{"time":"2026-07-18T00:17:01.908399859-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:17:16.563926552-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":42,"events_lines":2}
+{"time":"2026-07-18T00:17:16.927531917-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:17:31.563684081-07:00","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":5,"history_lines":1,"events_offset":44,"events_lines":2,"console_offset":9,"console_lines":1}
+{"time":"2026-07-18T00:17:31.921757883-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:17:46.564198255-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":46,"events_lines":2}
+{"time":"2026-07-18T00:17:46.805905292-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:18:01.563526169-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":48,"events_lines":2}
+{"time":"2026-07-18T00:18:01.840118346-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:18:16.564432106-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":50,"events_lines":2}
+{"time":"2026-07-18T00:18:16.794446148-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:18:31.564703002-07:00","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":6,"history_lines":1,"events_offset":52,"events_lines":2,"console_offset":10,"console_lines":1}
+{"time":"2026-07-18T00:18:31.819597064-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:18:46.563940664-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":54,"events_lines":2}
+{"time":"2026-07-18T00:18:46.952695178-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:19:01.564288543-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":56,"events_lines":2}
+{"time":"2026-07-18T00:19:01.918920022-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:19:16.563901067-07:00","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":7,"history_lines":1,"events_offset":58,"events_lines":2,"console_offset":11,"console_lines":1}
+{"time":"2026-07-18T00:19:17.074096414-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:19:31.563998077-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":60,"events_lines":2}
+{"time":"2026-07-18T00:19:31.907191739-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:19:46.56447855-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":62,"events_lines":2}
+{"time":"2026-07-18T00:19:46.90061879-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:20:01.563935279-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":64,"events_lines":2}
+{"time":"2026-07-18T00:20:01.898735204-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:20:16.564440073-07:00","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":8,"history_lines":1,"events_offset":66,"events_lines":2,"console_offset":12,"console_lines":1}
+{"time":"2026-07-18T00:20:16.936823187-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:20:31.564514472-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":68,"events_lines":2}
+{"time":"2026-07-18T00:20:31.868945216-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:20:46.56422581-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":70,"events_lines":2}
+{"time":"2026-07-18T00:20:46.897147723-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:21:01.564514706-07:00","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":9,"history_lines":1,"events_offset":72,"events_lines":2,"console_offset":13,"console_lines":3}
+{"time":"2026-07-18T00:21:01.869355794-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:21:16.563734678-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":74,"events_lines":2}
+{"time":"2026-07-18T00:21:16.831598167-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:21:31.563995103-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":76,"events_lines":2}
+{"time":"2026-07-18T00:21:31.899815972-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:21:46.564372652-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":78,"events_lines":2}
+{"time":"2026-07-18T00:21:46.881910279-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:22:01.56378357-07:00","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":10,"history_lines":1,"events_offset":80,"events_lines":2,"console_offset":16,"console_lines":1}
+{"time":"2026-07-18T00:22:01.818130863-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:22:16.563810317-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":82,"events_lines":2}
+{"time":"2026-07-18T00:22:16.836291269-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:22:31.564547399-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":84,"events_lines":2}
+{"time":"2026-07-18T00:22:31.934543744-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:22:46.563939846-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":86,"events_lines":2}
+{"time":"2026-07-18T00:22:46.827645978-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:23:01.564262863-07:00","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":11,"history_lines":1,"events_offset":88,"events_lines":2,"console_offset":17,"console_lines":1}
+{"time":"2026-07-18T00:23:01.900870788-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:23:16.56426557-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":90,"events_lines":2}
+{"time":"2026-07-18T00:23:16.964132741-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:23:31.564312359-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":92,"events_lines":2}
+{"time":"2026-07-18T00:23:32.187248072-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:23:46.563891183-07:00","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":12,"history_lines":1,"events_offset":94,"events_lines":2,"console_offset":18,"console_lines":1}
+{"time":"2026-07-18T00:23:46.912457772-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:24:01.564613043-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":96,"events_lines":2}
+{"time":"2026-07-18T00:24:02.043635822-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:24:16.564234957-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":98,"events_lines":2}
+{"time":"2026-07-18T00:24:16.928796924-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:24:31.564025628-07:00","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":13,"history_lines":1,"events_offset":100,"events_lines":2,"console_offset":19,"console_lines":1}
+{"time":"2026-07-18T00:24:31.871032387-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:24:46.564070206-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":102,"events_lines":2}
+{"time":"2026-07-18T00:24:46.859114451-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:25:01.564600511-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":104,"events_lines":2}
+{"time":"2026-07-18T00:25:01.973402906-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:25:16.563999868-07:00","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":14,"history_lines":1,"events_offset":106,"events_lines":2,"console_offset":20,"console_lines":1}
+{"time":"2026-07-18T00:25:16.959625639-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:25:31.563705186-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":108,"events_lines":2}
+{"time":"2026-07-18T00:25:31.876698292-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:25:46.564018263-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":110,"events_lines":2}
+{"time":"2026-07-18T00:25:46.916899074-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:26:01.564305505-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":112,"events_lines":2}
+{"time":"2026-07-18T00:26:01.905051084-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:26:16.564396279-07:00","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":15,"history_lines":1,"events_offset":114,"events_lines":2,"console_offset":21,"console_lines":1}
+{"time":"2026-07-18T00:26:16.864349449-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:26:31.563716308-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":116,"events_lines":2}
+{"time":"2026-07-18T00:26:31.755454145-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:26:46.564352421-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":118,"events_lines":2}
+{"time":"2026-07-18T00:26:46.876704338-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:27:01.563811525-07:00","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":16,"history_lines":1,"events_offset":120,"events_lines":2,"console_offset":22,"console_lines":1}
+{"time":"2026-07-18T00:27:01.912986151-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:27:16.564115906-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":122,"events_lines":2}
+{"time":"2026-07-18T00:27:16.843111418-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:27:31.563884271-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":124,"events_lines":2}
+{"time":"2026-07-18T00:27:31.834246138-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:27:46.564066778-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":126,"events_lines":2}
+{"time":"2026-07-18T00:27:46.847503645-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:28:01.564107071-07:00","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":17,"history_lines":1,"events_offset":128,"events_lines":2,"console_offset":23,"console_lines":1}
+{"time":"2026-07-18T00:28:01.853620617-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:28:16.564010729-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":130,"events_lines":2}
+{"time":"2026-07-18T00:28:16.868165606-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:28:31.563807194-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":132,"events_lines":2}
+{"time":"2026-07-18T00:28:31.851921055-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:28:46.564422876-07:00","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":18,"history_lines":1,"events_offset":134,"events_lines":2,"console_offset":24,"console_lines":1}
+{"time":"2026-07-18T00:28:46.906155937-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:29:01.563639631-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":136,"events_lines":2}
+{"time":"2026-07-18T00:29:02.014402453-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:29:16.563988264-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":138,"events_lines":2}
+{"time":"2026-07-18T00:29:16.859624352-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:29:31.564477672-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":140,"events_lines":2}
+{"time":"2026-07-18T00:29:31.979688846-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:29:46.56404689-07:00","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":19,"history_lines":1,"events_offset":142,"events_lines":2,"console_offset":25,"console_lines":2}
+{"time":"2026-07-18T00:29:46.924894359-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:30:01.564497866-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":144,"events_lines":2}
+{"time":"2026-07-18T00:30:01.975064748-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:30:16.563591708-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":146,"events_lines":2}
+{"time":"2026-07-18T00:30:16.958259468-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:30:31.56420643-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":148,"events_lines":2}
+{"time":"2026-07-18T00:30:31.859500796-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:30:46.564381467-07:00","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":20,"history_lines":1,"events_offset":150,"events_lines":2,"console_offset":27,"console_lines":1}
+{"time":"2026-07-18T00:30:47.045676291-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:31:01.564530373-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":152,"events_lines":2}
+{"time":"2026-07-18T00:31:01.92285196-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:31:16.563880423-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":154,"events_lines":2}
+{"time":"2026-07-18T00:31:16.940951385-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:31:31.563804932-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":156,"events_lines":2}
+{"time":"2026-07-18T00:31:31.959272248-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:31:46.563917678-07:00","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":21,"history_lines":1,"events_offset":158,"events_lines":2,"console_offset":28,"console_lines":1}
+{"time":"2026-07-18T00:31:46.963435336-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:32:01.563585788-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":160,"events_lines":2}
+{"time":"2026-07-18T00:32:01.939591775-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:32:16.563783507-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":162,"events_lines":2}
+{"time":"2026-07-18T00:32:16.840744651-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:32:31.564470701-07:00","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":22,"history_lines":1,"events_offset":164,"events_lines":2,"console_offset":29,"console_lines":1}
+{"time":"2026-07-18T00:32:31.868900727-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:32:46.564606487-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":166,"events_lines":2}
+{"time":"2026-07-18T00:32:46.828163081-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:33:01.56404467-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":168,"events_lines":2}
+{"time":"2026-07-18T00:33:01.837324004-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:33:16.563946588-07:00","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":23,"history_lines":1,"events_offset":170,"events_lines":2,"console_offset":30,"console_lines":1}
+{"time":"2026-07-18T00:33:16.795555453-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:33:31.564048153-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":172,"events_lines":2}
+{"time":"2026-07-18T00:33:31.957627797-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:33:46.564462138-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":174,"events_lines":2}
+{"time":"2026-07-18T00:33:46.942871664-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:34:01.564659266-07:00","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":24,"history_lines":1,"events_offset":176,"events_lines":2,"console_offset":31,"console_lines":1}
+{"time":"2026-07-18T00:34:01.883060771-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:34:16.563556351-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":178,"events_lines":2}
+{"time":"2026-07-18T00:34:17.267231167-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:34:31.564078689-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":180,"events_lines":2}
+{"time":"2026-07-18T00:34:31.930413569-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:34:46.564487594-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":182,"events_lines":2}
+{"time":"2026-07-18T00:34:47.320596392-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:35:01.564179726-07:00","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":25,"history_lines":1,"events_offset":184,"events_lines":2,"console_offset":32,"console_lines":1}
+{"time":"2026-07-18T00:35:01.892639891-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:35:16.564234779-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":186,"events_lines":2}
+{"time":"2026-07-18T00:35:17.054933692-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:35:31.564577202-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":188,"events_lines":2}
+{"time":"2026-07-18T00:35:31.901152183-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:35:46.564391577-07:00","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":26,"history_lines":1,"events_offset":190,"events_lines":2,"console_offset":33,"console_lines":1}
+{"time":"2026-07-18T00:35:46.856249801-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:36:01.564609006-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":192,"events_lines":2}
+{"time":"2026-07-18T00:36:01.858482536-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:36:16.563577883-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":194,"events_lines":2}
+{"time":"2026-07-18T00:36:17.044588378-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:36:31.563963788-07:00","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":27,"history_lines":1,"events_offset":196,"events_lines":2,"console_offset":34,"console_lines":1}
+{"time":"2026-07-18T00:36:32.810869782-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:36:46.56368993-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":198,"events_lines":2}
+{"time":"2026-07-18T00:36:47.815039621-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:37:01.564345743-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":200,"events_lines":2}
+{"time":"2026-07-18T00:37:02.55826198-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:37:16.564314004-07:00","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":28,"history_lines":1,"events_offset":202,"events_lines":2,"console_offset":35,"console_lines":1}
+{"time":"2026-07-18T00:37:16.879405821-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:37:31.56444287-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":204,"events_lines":2}
+{"time":"2026-07-18T00:37:31.867552193-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:37:46.564050518-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":206,"events_lines":2}
+{"time":"2026-07-18T00:37:46.887679655-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:38:01.563708229-07:00","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":29,"history_lines":1,"events_offset":208,"events_lines":2,"console_offset":36,"console_lines":2}
+{"time":"2026-07-18T00:38:01.86890415-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:38:16.563635908-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":210,"events_lines":2}
+{"time":"2026-07-18T00:38:22.181210541-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+

healed/calib/glean_keep50_s1224/wandb/debug.log

@@ -0,0 +1,19 @@
+2026-07-18 00:11:44,791 INFO    MainThread:418239 [wandb_setup.py:_flush():81] Current SDK version is 0.28.0
+2026-07-18 00:11:44,791 INFO    MainThread:418239 [wandb_setup.py:_flush():81] Configure stats pid to 418239
+2026-07-18 00:11:44,791 INFO    MainThread:418239 [wandb_setup.py:_flush():81] Loading settings from environment variables
+2026-07-18 00:11:44,791 INFO    MainThread:418239 [wandb_init.py:setup_run_log_directory():725] Logging user logs to outputs/healed/calib/glean_keep50_s1224/wandb/run-20260718_001144-yv1sfljs/logs/debug.log
+2026-07-18 00:11:44,791 INFO    MainThread:418239 [wandb_init.py:setup_run_log_directory():726] Logging internal logs to outputs/healed/calib/glean_keep50_s1224/wandb/run-20260718_001144-yv1sfljs/logs/debug-internal.log
+2026-07-18 00:11:44,791 INFO    MainThread:418239 [wandb_init.py:init():768] calling init triggers
+2026-07-18 00:11:44,791 INFO    MainThread:418239 [wandb_init.py:init():773] wandb.init called with sweep_config: {}
+config: {'student': 'outputs/pruned/glean-0125inst-general-keep50', 'teacher': 'allenai/OLMoE-1B-7B-0125-Instruct', 'training_mode': 'off-policy', 'kl_direction': 'forward', 'dataset': 'allenai/RLVR-MATH', 'dataset_sources': None, 'max_difficulty': None, 'trajectories': 'outputs/teacher_trajectories/dolci_math_curated.jsonl', 'trajectory_dataset': 'allenai/Dolci-Instruct-RL', 'off_policy_frames': 'chat', 'off_policy_max_seq_len': 2048, 'topk_targets': 'outputs/teacher_trajectories/dolci_combined_top128', 'max_loss_tokens': None, 'loss_tokens_per_step': 120000, 'teacher_device': 'cuda:0', 'student_device': 'cuda:0', 'lr': 3e-05, 'optimizer': 'adamw8bit', 'weight_decay': 0.1, 'epochs': 9, 'prompts_per_step': 256, 'group_size': 1, 'rollout_batch': 64, 'micro_batch': 3, 'max_new_tokens': 256, 'max_prompt_len': 1024, 'warmup_steps': 10, 'max_grad_norm': 1.0, 'eval_every': 10, 'gsm8k_every': 0, 'gsm8k_n': 256, 'gsm8k_batch': 16, 'gsm8k_max_new_tokens': 512, 'gsm8k_frames': 'chat', 'save_every': 400, 'out_dir': 'outputs/healed/calib/glean_keep50_s1224', 'sweep': 400, 'wandb': True, 'wandb_project': 'glean-calib', 'wandb_run_name': 'calib_keep50_400', 'wandb_run_id': None, 'wandb_resume': None, 'wandb_mode': 'online', 'no_wandb_sync': False, 'debug': False, 'resume_from': 'outputs/healed/calib/glean_keep50_s1224/step0300', 'start_step': 0, 'no_grad_checkpointing': False, 'seed': 1224, 'no_teacher_overlap': False, 'sync_checkpoints': False, 'rollout_engine': 'hf', 'vllm_gpu': None, 'vllm_port': 8377, 'vllm_refresh_every': 5, 'vllm_serve_bin': 'vllm-plugin/.venv/bin/python', 'vllm_gpu_mem_util': 0.85, 'vllm_refresh_mode': 'reload', 'vllm_live_dir': None, 'resolved_kl_direction': 'forward', '_wandb': {}}
+2026-07-18 00:11:44,791 INFO    MainThread:418239 [wandb_init.py:init():816] starting backend
+2026-07-18 00:11:44,791 INFO    MainThread:418239 [wandb_init.py:init():831] sending inform_init request
+2026-07-18 00:11:45,014 INFO    MainThread:418239 [wandb_init.py:init():836] backend started and connected
+2026-07-18 00:11:45,017 INFO    MainThread:418239 [wandb_init.py:init():906] updated telemetry
+2026-07-18 00:11:45,022 INFO    MainThread:418239 [wandb_init.py:init():929] communicating run to backend with 90.0 second timeout
+2026-07-18 00:11:45,312 INFO    MainThread:418239 [wandb_init.py:init():974] starting run threads in backend
+2026-07-18 00:11:45,445 INFO    MainThread:418239 [wandb_run.py:_console_start():2523] atexit reg
+2026-07-18 00:11:45,445 INFO    MainThread:418239 [wandb_run.py:_redirect():2373] redirect: wrap_raw
+2026-07-18 00:11:45,445 INFO    MainThread:418239 [wandb_run.py:_redirect():2442] Wrapping output streams.
+2026-07-18 00:11:45,445 INFO    MainThread:418239 [wandb_run.py:_redirect():2465] Redirects installed.
+2026-07-18 00:11:45,447 INFO    MainThread:418239 [wandb_init.py:init():1012] run started, returning control to user process

healed/calib/glean_keep50_s1224/wandb/run-20260717_194504-jkxl7c7u/files/requirements.txt

@@ -0,0 +1,129 @@
+certifi==2026.6.17
+fonttools==4.63.0
+urllib3==2.7.0
+requests==2.34.2
+nvidia-cublas-cu12==12.6.4.1
+packaging==26.2
+regex==2026.6.28
+portalocker==3.2.0
+safetensors==0.8.0
+datasets==5.0.0
+sqlitedict==2.1.0
+narwhals==2.23.0
+latex2sympy2_extended==1.11.0
+smmap==5.0.3
+colorama==0.4.6
+anyio==4.14.1
+cycler==0.12.1
+pytablewriter==1.2.1
+pydantic_core==2.46.4
+huggingface_hub==0.36.2
+charset-normalizer==3.4.8
+nvidia-cusolver-cu12==11.7.1.2
+joblib==1.5.3
+typing-inspection==0.4.2
+wandb==0.28.0
+bitsandbytes==0.49.2
+kiwisolver==1.5.0
+cuda-toolkit==12.6.3
+propcache==0.5.2
+Jinja2==3.1.6
+cuda-bindings==12.9.7
+pluggy==1.6.0
+pytest==9.1.1
+fsspec==2026.4.0
+tqdm==4.68.3
+psutil==7.2.2
+pydantic==2.13.4
+typepy==1.3.5
+xxhash==3.8.1
+DataProperty==1.1.1
+aiosignal==1.4.0
+annotated-types==0.7.0
+Pygments==2.20.0
+zstandard==0.25.0
+threadpoolctl==3.6.0
+nvidia-cudnn-cu12==9.10.2.21
+evaluate==0.4.6
+pillow==12.3.0
+GitPython==3.1.50
+httpx==0.28.1
+lm_eval==0.4.12
+pytz==2026.2
+mbstrdecoder==1.1.5
+immutabledict==4.3.1
+dill==0.4.1
+mpmath==1.3.0
+yarl==1.24.2
+word2number==1.1
+networkx==3.6.1
+einops==0.8.2
+tabulate==0.10.0
+pathvalidate==3.3.1
+tenacity==9.1.4
+sentry-sdk==2.64.0
+nvidia-nvshmem-cu12==3.4.5
+hf-xet==1.5.1
+six==1.17.0
+pyarrow==24.0.0
+nvidia-curand-cu12==10.3.7.77
+cuda-pathfinder==1.5.6
+sacrebleu==2.6.0
+triton==3.7.1
+frozenlist==1.8.0
+antlr4-python3-runtime==4.11.0
+tabledata==1.3.5
+h11==0.16.0
+nvidia-cusparselt-cu12==0.7.1
+nvidia-nvjitlink-cu12==12.6.85
+scipy==1.18.0
+rouge_score==0.1.2
+transformers==4.57.6
+multiprocess==0.70.19
+absl-py==2.5.0
+accelerate==1.14.0
+click==8.4.2
+lxml==6.1.1
+defusedxml==0.7.1
+setuptools==81.0.0
+nvidia-cuda-cupti-cu12==12.6.80
+httpcore==1.0.9
+langdetect==1.0.9
+nltk==3.10.0
+iniconfig==2.3.0
+nvidia-cuda-runtime-cu12==12.6.77
+nvidia-cusparse-cu12==12.5.4.2
+matplotlib==3.11.0
+protobuf==7.35.1
+nvidia-cufft-cu12==11.3.0.4
+pyparsing==3.3.2
+chardet==6.0.0.post1
+typing_extensions==4.16.0
+stanford-stk==0.7.1
+aiohappyeyeballs==2.7.1
+MarkupSafe==3.0.3
+nvidia-cufile-cu12==1.11.1.6
+tcolorpy==0.1.7
+scikit-learn==1.9.0
+tokenizers==0.22.2
+attrs==26.1.0
+gitdb==4.0.12
+aiohttp==3.14.1
+glean==0.0.1
+platformdirs==4.10.0
+megablocks==0.11.0.dev0
+nvidia-cuda-nvrtc-cu12==12.6.85
+pandas==3.0.3
+multidict==6.7.1
+torch==2.12.1+cu126
+contourpy==1.3.3
+nvidia-nvtx-cu12==12.6.77
+math-verify==0.9.0
+filelock==3.29.5
+idna==3.18
+nvidia-nccl-cu12==2.29.3
+python-dateutil==2.9.0.post0
+more-itertools==11.1.0
+numpy==2.0.2
+sympy==1.14.0
+PyYAML==6.0.3

healed/calib/glean_keep50_s1224/wandb/run-20260717_194504-jkxl7c7u/logs/debug-core.log

@@ -0,0 +1,20 @@
+{"time":"2026-07-17T19:45:03.674982902-07:00","level":"INFO","msg":"main: starting server","port-filename":"/tmp/tmpbbhq929j/port-387507.txt","pid":387507,"detached":false,"idle-timeout":600000000000,"log-level":0,"disable-analytics":false,"shutdown-on-parent-exit":false,"enable-dcgm-profiling":false}
+{"time":"2026-07-17T19:45:03.676638476-07:00","level":"INFO","msg":"server: will exit if parent process dies","ppid":387507}
+{"time":"2026-07-17T19:45:03.676580344-07:00","level":"INFO","msg":"server: accepting connections","addr":{"Name":"/tmp/wandb-387507-387617-4269361802/socket","Net":"unix"}}
+{"time":"2026-07-17T19:45:03.858366542-07:00","level":"INFO","msg":"connection: ManageConnectionData: new connection created","id":"1(@)"}
+{"time":"2026-07-17T19:45:04.067564724-07:00","level":"INFO","msg":"handleInformInit: received","streamId":"jkxl7c7u","id":"1(@)"}
+{"time":"2026-07-17T19:45:04.305047118-07:00","level":"INFO","msg":"handleInformInit: stream started","streamId":"jkxl7c7u","id":"1(@)"}
+{"time":"2026-07-17T19:45:09.734710959-07:00","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"ho1t2e7mfh14"}
+{"time":"2026-07-17T21:12:34.351042664-07:00","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"ho1t2e7mfh14"}
+{"time":"2026-07-17T21:12:35.249866376-07:00","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"ho1t2e7mfh14"}
+{"time":"2026-07-17T21:12:35.253254984-07:00","level":"INFO","msg":"handleInformFinish: finish message received","streamId":"jkxl7c7u","id":"1(@)"}
+{"time":"2026-07-17T21:12:35.259032334-07:00","level":"INFO","msg":"handleInformFinish: stream closed","streamId":"jkxl7c7u","id":"1(@)"}
+{"time":"2026-07-17T21:12:35.324929145-07:00","level":"INFO","msg":"handleInformTeardown: server teardown initiated","id":"1(@)"}
+{"time":"2026-07-17T21:12:35.325007457-07:00","level":"INFO","msg":"handleInformTeardown: server shutdown complete","id":"1(@)"}
+{"time":"2026-07-17T21:12:35.325034578-07:00","level":"INFO","msg":"server is shutting down"}
+{"time":"2026-07-17T21:12:35.325027938-07:00","level":"INFO","msg":"connection: closing","id":"1(@)"}
+{"time":"2026-07-17T21:12:35.32507588-07:00","level":"INFO","msg":"processOutgoingData: finished","id":"1(@)"}
+{"time":"2026-07-17T21:12:35.325120811-07:00","level":"INFO","msg":"connection: closed successfully","id":"1(@)"}
+{"time":"2026-07-17T21:12:35.325131832-07:00","level":"INFO","msg":"connection: ManageConnectionData: connection closed","id":"1(@)"}
+{"time":"2026-07-17T21:12:35.325480104-07:00","level":"INFO","msg":"server: listener closed","addr":{"Name":"/tmp/wandb-387507-387617-4269361802/socket","Net":"unix"}}
+{"time":"2026-07-17T21:12:35.325535646-07:00","level":"INFO","msg":"server is closed"}

healed/calib/glean_keep50_s1224/wandb/run-20260717_224325-kvt72u16/files/config.yaml

@@ -0,0 +1,246 @@
+_wandb:
+    value:
+        cli_version: 0.28.0
+        e:
+            v19kdketdy6gkgkcksw06943ad61z958:
+                args:
+                    - --training-mode
+                    - off-policy
+                    - --kl-direction
+                    - forward
+                    - --topk-targets
+                    - outputs/teacher_trajectories/dolci_combined_top128
+                    - --off-policy-frames
+                    - chat
+                    - --student-device
+                    - cuda:0
+                    - --student
+                    - outputs/pruned/glean-0125inst-general-keep50
+                    - --resume-from
+                    - outputs/healed/calib/glean_keep50_s1224/step0200
+                    - --lr
+                    - "3e-5"
+                    - --optimizer
+                    - adamw8bit
+                    - --epochs
+                    - "9"
+                    - --sweep
+                    - "300"
+                    - --micro-batch
+                    - "3"
+                    - --loss-tokens-per-step
+                    - "120000"
+                    - --gsm8k-every
+                    - "0"
+                    - --save-every
+                    - "300"
+                    - --seed
+                    - "1224"
+                    - --out-dir
+                    - outputs/healed/calib/glean_keep50_s1224
+                    - --wandb
+                    - --wandb-mode
+                    - online
+                    - --wandb-project
+                    - glean-calib
+                    - --wandb-run-name
+                    - calib_keep50_300
+                codePath: scripts/11_distill_on_policy.py
+                codePathLocal: scripts/11_distill_on_policy.py
+                cpu_count: 24
+                cpu_count_logical: 48
+                cudaVersion: "13.0"
+                disk:
+                    /:
+                        total: "1958315118592"
+                        used: "1668431851520"
+                email: hbfreed@protonmail.com
+                executable: /home/henry/Documents/PythonProjects/variable-reap/.venv/bin/python3
+                git:
+                    commit: 247c7f0113d7a058b1db8044deddfd4456cc998d
+                    remote: https://github.com/hbfreed/variable-reap.git
+                gpu: NVIDIA GeForce RTX 3090
+                gpu_count: 3
+                gpu_nvidia:
+                    - architecture: Ampere
+                      cudaCores: 10496
+                      memoryTotal: "25769803776"
+                      name: NVIDIA GeForce RTX 3090
+                      uuid: GPU-8ca70870-ddf2-d274-bcc5-182c2075bced
+                    - architecture: Ampere
+                      cudaCores: 10496
+                      memoryTotal: "25769803776"
+                      name: NVIDIA GeForce RTX 3090
+                      uuid: GPU-a6acf07f-31f5-618f-a5d0-c0017e7e2e27
+                    - architecture: Ampere
+                      cudaCores: 10496
+                      memoryTotal: "25769803776"
+                      name: NVIDIA GeForce RTX 3090
+                      uuid: GPU-864c54df-0130-7780-e271-8a5551d1733f
+                host: pop-os
+                memory:
+                    total: "134900756480"
+                os: Linux-7.0.11-76070011-generic-x86_64-with-glibc2.35
+                program: /home/henry/Documents/PythonProjects/variable-reap/scripts/11_distill_on_policy.py
+                python: CPython 3.12.12
+                root: outputs/healed/calib/glean_keep50_s1224
+                startedAt: "2026-07-18T05:43:25.786433Z"
+                writerId: v19kdketdy6gkgkcksw06943ad61z958
+        m:
+            - "1": step
+              "6":
+                - 3
+              "7": []
+            - "2": '*'
+              "5": 1
+              "6":
+                - 1
+              "7": []
+        python_version: 3.12.12
+        t:
+            "1":
+                - 1
+                - 5
+                - 11
+                - 49
+                - 51
+                - 53
+                - 71
+            "2":
+                - 1
+                - 5
+                - 11
+                - 49
+                - 51
+                - 53
+                - 71
+            "3":
+                - 2
+                - 7
+                - 13
+                - 16
+            "4": 3.12.12
+            "5": 0.28.0
+            "6": 4.57.6
+            "12": 0.28.0
+            "13": linux-x86_64
+dataset:
+    value: allenai/RLVR-MATH
+dataset_sources:
+    value: null
+debug:
+    value: false
+epochs:
+    value: 9
+eval_every:
+    value: 10
+group_size:
+    value: 1
+gsm8k_batch:
+    value: 16
+gsm8k_every:
+    value: 0
+gsm8k_frames:
+    value: chat
+gsm8k_max_new_tokens:
+    value: 512
+gsm8k_n:
+    value: 256
+kl_direction:
+    value: forward
+loss_tokens_per_step:
+    value: 120000
+lr:
+    value: 3e-05
+max_difficulty:
+    value: null
+max_grad_norm:
+    value: 1
+max_loss_tokens:
+    value: null
+max_new_tokens:
+    value: 256
+max_prompt_len:
+    value: 1024
+micro_batch:
+    value: 3
+no_grad_checkpointing:
+    value: false
+no_teacher_overlap:
+    value: false
+no_wandb_sync:
+    value: false
+off_policy_frames:
+    value: chat
+off_policy_max_seq_len:
+    value: 2048
+optimizer:
+    value: adamw8bit
+out_dir:
+    value: outputs/healed/calib/glean_keep50_s1224
+prompts_per_step:
+    value: 256
+resolved_kl_direction:
+    value: forward
+resume_from:
+    value: outputs/healed/calib/glean_keep50_s1224/step0200
+rollout_batch:
+    value: 64
+rollout_engine:
+    value: hf
+save_every:
+    value: 300
+seed:
+    value: 1224
+start_step:
+    value: 0
+student:
+    value: outputs/pruned/glean-0125inst-general-keep50
+student_device:
+    value: cuda:0
+sweep:
+    value: 300
+sync_checkpoints:
+    value: false
+teacher:
+    value: allenai/OLMoE-1B-7B-0125-Instruct
+teacher_device:
+    value: cuda:0
+topk_targets:
+    value: outputs/teacher_trajectories/dolci_combined_top128
+training_mode:
+    value: off-policy
+trajectories:
+    value: outputs/teacher_trajectories/dolci_math_curated.jsonl
+trajectory_dataset:
+    value: allenai/Dolci-Instruct-RL
+vllm_gpu:
+    value: null
+vllm_gpu_mem_util:
+    value: 0.85
+vllm_live_dir:
+    value: null
+vllm_port:
+    value: 8377
+vllm_refresh_every:
+    value: 5
+vllm_refresh_mode:
+    value: reload
+vllm_serve_bin:
+    value: vllm-plugin/.venv/bin/python
+wandb:
+    value: true
+wandb_mode:
+    value: online
+wandb_project:
+    value: glean-calib
+wandb_resume:
+    value: null
+wandb_run_id:
+    value: null
+wandb_run_name:
+    value: calib_keep50_300
+warmup_steps:
+    value: 10
+weight_decay:
+    value: 0.1

healed/calib/glean_keep50_s1224/wandb/run-20260717_224325-kvt72u16/files/output.log

@@ -0,0 +1,116 @@
+Loading checkpoint shards: 100%|██████████| 2/2 [00:00<00:00,  7.92it/s]
+resumed student weights from outputs/healed/calib/glean_keep50_s1224/step0200 (fresh optimizer, step counter at 0)
+58360 cached top-128 chat trajectories / 22,295,631 unique tokens | 185 steps/epoch | 300 total steps | student params 3.70B | teacher overlap=False
+restored optimizer/scheduler state from step 200; rebuilt 256 paged buffers
+{"step": 201, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.1672117078253068, "tokens": 120000, "cumulative_loss_tokens": 24120000, "grad_norm": 0.400390625, "lr": 3e-05, "finish_rate": 0.924, "comp_len": 434.8, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 53.1, "frames": {"chat": 276}, "mem_gb": 15.9}
+{"step": 202, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.2827735594579019, "tokens": 120000, "cumulative_loss_tokens": 24240000, "grad_norm": 0.474609375, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 337.1, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 58.0, "frames": {"chat": 356}, "mem_gb": 15.95}
+{"step": 203, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.2586930164888501, "tokens": 120000, "cumulative_loss_tokens": 24360000, "grad_norm": 0.48046875, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 394.7, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 48.7, "frames": {"chat": 304}, "mem_gb": 15.77}
+{"step": 204, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.26525049426544767, "tokens": 120000, "cumulative_loss_tokens": 24480000, "grad_norm": 0.490234375, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 320.0, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 58.5, "frames": {"chat": 375}, "mem_gb": 15.68}
+{"step": 205, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.1216446100747368, "tokens": 120000, "cumulative_loss_tokens": 24600000, "grad_norm": 0.3828125, "lr": 3e-05, "finish_rate": 0.939, "comp_len": 408.2, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 48.3, "frames": {"chat": 294}, "mem_gb": 16.04}
+{"step": 206, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.23771027357218166, "tokens": 120000, "cumulative_loss_tokens": 24720000, "grad_norm": 0.447265625, "lr": 3e-05, "finish_rate": 0.98, "comp_len": 401.3, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 52.3, "frames": {"chat": 299}, "mem_gb": 16.05}
+{"step": 207, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.1714890734677979, "tokens": 120000, "cumulative_loss_tokens": 24840000, "grad_norm": 0.400390625, "lr": 3e-05, "finish_rate": 0.98, "comp_len": 393.4, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 51.1, "frames": {"chat": 305}, "mem_gb": 15.83}
+{"step": 208, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.10487559535800635, "tokens": 120000, "cumulative_loss_tokens": 24960000, "grad_norm": 0.310546875, "lr": 3e-05, "finish_rate": 0.931, "comp_len": 458.0, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 43.7, "frames": {"chat": 262}, "mem_gb": 15.88}
+{"step": 209, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.15315499292137102, "tokens": 120000, "cumulative_loss_tokens": 25080000, "grad_norm": 0.38671875, "lr": 3e-05, "finish_rate": 0.95, "comp_len": 396.0, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 46.2, "frames": {"chat": 303}, "mem_gb": 15.94}
+{"step": 210, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.2631195893732831, "tokens": 120000, "cumulative_loss_tokens": 25200000, "grad_norm": 0.51953125, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 315.8, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 56.8, "frames": {"chat": 380}, "mem_gb": 15.82}
+The attention mask is not set and cannot be inferred from input because pad token is same as eos token. As a consequence, you may observe unexpected behavior. Please pass your input's `attention_mask` to obtain reliable results.
+[eval step 210] sample: "To find the rank of a matrix, we need to determine the maximum number of linearly independent rows or columns in the matrix. \n\nLet's represent the given matrix as \\( A \\):\n\n\\[ A = \\begin{bmatrix}\n12 &"
+{"step": 211, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.1175969291699895, "tokens": 120000, "cumulative_loss_tokens": 25320000, "grad_norm": 0.33203125, "lr": 3e-05, "finish_rate": 0.95, "comp_len": 425.5, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 46.1, "frames": {"chat": 282}, "mem_gb": 15.78}
+{"step": 212, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.241187597109812, "tokens": 120000, "cumulative_loss_tokens": 25440000, "grad_norm": 0.455078125, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 331.5, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 58.3, "frames": {"chat": 362}, "mem_gb": 15.79}
+{"step": 213, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.2343245160275294, "tokens": 120000, "cumulative_loss_tokens": 25560000, "grad_norm": 0.431640625, "lr": 3e-05, "finish_rate": 0.967, "comp_len": 390.9, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 51.8, "frames": {"chat": 307}, "mem_gb": 16.04}
+{"step": 214, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.15173778677885422, "tokens": 120000, "cumulative_loss_tokens": 25680000, "grad_norm": 0.388671875, "lr": 3e-05, "finish_rate": 0.926, "comp_len": 425.5, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 46.9, "frames": {"chat": 282}, "mem_gb": 15.9}
+{"step": 215, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.249474069713708, "tokens": 120000, "cumulative_loss_tokens": 25800000, "grad_norm": 0.462890625, "lr": 3e-05, "finish_rate": 0.979, "comp_len": 356.1, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 53.9, "frames": {"chat": 337}, "mem_gb": 15.93}
+{"step": 216, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.14135323118637316, "tokens": 120000, "cumulative_loss_tokens": 25920000, "grad_norm": 0.52734375, "lr": 3e-05, "finish_rate": 0.891, "comp_len": 436.4, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 47.8, "frames": {"chat": 275}, "mem_gb": 15.99}
+{"step": 217, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.26264052670868115, "tokens": 120000, "cumulative_loss_tokens": 26040000, "grad_norm": 0.46875, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 377.4, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 52.8, "frames": {"chat": 318}, "mem_gb": 15.8}
+{"step": 218, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.3055614206787354, "tokens": 120000, "cumulative_loss_tokens": 26160000, "grad_norm": 0.5078125, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 311.7, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 58.4, "frames": {"chat": 385}, "mem_gb": 15.83}
+{"step": 219, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.30217641376073784, "tokens": 120000, "cumulative_loss_tokens": 26280000, "grad_norm": 0.5078125, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 379.7, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 52.3, "frames": {"chat": 316}, "mem_gb": 15.78}
+{"step": 220, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.1054507888781062, "tokens": 120000, "cumulative_loss_tokens": 26400000, "grad_norm": 0.3671875, "lr": 3e-05, "finish_rate": 0.887, "comp_len": 451.1, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 48.2, "frames": {"chat": 266}, "mem_gb": 16.07}
+[eval step 220] sample: "To find the rank of a matrix, we need to determine the maximum number of linearly independent rows or columns in the matrix. \n\nLet's represent the given matrix as \\( A \\):\n\n\\[ A = \\begin{bmatrix}\n12 &"
+{"step": 221, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.2681111897527861, "tokens": 120000, "cumulative_loss_tokens": 26520000, "grad_norm": 0.458984375, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 323.5, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 60.2, "frames": {"chat": 371}, "mem_gb": 15.8}
+{"step": 222, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.13136133254119195, "tokens": 120000, "cumulative_loss_tokens": 26640000, "grad_norm": 0.36328125, "lr": 3e-05, "finish_rate": 0.878, "comp_len": 458.0, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 46.9, "frames": {"chat": 262}, "mem_gb": 16.01}
+{"step": 223, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.2881968457552294, "tokens": 120000, "cumulative_loss_tokens": 26760000, "grad_norm": 0.49609375, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 344.8, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 57.0, "frames": {"chat": 348}, "mem_gb": 15.83}
+{"step": 224, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.04368866277068543, "tokens": 120000, "cumulative_loss_tokens": 26880000, "grad_norm": 0.3046875, "lr": 3e-05, "finish_rate": 0.811, "comp_len": 540.5, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 39.5, "frames": {"chat": 222}, "mem_gb": 16.07}
+{"step": 225, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.24014004462277516, "tokens": 120000, "cumulative_loss_tokens": 27000000, "grad_norm": 0.486328125, "lr": 3e-05, "finish_rate": 0.99, "comp_len": 294.1, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 60.5, "frames": {"chat": 408}, "mem_gb": 15.64}
+{"step": 226, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.23258645670513312, "tokens": 120000, "cumulative_loss_tokens": 27120000, "grad_norm": 0.4453125, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 359.3, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 53.1, "frames": {"chat": 334}, "mem_gb": 15.7}
+{"step": 227, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.11805800152181958, "tokens": 120000, "cumulative_loss_tokens": 27240000, "grad_norm": 0.3359375, "lr": 3e-05, "finish_rate": 0.968, "comp_len": 384.6, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 47.9, "frames": {"chat": 312}, "mem_gb": 15.75}
+{"step": 228, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.23752855263810294, "tokens": 120000, "cumulative_loss_tokens": 27360000, "grad_norm": 0.46484375, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 340.9, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 56.1, "frames": {"chat": 352}, "mem_gb": 15.73}
+{"step": 229, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.20314603109608093, "tokens": 120000, "cumulative_loss_tokens": 27480000, "grad_norm": 0.42578125, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 323.5, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 56.6, "frames": {"chat": 371}, "mem_gb": 15.54}
+{"step": 230, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.13498149184971117, "tokens": 120000, "cumulative_loss_tokens": 27600000, "grad_norm": 0.408203125, "lr": 3e-05, "finish_rate": 0.898, "comp_len": 422.5, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 48.7, "frames": {"chat": 284}, "mem_gb": 15.99}
+[eval step 230] sample: "To find the rank of a matrix, we need to determine the maximum number of linearly independent rows or columns in the matrix. \n\nLet's represent the given matrix as \\( A \\):\n\n\\[ A = \\begin{bmatrix}\n12 &"
+{"step": 231, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.11061588493684928, "tokens": 120000, "cumulative_loss_tokens": 27720000, "grad_norm": 0.349609375, "lr": 3e-05, "finish_rate": 0.923, "comp_len": 418.1, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 45.3, "frames": {"chat": 287}, "mem_gb": 16.06}
+{"step": 232, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.09496309262601038, "tokens": 120000, "cumulative_loss_tokens": 27840000, "grad_norm": 0.314453125, "lr": 3e-05, "finish_rate": 0.916, "comp_len": 421.1, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 45.5, "frames": {"chat": 285}, "mem_gb": 15.95}
+{"step": 233, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.058402567499813936, "tokens": 120000, "cumulative_loss_tokens": 27960000, "grad_norm": 0.263671875, "lr": 3e-05, "finish_rate": 0.873, "comp_len": 476.2, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 42.4, "frames": {"chat": 252}, "mem_gb": 16.04}
+{"step": 234, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.24344681201429727, "tokens": 120000, "cumulative_loss_tokens": 28080000, "grad_norm": 0.486328125, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 333.3, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 55.9, "frames": {"chat": 360}, "mem_gb": 15.83}
+{"step": 235, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.2581301371839363, "tokens": 120000, "cumulative_loss_tokens": 28200000, "grad_norm": 0.4921875, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 334.3, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 55.6, "frames": {"chat": 359}, "mem_gb": 15.82}
+{"step": 236, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.25698635234504086, "tokens": 120000, "cumulative_loss_tokens": 28320000, "grad_norm": 0.5078125, "lr": 3e-05, "finish_rate": 1.0, "comp_len": 330.6, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 56.2, "frames": {"chat": 363}, "mem_gb": 15.58}
+{"step": 237, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.22146621859675894, "tokens": 120000, "cumulative_loss_tokens": 28440000, "grad_norm": 0.453125, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 332.4, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 56.0, "frames": {"chat": 361}, "mem_gb": 15.67}
+{"step": 238, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.13498240058332062, "tokens": 120000, "cumulative_loss_tokens": 28560000, "grad_norm": 0.384765625, "lr": 3e-05, "finish_rate": 0.944, "comp_len": 419.6, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 47.9, "frames": {"chat": 286}, "mem_gb": 15.86}
+{"step": 239, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.25087619634370006, "tokens": 120000, "cumulative_loss_tokens": 28680000, "grad_norm": 0.462890625, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 327.0, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 60.0, "frames": {"chat": 367}, "mem_gb": 15.71}
+{"step": 240, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.2659638972638175, "tokens": 120000, "cumulative_loss_tokens": 28800000, "grad_norm": 0.470703125, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 340.9, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 56.6, "frames": {"chat": 352}, "mem_gb": 15.74}
+[eval step 240] sample: "To find the rank of a matrix, we need to determine the maximum number of linearly independent rows or columns. \n\nLet's start by examining the given matrix:\n\n\\[\n\\begin{bmatrix}\n12 & -16 & 4 & 16 \\\\\n-9 "
+{"step": 241, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.152445907362209, "tokens": 120000, "cumulative_loss_tokens": 28920000, "grad_norm": 0.37890625, "lr": 3e-05, "finish_rate": 0.962, "comp_len": 382.2, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 50.0, "frames": {"chat": 314}, "mem_gb": 15.69}
+{"step": 242, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.23061156271438424, "tokens": 120000, "cumulative_loss_tokens": 29040000, "grad_norm": 0.47265625, "lr": 3e-05, "finish_rate": 0.995, "comp_len": 322.6, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 57.6, "frames": {"chat": 372}, "mem_gb": 15.66}
+{"step": 243, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.2019635400752537, "tokens": 120000, "cumulative_loss_tokens": 29160000, "grad_norm": 0.4140625, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 359.3, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 56.9, "frames": {"chat": 334}, "mem_gb": 15.81}
+{"step": 244, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.2732710076200077, "tokens": 120000, "cumulative_loss_tokens": 29280000, "grad_norm": 0.71875, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 326.1, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 57.0, "frames": {"chat": 368}, "mem_gb": 15.75}
+{"step": 245, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.20303159307703997, "tokens": 120000, "cumulative_loss_tokens": 29400000, "grad_norm": 0.42578125, "lr": 3e-05, "finish_rate": 0.994, "comp_len": 339.0, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 57.3, "frames": {"chat": 354}, "mem_gb": 15.9}
+{"step": 246, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.11647335124739135, "tokens": 120000, "cumulative_loss_tokens": 29520000, "grad_norm": 0.40625, "lr": 3e-05, "finish_rate": 0.973, "comp_len": 409.6, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 46.8, "frames": {"chat": 293}, "mem_gb": 15.79}
+{"step": 247, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.0634124994080669, "tokens": 120000, "cumulative_loss_tokens": 29640000, "grad_norm": 0.37890625, "lr": 3e-05, "finish_rate": 0.835, "comp_len": 519.5, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 42.7, "frames": {"chat": 231}, "mem_gb": 15.95}
+{"step": 248, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.18457725901436062, "tokens": 120000, "cumulative_loss_tokens": 29760000, "grad_norm": 0.466796875, "lr": 3e-05, "finish_rate": 0.964, "comp_len": 397.4, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 49.0, "frames": {"chat": 302}, "mem_gb": 15.92}
+{"step": 249, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.12263697761416746, "tokens": 120000, "cumulative_loss_tokens": 29880000, "grad_norm": 0.333984375, "lr": 3e-05, "finish_rate": 0.878, "comp_len": 456.3, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 48.3, "frames": {"chat": 263}, "mem_gb": 16.06}
+{"step": 250, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.13431983752972446, "tokens": 120000, "cumulative_loss_tokens": 30000000, "grad_norm": 0.361328125, "lr": 3e-05, "finish_rate": 0.889, "comp_len": 476.2, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 44.4, "frames": {"chat": 252}, "mem_gb": 15.99}
+[eval step 250] sample: "To compute the rank of a matrix, we need to determine the maximum number of linearly independent rows or columns in the matrix. \n\nLet's represent the given matrix \\( A \\) as:\n\\[ A = \\begin{bmatrix}\n12"
+{"step": 251, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.19803441493669524, "tokens": 120000, "cumulative_loss_tokens": 30120000, "grad_norm": 0.458984375, "lr": 3e-05, "finish_rate": 0.968, "comp_len": 348.8, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 52.8, "frames": {"chat": 344}, "mem_gb": 16.02}
+{"step": 252, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.1865262628025841, "tokens": 120000, "cumulative_loss_tokens": 30240000, "grad_norm": 0.455078125, "lr": 3e-05, "finish_rate": 0.981, "comp_len": 383.4, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 52.3, "frames": {"chat": 313}, "mem_gb": 15.95}
+{"step": 253, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.22638036701406042, "tokens": 120000, "cumulative_loss_tokens": 30360000, "grad_norm": 0.4609375, "lr": 3e-05, "finish_rate": 0.992, "comp_len": 335.2, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 53.6, "frames": {"chat": 358}, "mem_gb": 15.65}
+{"step": 254, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.10856965826541806, "tokens": 120000, "cumulative_loss_tokens": 30480000, "grad_norm": 0.349609375, "lr": 3e-05, "finish_rate": 0.951, "comp_len": 449.4, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 42.9, "frames": {"chat": 267}, "mem_gb": 15.87}
+{"step": 255, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.2210595068858781, "tokens": 120000, "cumulative_loss_tokens": 30600000, "grad_norm": 0.453125, "lr": 3e-05, "finish_rate": 0.98, "comp_len": 400.0, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 48.6, "frames": {"chat": 300}, "mem_gb": 15.74}
+{"step": 256, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.18661262825395292, "tokens": 120000, "cumulative_loss_tokens": 30720000, "grad_norm": 0.4375, "lr": 3e-05, "finish_rate": 0.994, "comp_len": 344.8, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 52.8, "frames": {"chat": 348}, "mem_gb": 15.68}
+{"step": 257, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.23165423854584805, "tokens": 120000, "cumulative_loss_tokens": 30840000, "grad_norm": 0.451171875, "lr": 3e-05, "finish_rate": 0.988, "comp_len": 364.7, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 52.6, "frames": {"chat": 329}, "mem_gb": 15.77}
+{"step": 258, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.11868625700032959, "tokens": 120000, "cumulative_loss_tokens": 30960000, "grad_norm": 0.34765625, "lr": 3e-05, "finish_rate": 0.948, "comp_len": 446.1, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 45.6, "frames": {"chat": 269}, "mem_gb": 15.93}
+{"step": 259, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.24212703010075104, "tokens": 120000, "cumulative_loss_tokens": 31080000, "grad_norm": 0.470703125, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 335.2, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 55.1, "frames": {"chat": 358}, "mem_gb": 15.6}
+{"step": 260, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.1401378205653125, "tokens": 120000, "cumulative_loss_tokens": 31200000, "grad_norm": 0.375, "lr": 3e-05, "finish_rate": 0.96, "comp_len": 397.4, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 48.5, "frames": {"chat": 302}, "mem_gb": 15.86}
+[eval step 260] sample: 'To compute the rank of a matrix, we need to determine the maximum number of linearly independent rows or columns in the matrix. \n\nGiven the matrix:\n\n\\[\n\\begin{bmatrix}\n12 & -16 & 4 & 16 \\\\\n-9 & 11 & -'
+{"step": 261, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.2441046292551638, "tokens": 120000, "cumulative_loss_tokens": 31320000, "grad_norm": 0.458984375, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 315.0, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 56.7, "frames": {"chat": 381}, "mem_gb": 15.87}
+{"step": 262, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.26715706478791934, "tokens": 120000, "cumulative_loss_tokens": 31440000, "grad_norm": 0.50390625, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 316.6, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 60.8, "frames": {"chat": 379}, "mem_gb": 15.82}
+{"step": 263, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.26412971248012035, "tokens": 120000, "cumulative_loss_tokens": 31560000, "grad_norm": 0.6171875, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 341.9, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 56.0, "frames": {"chat": 351}, "mem_gb": 15.64}
+{"step": 264, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.21945662072741737, "tokens": 120000, "cumulative_loss_tokens": 31680000, "grad_norm": 0.4140625, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 363.6, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 53.5, "frames": {"chat": 330}, "mem_gb": 15.84}
+{"step": 265, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.22621997363331417, "tokens": 120000, "cumulative_loss_tokens": 31800000, "grad_norm": 0.45703125, "lr": 3e-05, "finish_rate": 0.991, "comp_len": 379.7, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 50.5, "frames": {"chat": 316}, "mem_gb": 15.87}
+{"step": 266, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.1477637799535102, "tokens": 120000, "cumulative_loss_tokens": 31920000, "grad_norm": 0.451171875, "lr": 3e-05, "finish_rate": 0.891, "comp_len": 434.8, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 49.3, "frames": {"chat": 276}, "mem_gb": 16.05}
+{"step": 267, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.12686351013585614, "tokens": 120000, "cumulative_loss_tokens": 32040000, "grad_norm": 0.408203125, "lr": 3e-05, "finish_rate": 0.833, "comp_len": 500.0, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 44.9, "frames": {"chat": 240}, "mem_gb": 15.99}
+{"step": 268, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.1362769400940432, "tokens": 120000, "cumulative_loss_tokens": 32160000, "grad_norm": 0.349609375, "lr": 3e-05, "finish_rate": 0.948, "comp_len": 416.7, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 47.7, "frames": {"chat": 288}, "mem_gb": 15.9}
+{"step": 269, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.12843327870285137, "tokens": 120000, "cumulative_loss_tokens": 32280000, "grad_norm": 0.36328125, "lr": 3e-05, "finish_rate": 0.937, "comp_len": 442.8, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 46.5, "frames": {"chat": 271}, "mem_gb": 15.89}
+{"step": 270, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.2064100126405557, "tokens": 120000, "cumulative_loss_tokens": 32400000, "grad_norm": 0.47265625, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 336.1, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 59.5, "frames": {"chat": 357}, "mem_gb": 15.84}
+[eval step 270] sample: 'To compute the rank of a matrix, we need to determine the maximum number of linearly independent rows or columns in the matrix. A matrix is said to be of rank \\( r \\) if it has \\( r \\) linearly indepe'
+{"step": 271, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.2752586573896774, "tokens": 120000, "cumulative_loss_tokens": 32520000, "grad_norm": 0.52734375, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 339.0, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 57.4, "frames": {"chat": 354}, "mem_gb": 15.81}
+{"step": 272, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.2640283962545606, "tokens": 120000, "cumulative_loss_tokens": 32640000, "grad_norm": 0.453125, "lr": 3e-05, "finish_rate": 0.927, "comp_len": 397.4, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 51.0, "frames": {"chat": 302}, "mem_gb": 16.05}
+{"step": 273, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.1916471611869211, "tokens": 120000, "cumulative_loss_tokens": 32760000, "grad_norm": 0.412109375, "lr": 3e-05, "finish_rate": 0.923, "comp_len": 404.0, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 52.2, "frames": {"chat": 297}, "mem_gb": 16.04}
+{"step": 274, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.0575785630961104, "tokens": 120000, "cumulative_loss_tokens": 32880000, "grad_norm": 0.31640625, "lr": 3e-05, "finish_rate": 0.799, "comp_len": 547.9, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 39.9, "frames": {"chat": 219}, "mem_gb": 16.06}
+{"step": 275, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.2618522337922516, "tokens": 120000, "cumulative_loss_tokens": 33000000, "grad_norm": 0.44140625, "lr": 3e-05, "finish_rate": 0.99, "comp_len": 392.2, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 51.5, "frames": {"chat": 306}, "mem_gb": 15.76}
+{"step": 276, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.2590624229635112, "tokens": 120000, "cumulative_loss_tokens": 33120000, "grad_norm": 0.486328125, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 349.9, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 54.0, "frames": {"chat": 343}, "mem_gb": 15.7}
+{"step": 277, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.23909416454567883, "tokens": 120000, "cumulative_loss_tokens": 33240000, "grad_norm": 0.470703125, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 351.9, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 56.5, "frames": {"chat": 341}, "mem_gb": 15.81}
+{"step": 278, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.15004109006367314, "tokens": 120000, "cumulative_loss_tokens": 33360000, "grad_norm": 0.42578125, "lr": 3e-05, "finish_rate": 0.965, "comp_len": 377.4, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 50.4, "frames": {"chat": 318}, "mem_gb": 15.76}
+{"step": 279, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.2516725248177846, "tokens": 120000, "cumulative_loss_tokens": 33480000, "grad_norm": 0.48046875, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 346.8, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 53.3, "frames": {"chat": 346}, "mem_gb": 15.65}
+{"step": 280, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.1229936876871235, "tokens": 120000, "cumulative_loss_tokens": 33600000, "grad_norm": 0.376953125, "lr": 3e-05, "finish_rate": 0.841, "comp_len": 433.2, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 47.3, "frames": {"chat": 277}, "mem_gb": 16.13}
+[eval step 280] sample: 'To find the rank of a matrix, we need to determine the maximum number of linearly independent rows or columns. In this 4x4 matrix:\n\n\\[\n\\begin{bmatrix}\n12 & -16 & 4 & 16 \\\\\n-9 & 11 & -1 & -10 \\\\\n0 & 1'
+{"step": 281, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.11821835112910097, "tokens": 120000, "cumulative_loss_tokens": 33720000, "grad_norm": 0.333984375, "lr": 3e-05, "finish_rate": 0.956, "comp_len": 444.4, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 46.5, "frames": {"chat": 270}, "mem_gb": 15.86}
+{"step": 282, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.15829591915813895, "tokens": 120000, "cumulative_loss_tokens": 33840000, "grad_norm": 0.37109375, "lr": 3e-05, "finish_rate": 0.903, "comp_len": 466.9, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 45.3, "frames": {"chat": 257}, "mem_gb": 16.04}
+{"step": 283, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.24422800888229781, "tokens": 120000, "cumulative_loss_tokens": 33960000, "grad_norm": 0.478515625, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 330.6, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 55.5, "frames": {"chat": 363}, "mem_gb": 15.65}
+{"step": 284, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.25823034505227893, "tokens": 120000, "cumulative_loss_tokens": 34080000, "grad_norm": 0.474609375, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 337.1, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 54.8, "frames": {"chat": 356}, "mem_gb": 15.79}
+{"step": 285, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.2246625458458904, "tokens": 120000, "cumulative_loss_tokens": 34200000, "grad_norm": 0.416015625, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 371.5, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 52.0, "frames": {"chat": 323}, "mem_gb": 15.82}
+{"step": 286, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.10769453969344807, "tokens": 120000, "cumulative_loss_tokens": 34320000, "grad_norm": 0.341796875, "lr": 3e-05, "finish_rate": 0.868, "comp_len": 493.8, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 42.9, "frames": {"chat": 243}, "mem_gb": 16.07}
+{"step": 287, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.22839496972688164, "tokens": 120000, "cumulative_loss_tokens": 34440000, "grad_norm": 0.4375, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 390.9, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 52.9, "frames": {"chat": 307}, "mem_gb": 15.75}
+{"step": 288, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.21778279120173927, "tokens": 120000, "cumulative_loss_tokens": 34560000, "grad_norm": 0.416015625, "lr": 3e-05, "finish_rate": 0.988, "comp_len": 346.8, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 52.8, "frames": {"chat": 346}, "mem_gb": 15.99}
+{"step": 289, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.1388638691237817, "tokens": 120000, "cumulative_loss_tokens": 34680000, "grad_norm": 0.408203125, "lr": 3e-05, "finish_rate": 0.905, "comp_len": 439.6, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 46.7, "frames": {"chat": 273}, "mem_gb": 15.99}
+{"step": 290, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.14722677345586319, "tokens": 120000, "cumulative_loss_tokens": 34800000, "grad_norm": 0.375, "lr": 3e-05, "finish_rate": 0.957, "comp_len": 397.4, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 49.3, "frames": {"chat": 302}, "mem_gb": 15.92}
+[eval step 290] sample: 'To find the rank of a matrix, we need to determine the maximum number of linearly independent rows or columns. In this 4x4 matrix:\n\n\\[\n\\begin{bmatrix}\n12 & -16 & 4 & 16 \\\\\n-9 & 11 & -1 & -10 \\\\\n0 & 1'
+{"step": 291, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.22137576552818838, "tokens": 120000, "cumulative_loss_tokens": 34920000, "grad_norm": 0.443359375, "lr": 3e-05, "finish_rate": 0.97, "comp_len": 356.1, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 50.6, "frames": {"chat": 337}, "mem_gb": 16.05}
+{"step": 292, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.16152701778401, "tokens": 120000, "cumulative_loss_tokens": 35040000, "grad_norm": 0.3671875, "lr": 3e-05, "finish_rate": 0.914, "comp_len": 428.6, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 46.5, "frames": {"chat": 280}, "mem_gb": 15.93}
+{"step": 293, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.2765374315341003, "tokens": 120000, "cumulative_loss_tokens": 35160000, "grad_norm": 0.52734375, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 362.5, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 52.8, "frames": {"chat": 331}, "mem_gb": 15.83}
+{"step": 294, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.1624435577841165, "tokens": 120000, "cumulative_loss_tokens": 35280000, "grad_norm": 0.388671875, "lr": 3e-05, "finish_rate": 0.959, "comp_len": 382.2, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 47.3, "frames": {"chat": 314}, "mem_gb": 15.93}
+{"step": 295, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.2818843329635759, "tokens": 120000, "cumulative_loss_tokens": 35400000, "grad_norm": 0.486328125, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 408.2, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 49.3, "frames": {"chat": 294}, "mem_gb": 15.86}
+{"step": 296, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.12327816759125951, "tokens": 120000, "cumulative_loss_tokens": 35520000, "grad_norm": 0.361328125, "lr": 3e-05, "finish_rate": 0.894, "comp_len": 438.0, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 46.4, "frames": {"chat": 274}, "mem_gb": 15.99}
+{"step": 297, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.2160477212319461, "tokens": 120000, "cumulative_loss_tokens": 35640000, "grad_norm": 0.474609375, "lr": 3e-05, "finish_rate": 0.995, "comp_len": 328.8, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 56.6, "frames": {"chat": 365}, "mem_gb": 15.71}
+{"step": 298, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.1336963133474191, "tokens": 120000, "cumulative_loss_tokens": 35760000, "grad_norm": 0.36328125, "lr": 3e-05, "finish_rate": 0.972, "comp_len": 371.5, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 50.9, "frames": {"chat": 323}, "mem_gb": 15.77}
+{"step": 299, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.09222794443805081, "tokens": 120000, "cumulative_loss_tokens": 35880000, "grad_norm": 0.4296875, "lr": 3e-05, "finish_rate": 0.829, "comp_len": 512.8, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 42.0, "frames": {"chat": 234}, "mem_gb": 16.06}
+{"step": 300, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.12722712953349885, "tokens": 120000, "cumulative_loss_tokens": 36000000, "grad_norm": 0.365234375, "lr": 3e-05, "finish_rate": 0.891, "comp_len": 449.4, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 44.0, "frames": {"chat": 267}, "mem_gb": 16.06}
+[eval step 300] sample: "To find the rank of a matrix, we need to determine the maximum number of linearly independent rows or columns in the matrix. \n\nLet's represent the given 4x4 matrix \\( A \\) as follows:\n\\[ A = \\begin{pm"
+checkpoint snapshot queued -> outputs/healed/calib/glean_keep50_s1224/step0300

healed/calib/glean_keep50_s1224/wandb/run-20260717_224325-kvt72u16/files/requirements.txt

@@ -0,0 +1,129 @@
+certifi==2026.6.17
+fonttools==4.63.0
+urllib3==2.7.0
+requests==2.34.2
+nvidia-cublas-cu12==12.6.4.1
+packaging==26.2
+regex==2026.6.28
+portalocker==3.2.0
+safetensors==0.8.0
+datasets==5.0.0
+sqlitedict==2.1.0
+narwhals==2.23.0
+latex2sympy2_extended==1.11.0
+smmap==5.0.3
+colorama==0.4.6
+anyio==4.14.1
+cycler==0.12.1
+pytablewriter==1.2.1
+pydantic_core==2.46.4
+huggingface_hub==0.36.2
+charset-normalizer==3.4.8
+nvidia-cusolver-cu12==11.7.1.2
+joblib==1.5.3
+typing-inspection==0.4.2
+wandb==0.28.0
+bitsandbytes==0.49.2
+kiwisolver==1.5.0
+cuda-toolkit==12.6.3
+propcache==0.5.2
+Jinja2==3.1.6
+cuda-bindings==12.9.7
+pluggy==1.6.0
+pytest==9.1.1
+fsspec==2026.4.0
+tqdm==4.68.3
+psutil==7.2.2
+pydantic==2.13.4
+typepy==1.3.5
+xxhash==3.8.1
+DataProperty==1.1.1
+aiosignal==1.4.0
+annotated-types==0.7.0
+Pygments==2.20.0
+zstandard==0.25.0
+threadpoolctl==3.6.0
+nvidia-cudnn-cu12==9.10.2.21
+evaluate==0.4.6
+pillow==12.3.0
+GitPython==3.1.50
+httpx==0.28.1
+lm_eval==0.4.12
+pytz==2026.2
+mbstrdecoder==1.1.5
+immutabledict==4.3.1
+dill==0.4.1
+mpmath==1.3.0
+yarl==1.24.2
+word2number==1.1
+networkx==3.6.1
+einops==0.8.2
+tabulate==0.10.0
+pathvalidate==3.3.1
+tenacity==9.1.4
+sentry-sdk==2.64.0
+nvidia-nvshmem-cu12==3.4.5
+hf-xet==1.5.1
+six==1.17.0
+pyarrow==24.0.0
+nvidia-curand-cu12==10.3.7.77
+cuda-pathfinder==1.5.6
+sacrebleu==2.6.0
+triton==3.7.1
+frozenlist==1.8.0
+antlr4-python3-runtime==4.11.0
+tabledata==1.3.5
+h11==0.16.0
+nvidia-cusparselt-cu12==0.7.1
+nvidia-nvjitlink-cu12==12.6.85
+scipy==1.18.0
+rouge_score==0.1.2
+transformers==4.57.6
+multiprocess==0.70.19
+absl-py==2.5.0
+accelerate==1.14.0
+click==8.4.2
+lxml==6.1.1
+defusedxml==0.7.1
+setuptools==81.0.0
+nvidia-cuda-cupti-cu12==12.6.80
+httpcore==1.0.9
+langdetect==1.0.9
+nltk==3.10.0
+iniconfig==2.3.0
+nvidia-cuda-runtime-cu12==12.6.77
+nvidia-cusparse-cu12==12.5.4.2
+matplotlib==3.11.0
+protobuf==7.35.1
+nvidia-cufft-cu12==11.3.0.4
+pyparsing==3.3.2
+chardet==6.0.0.post1
+typing_extensions==4.16.0
+stanford-stk==0.7.1
+aiohappyeyeballs==2.7.1
+MarkupSafe==3.0.3
+nvidia-cufile-cu12==1.11.1.6
+tcolorpy==0.1.7
+scikit-learn==1.9.0
+tokenizers==0.22.2
+attrs==26.1.0
+gitdb==4.0.12
+aiohttp==3.14.1
+glean==0.0.1
+platformdirs==4.10.0
+megablocks==0.11.0.dev0
+nvidia-cuda-nvrtc-cu12==12.6.85
+pandas==3.0.3
+multidict==6.7.1
+torch==2.12.1+cu126
+contourpy==1.3.3
+nvidia-nvtx-cu12==12.6.77
+math-verify==0.9.0
+filelock==3.29.5
+idna==3.18
+nvidia-nccl-cu12==2.29.3
+python-dateutil==2.9.0.post0
+more-itertools==11.1.0
+numpy==2.0.2
+sympy==1.14.0
+PyYAML==6.0.3

healed/calib/glean_keep50_s1224/wandb/run-20260717_224325-kvt72u16/files/wandb-metadata.json

@@ -0,0 +1,97 @@
+{
+  "os": "Linux-7.0.11-76070011-generic-x86_64-with-glibc2.35",
+  "python": "CPython 3.12.12",
+  "startedAt": "2026-07-18T05:43:25.786433Z",
+  "args": [
+    "--training-mode",
+    "off-policy",
+    "--kl-direction",
+    "forward",
+    "--topk-targets",
+    "outputs/teacher_trajectories/dolci_combined_top128",
+    "--off-policy-frames",
+    "chat",
+    "--student-device",
+    "cuda:0",
+    "--student",
+    "outputs/pruned/glean-0125inst-general-keep50",
+    "--resume-from",
+    "outputs/healed/calib/glean_keep50_s1224/step0200",
+    "--lr",
+    "3e-5",
+    "--optimizer",
+    "adamw8bit",
+    "--epochs",
+    "9",
+    "--sweep",
+    "300",
+    "--micro-batch",
+    "3",
+    "--loss-tokens-per-step",
+    "120000",
+    "--gsm8k-every",
+    "0",
+    "--save-every",
+    "300",
+    "--seed",
+    "1224",
+    "--out-dir",
+    "outputs/healed/calib/glean_keep50_s1224",
+    "--wandb",
+    "--wandb-mode",
+    "online",
+    "--wandb-project",
+    "glean-calib",
+    "--wandb-run-name",
+    "calib_keep50_300"
+  ],
+  "program": "/home/henry/Documents/PythonProjects/variable-reap/scripts/11_distill_on_policy.py",
+  "codePath": "scripts/11_distill_on_policy.py",
+  "codePathLocal": "scripts/11_distill_on_policy.py",
+  "git": {
+    "remote": "https://github.com/hbfreed/variable-reap.git",
+    "commit": "247c7f0113d7a058b1db8044deddfd4456cc998d"
+  },
+  "email": "hbfreed@protonmail.com",
+  "root": "outputs/healed/calib/glean_keep50_s1224",
+  "host": "pop-os",
+  "executable": "/home/henry/Documents/PythonProjects/variable-reap/.venv/bin/python3",
+  "cpu_count": 24,
+  "cpu_count_logical": 48,
+  "gpu": "NVIDIA GeForce RTX 3090",
+  "gpu_count": 3,
+  "disk": {
+    "/": {
+      "total": "1958315118592",
+      "used": "1668431851520"
+    }
+  },
+  "memory": {
+    "total": "134900756480"
+  },
+  "gpu_nvidia": [
+    {
+      "name": "NVIDIA GeForce RTX 3090",
+      "memoryTotal": "25769803776",
+      "cudaCores": 10496,
+      "architecture": "Ampere",
+      "uuid": "GPU-8ca70870-ddf2-d274-bcc5-182c2075bced"
+    },
+    {
+      "name": "NVIDIA GeForce RTX 3090",
+      "memoryTotal": "25769803776",
+      "cudaCores": 10496,
+      "architecture": "Ampere",
+      "uuid": "GPU-a6acf07f-31f5-618f-a5d0-c0017e7e2e27"
+    },
+    {
+      "name": "NVIDIA GeForce RTX 3090",
+      "memoryTotal": "25769803776",
+      "cudaCores": 10496,
+      "architecture": "Ampere",
+      "uuid": "GPU-864c54df-0130-7780-e271-8a5551d1733f"
+    }
+  ],
+  "cudaVersion": "13.0",
+  "writerId": "v19kdketdy6gkgkcksw06943ad61z958"
+}

healed/calib/glean_keep50_s1224/wandb/run-20260717_224325-kvt72u16/files/wandb-summary.json

@@ -0,0 +1 @@
+{"t_data_s":0,"comp_len":449.4,"step":300,"frames":{"chat":267},"cumulative_loss_tokens":36000000,"_step":99,"grad_norm":0.365234375,"t_step_s":44,"finish_rate":0.891,"tokens":120000,"mem_gb":16.06,"epoch":1,"training_mode":"off-policy","lr":3e-05,"t_rollout_s":0,"_wandb":{"runtime":5186},"_timestamp":1.7843585644060068e+09,"_runtime":5186,"forward_topk_kl":0.12722712953349885}

healed/calib/glean_keep50_s1224/wandb/run-20260717_224325-kvt72u16/logs/debug-core.log

@@ -0,0 +1,20 @@
+{"time":"2026-07-17T22:43:25.45660312-07:00","level":"INFO","msg":"main: starting server","port-filename":"/tmp/tmpe88yxig2/port-414016.txt","pid":414016,"detached":false,"idle-timeout":600000000000,"log-level":0,"disable-analytics":false,"shutdown-on-parent-exit":false,"enable-dcgm-profiling":false}
+{"time":"2026-07-17T22:43:25.457476459-07:00","level":"INFO","msg":"server: will exit if parent process dies","ppid":414016}
+{"time":"2026-07-17T22:43:25.457419897-07:00","level":"INFO","msg":"server: accepting connections","addr":{"Name":"/tmp/wandb-414016-414126-683263083/socket","Net":"unix"}}
+{"time":"2026-07-17T22:43:25.641547645-07:00","level":"INFO","msg":"connection: ManageConnectionData: new connection created","id":"1(@)"}
+{"time":"2026-07-17T22:43:25.790393144-07:00","level":"INFO","msg":"handleInformInit: received","streamId":"kvt72u16","id":"1(@)"}
+{"time":"2026-07-17T22:43:26.012263992-07:00","level":"INFO","msg":"handleInformInit: stream started","streamId":"kvt72u16","id":"1(@)"}
+{"time":"2026-07-17T22:43:31.433003228-07:00","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"abp5w6dr7qbh"}
+{"time":"2026-07-18T00:09:53.025678913-07:00","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"abp5w6dr7qbh"}
+{"time":"2026-07-18T00:09:53.93028554-07:00","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"abp5w6dr7qbh"}
+{"time":"2026-07-18T00:09:53.933179485-07:00","level":"INFO","msg":"handleInformFinish: finish message received","streamId":"kvt72u16","id":"1(@)"}
+{"time":"2026-07-18T00:09:53.938049756-07:00","level":"INFO","msg":"handleInformFinish: stream closed","streamId":"kvt72u16","id":"1(@)"}
+{"time":"2026-07-18T00:09:54.793162622-07:00","level":"INFO","msg":"handleInformTeardown: server teardown initiated","id":"1(@)"}
+{"time":"2026-07-18T00:09:54.793204993-07:00","level":"INFO","msg":"handleInformTeardown: server shutdown complete","id":"1(@)"}
+{"time":"2026-07-18T00:09:54.793212663-07:00","level":"INFO","msg":"connection: closing","id":"1(@)"}
+{"time":"2026-07-18T00:09:54.793230924-07:00","level":"INFO","msg":"server is shutting down"}
+{"time":"2026-07-18T00:09:54.793254745-07:00","level":"INFO","msg":"connection: closed successfully","id":"1(@)"}
+{"time":"2026-07-18T00:09:54.793256745-07:00","level":"INFO","msg":"processOutgoingData: finished","id":"1(@)"}
+{"time":"2026-07-18T00:09:54.793290266-07:00","level":"INFO","msg":"connection: ManageConnectionData: connection closed","id":"1(@)"}
+{"time":"2026-07-18T00:09:54.793393689-07:00","level":"INFO","msg":"server: listener closed","addr":{"Name":"/tmp/wandb-414016-414126-683263083/socket","Net":"unix"}}
+{"time":"2026-07-18T00:09:54.793446271-07:00","level":"INFO","msg":"server is closed"}

healed/calib/glean_keep50_s1224/wandb/run-20260717_224325-kvt72u16/logs/debug.log

@@ -0,0 +1,24 @@
+2026-07-17 22:43:25,788 INFO    MainThread:414016 [wandb_setup.py:_flush():81] Current SDK version is 0.28.0
+2026-07-17 22:43:25,788 INFO    MainThread:414016 [wandb_setup.py:_flush():81] Configure stats pid to 414016
+2026-07-17 22:43:25,788 INFO    MainThread:414016 [wandb_setup.py:_flush():81] Loading settings from environment variables
+2026-07-17 22:43:25,788 INFO    MainThread:414016 [wandb_init.py:setup_run_log_directory():725] Logging user logs to outputs/healed/calib/glean_keep50_s1224/wandb/run-20260717_224325-kvt72u16/logs/debug.log
+2026-07-17 22:43:25,788 INFO    MainThread:414016 [wandb_init.py:setup_run_log_directory():726] Logging internal logs to outputs/healed/calib/glean_keep50_s1224/wandb/run-20260717_224325-kvt72u16/logs/debug-internal.log
+2026-07-17 22:43:25,788 INFO    MainThread:414016 [wandb_init.py:init():768] calling init triggers
+2026-07-17 22:43:25,788 INFO    MainThread:414016 [wandb_init.py:init():773] wandb.init called with sweep_config: {}
+config: {'student': 'outputs/pruned/glean-0125inst-general-keep50', 'teacher': 'allenai/OLMoE-1B-7B-0125-Instruct', 'training_mode': 'off-policy', 'kl_direction': 'forward', 'dataset': 'allenai/RLVR-MATH', 'dataset_sources': None, 'max_difficulty': None, 'trajectories': 'outputs/teacher_trajectories/dolci_math_curated.jsonl', 'trajectory_dataset': 'allenai/Dolci-Instruct-RL', 'off_policy_frames': 'chat', 'off_policy_max_seq_len': 2048, 'topk_targets': 'outputs/teacher_trajectories/dolci_combined_top128', 'max_loss_tokens': None, 'loss_tokens_per_step': 120000, 'teacher_device': 'cuda:0', 'student_device': 'cuda:0', 'lr': 3e-05, 'optimizer': 'adamw8bit', 'weight_decay': 0.1, 'epochs': 9, 'prompts_per_step': 256, 'group_size': 1, 'rollout_batch': 64, 'micro_batch': 3, 'max_new_tokens': 256, 'max_prompt_len': 1024, 'warmup_steps': 10, 'max_grad_norm': 1.0, 'eval_every': 10, 'gsm8k_every': 0, 'gsm8k_n': 256, 'gsm8k_batch': 16, 'gsm8k_max_new_tokens': 512, 'gsm8k_frames': 'chat', 'save_every': 300, 'out_dir': 'outputs/healed/calib/glean_keep50_s1224', 'sweep': 300, 'wandb': True, 'wandb_project': 'glean-calib', 'wandb_run_name': 'calib_keep50_300', 'wandb_run_id': None, 'wandb_resume': None, 'wandb_mode': 'online', 'no_wandb_sync': False, 'debug': False, 'resume_from': 'outputs/healed/calib/glean_keep50_s1224/step0200', 'start_step': 0, 'no_grad_checkpointing': False, 'seed': 1224, 'no_teacher_overlap': False, 'sync_checkpoints': False, 'rollout_engine': 'hf', 'vllm_gpu': None, 'vllm_port': 8377, 'vllm_refresh_every': 5, 'vllm_serve_bin': 'vllm-plugin/.venv/bin/python', 'vllm_gpu_mem_util': 0.85, 'vllm_refresh_mode': 'reload', 'vllm_live_dir': None, 'resolved_kl_direction': 'forward', '_wandb': {}}
+2026-07-17 22:43:25,788 INFO    MainThread:414016 [wandb_init.py:init():816] starting backend
+2026-07-17 22:43:25,788 INFO    MainThread:414016 [wandb_init.py:init():831] sending inform_init request
+2026-07-17 22:43:26,012 INFO    MainThread:414016 [wandb_init.py:init():836] backend started and connected
+2026-07-17 22:43:26,015 INFO    MainThread:414016 [wandb_init.py:init():906] updated telemetry
+2026-07-17 22:43:26,022 INFO    MainThread:414016 [wandb_init.py:init():929] communicating run to backend with 90.0 second timeout
+2026-07-17 22:43:26,281 INFO    MainThread:414016 [wandb_init.py:init():974] starting run threads in backend
+2026-07-17 22:43:26,412 INFO    MainThread:414016 [wandb_run.py:_console_start():2523] atexit reg
+2026-07-17 22:43:26,412 INFO    MainThread:414016 [wandb_run.py:_redirect():2373] redirect: wrap_raw
+2026-07-17 22:43:26,412 INFO    MainThread:414016 [wandb_run.py:_redirect():2442] Wrapping output streams.
+2026-07-17 22:43:26,412 INFO    MainThread:414016 [wandb_run.py:_redirect():2465] Redirects installed.
+2026-07-17 22:43:26,414 INFO    MainThread:414016 [wandb_init.py:init():1012] run started, returning control to user process
+2026-07-18 00:09:53,022 INFO    MainThread:414016 [wandb_run.py:_finish():2285] finishing run hbfreed/glean-calib/kvt72u16
+2026-07-18 00:09:53,024 INFO    MainThread:414016 [wandb_run.py:_atexit_cleanup():2490] got exitcode: 0
+2026-07-18 00:09:53,025 INFO    MainThread:414016 [wandb_run.py:_restore():2472] restore
+2026-07-18 00:09:53,025 INFO    MainThread:414016 [wandb_run.py:_restore():2478] restore done
+2026-07-18 00:09:53,932 INFO    MainThread:414016 [wandb_run.py:_footer_sync_info():3895] logging synced files

healed/calib/glean_keep75_s1224/args.json

@@ -0,0 +1,62 @@
+{
+  "student": "outputs/pruned/glean-0125inst-general-keep75",
+  "teacher": "allenai/OLMoE-1B-7B-0125-Instruct",
+  "training_mode": "off-policy",
+  "kl_direction": "forward",
+  "dataset": "allenai/RLVR-MATH",
+  "dataset_sources": null,
+  "max_difficulty": null,
+  "trajectories": "outputs/teacher_trajectories/dolci_math_curated.jsonl",
+  "trajectory_dataset": "allenai/Dolci-Instruct-RL",
+  "off_policy_frames": "chat",
+  "off_policy_max_seq_len": 2048,
+  "topk_targets": "outputs/teacher_trajectories/dolci_combined_top128",
+  "max_loss_tokens": null,
+  "loss_tokens_per_step": 120000,
+  "teacher_device": "cuda:0",
+  "student_device": "cuda:0",
+  "lr": 3e-05,
+  "optimizer": "adamw8bit",
+  "weight_decay": 0.1,
+  "epochs": 9,
+  "prompts_per_step": 256,
+  "group_size": 1,
+  "rollout_batch": 64,
+  "micro_batch": 3,
+  "max_new_tokens": 256,
+  "max_prompt_len": 1024,
+  "warmup_steps": 10,
+  "max_grad_norm": 1.0,
+  "eval_every": 10,
+  "gsm8k_every": 0,
+  "gsm8k_n": 256,
+  "gsm8k_batch": 16,
+  "gsm8k_max_new_tokens": 512,
+  "gsm8k_frames": "chat",
+  "save_every": 300,
+  "out_dir": "outputs/healed/calib/glean_keep75_s1224",
+  "sweep": 300,
+  "wandb": true,
+  "wandb_project": "glean-calib",
+  "wandb_run_name": "calib_keep75_300",
+  "wandb_run_id": null,
+  "wandb_resume": null,
+  "wandb_mode": "online",
+  "no_wandb_sync": false,
+  "debug": false,
+  "resume_from": "outputs/healed/calib/glean_keep75_s1224/step0200",
+  "start_step": 0,
+  "no_grad_checkpointing": false,
+  "seed": 1224,
+  "no_teacher_overlap": false,
+  "sync_checkpoints": false,
+  "rollout_engine": "hf",
+  "vllm_gpu": null,
+  "vllm_port": 8377,
+  "vllm_refresh_every": 5,
+  "vllm_serve_bin": "vllm-plugin/.venv/bin/python",
+  "vllm_gpu_mem_util": 0.85,
+  "vllm_refresh_mode": "reload",
+  "vllm_live_dir": null,
+  "resolved_kl_direction": "forward"
+}

healed/calib/glean_keep75_s1224/step0200/chat_template.jinja

@@ -0,0 +1,9 @@
+{{ bos_token }}{% for message in messages %}{% if message['role'] == 'system' %}{{ '<|system|>
+' + message['content'] + '
+' }}{% elif message['role'] == 'user' %}{{ '<|user|>
+' + message['content'] + '
+' }}{% elif message['role'] == 'assistant' %}{% if not loop.last %}{{ '<|assistant|>
+'  + message['content'] + eos_token + '
+' }}{% else %}{{ '<|assistant|>
+'  + message['content'] + eos_token }}{% endif %}{% endif %}{% if loop.last and add_generation_prompt %}{{ '<|assistant|>
+' }}{% endif %}{% endfor %}

healed/calib/glean_keep75_s1224/step0200/config.json

@@ -0,0 +1,1022 @@
+{
+  "architectures": [
+    "PrunedOlmoeForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "auto_map": {
+    "AutoConfig": "configuration_pruned_olmoe.PrunedOlmoeConfig",
+    "AutoModelForCausalLM": "modeling_pruned_olmoe.PrunedOlmoeForCausalLM"
+  },
+  "clip_qkv": null,
+  "dtype": "bfloat16",
+  "eos_token_id": 50279,
+  "expert_widths": [
+    [
+      1024,
+      768,
+      768,
+      768,
+      1024,
+      1024,
+      640,
+      1024,
+      640,
+      768,
+      1024,
+      896,
+      640,
+      256,
+      1024,
+      1024,
+      1024,
+      384,
+      896,
+      1024,
+      896,
+      768,
+      896,
+      896,
+      1024,
+      896,
+      384,
+      896,
+      896,
+      384,
+      512,
+      1024,
+      896,
+      1024,
+      768,
+      896,
+      1024,
+      1024,
+      1024,
+      640,
+      1024,
+      640,
+      768,
+      768,
+      1024,
+      896,
+      1024,
+      1024,
+      512,
+      256,
+      1024,
+      896,
+      896,
+      1024,
+      1024,
+      1024,
+      1024,
+      640,
+      256
+    ],
+    [
+      1024,
+      1024,
+      1024,
+      768,
+      1024,
+      640,
+      384,
+      1024,
+      256,
+      896,
+      1024,
+      768,
+      512,
+      768,
+      1024,
+      1024,
+      1024,
+      1024,
+      1024,
+      896,
+      128,
+      1024,
+      896,
+      896,
+      768,
+      640,
+      1024,
+      896,
+      1024,
+      512,
+      1024,
+      128,
+      1024,
+      1024,
+      1024,
+      1024,
+      1024,
+      384,
+      512,
+      1024,
+      896,
+      1024,
+      1024,
+      1024,
+      1024,
+      896,
+      512,
+      1024,
+      768,
+      1024,
+      768,
+      640,
+      1024,
+      1024,
+      896,
+      896,
+      768,
+      768
+    ],
+    [
+      1024,
+      1024,
+      896,
+      384,
+      1024,
+      1024,
+      384,
+      1024,
+      768,
+      1024,
+      1024,
+      1024,
+      512,
+      1024,
+      1024,
+      128,
+      1024,
+      1024,
+      1024,
+      1024,
+      768,
+      1024,
+      256,
+      1024,
+      1024,
+      1024,
+      384,
+      1024,
+      512,
+      1024,
+      1024,
+      1024,
+      1024,
+      1024,
+      256,
+      640,
+      1024,
+      1024,
+      768,
+      1024,
+      896,
+      256,
+      512,
+      1024,
+      768,
+      1024,
+      1024,
+      768,
+      1024,
+      1024,
+      128,
+      1024,
+      1024,
+      1024,
+      1024,
+      1024,
+      896,
+      384
+    ],
+    [
+      1024,
+      1024,
+      1024,
+      128,
+      896,
+      128,
+      1024,
+      1024,
+      896,
+      896,
+      768,
+      1024,
+      768,
+      1024,
+      1024,
+      1024,
+      640,
+      1024,
+      1024,
+      896,
+      1024,
+      256,
+      896,
+      1024,
+      1024,
+      1024,
+      1024,
+      1024,
+      1024,
+      1024,
+      1024,
+      512,
+      768,
+      1024,
+      384,
+      1024,
+      1024,
+      1024,
+      1024,
+      768,
+      1024,
+      1024,
+      1024,
+      1024,
+      1024,
+      1024,
+      256,
+      1024,
+      512,
+      1024,
+      1024,
+      1024,
+      1024,
+      896,
+      768,
+      256
+    ],
+    [
+      1024,
+      1024,
+      128,
+      768,
+      896,
+      1024,
+      1024,
+      1024,
+      1024,
+      1024,
+      896,
+      1024,
+      1024,
+      1024,
+      384,
+      1024,
+      1024,
+      1024,
+      896,
+      1024,
+      896,
+      640,
+      768,
+      640,
+      1024,
+      896,
+      1024,
+      768,
+      1024,
+      1024,
+      896,
+      1024,
+      1024,
+      896,
+      1024,
+      640,
+      768,
+      896,
+      896,
+      1024,
+      1024,
+      768,
+      1024,
+      896,
+      256,
+      1024,
+      512,
+      768,
+      1024,
+      1024,
+      1024,
+      896,
+      1024,
+      768,
+      1024
+    ],
+    [
+      1024,
+      768,
+      768,
+      768,
+      1024,
+      1024,
+      896,
+      1024,
+      896,
+      1024,
+      1024,
+      640,
+      1024,
+      896,
+      1024,
+      896,
+      1024,
+      896,
+      768,
+      1024,
+      896,
+      1024,
+      896,
+      896,
+      1024,
+      1024,
+      768,
+      512,
+      896,
+      1024,
+      1024,
+      896,
+      1024,
+      896,
+      896,
+      384,
+      128,
+      640,
+      1024,
+      896,
+      1024,
+      1024,
+      1024,
+      1024,
+      1024,
+      896,
+      128,
+      1024,
+      512,
+      1024,
+      256,
+      1024,
+      1024,
+      896,
+      896,
+      128,
+      128,
+      896
+    ],
+    [
+      1024,
+      768,
+      1024,
+      896,
+      1024,
+      1024,
+      1024,
+      1024,
+      896,
+      128,
+      256,
+      512,
+      1024,
+      896,
+      768,
+      768,
+      1024,
+      1024,
+      1024,
+      512,
+      1024,
+      768,
+      1024,
+      1024,
+      640,
+      1024,
+      256,
+      1024,
+      1024,
+      896,
+      128,
+      896,
+      1024,
+      1024,
+      896,
+      1024,
+      384,
+      768,
+      384,
+      896,
+      1024,
+      896,
+      384,
+      128,
+      512,
+      896,
+      1024,
+      1024,
+      768,
+      1024,
+      384,
+      256,
+      1024,
+      1024,
+      1024,
+      256,
+      640,
+      1024,
+      1024,
+      1024,
+      1024
+    ],
+    [
+      384,
+      1024,
+      640,
+      1024,
+      1024,
+      896,
+      512,
+      896,
+      640,
+      896,
+      384,
+      1024,
+      1024,
+      1024,
+      1024,
+      896,
+      1024,
+      896,
+      512,
+      896,
+      896,
+      896,
+      896,
+      1024,
+      1024,
+      384,
+      1024,
+      896,
+      768,
+      1024,
+      1024,
+      896,
+      1024,
+      1024,
+      768,
+      384,
+      896,
+      1024,
+      896,
+      512,
+      1024,
+      384,
+      896,
+      384,
+      768,
+      384,
+      1024,
+      1024,
+      768,
+      896,
+      896,
+      1024,
+      896,
+      1024,
+      896,
+      1024,
+      896,
+      1024
+    ],
+    [
+      896,
+      896,
+      768,
+      896,
+      896,
+      384,
+      768,
+      512,
+      896,
+      896,
+      896,
+      896,
+      1024,
+      512,
+      768,
+      1024,
+      1024,
+      640,
+      640,
+      1024,
+      768,
+      384,
+      1024,
+      1024,
+      384,
+      1024,
+      1024,
+      640,
+      1024,
+      896,
+      896,
+      640,
+      896,
+      896,
+      1024,
+      1024,
+      896,
+      640,
+      1024,
+      896,
+      768,
+      512,
+      1024,
+      896,
+      640,
+      896,
+      896,
+      1024,
+      896,
+      896,
+      512,
+      1024,
+      1024,
+      1024,
+      768,
+      1024,
+      256,
+      768,
+      1024
+    ],
+    [
+      1024,
+      1024,
+      1024,
+      768,
+      640,
+      1024,
+      640,
+      1024,
+      896,
+      128,
+      768,
+      896,
+      640,
+      768,
+      1024,
+      128,
+      896,
+      256,
+      256,
+      768,
+      1024,
+      896,
+      1024,
+      896,
+      128,
+      640,
+      640,
+      1024,
+      768,
+      1024,
+      1024,
+      256,
+      768,
+      1024,
+      256,
+      1024,
+      896,
+      1024,
+      1024,
+      384,
+      512,
+      768,
+      1024,
+      1024,
+      896,
+      1024,
+      1024,
+      1024,
+      896,
+      768,
+      1024,
+      1024,
+      1024,
+      384,
+      1024,
+      896,
+      896,
+      1024,
+      1024,
+      128,
+      896,
+      512
+    ],
+    [
+      896,
+      896,
+      1024,
+      1024,
+      128,
+      768,
+      640,
+      896,
+      640,
+      1024,
+      896,
+      1024,
+      896,
+      768,
+      1024,
+      640,
+      1024,
+      1024,
+      1024,
+      128,
+      1024,
+      128,
+      256,
+      512,
+      896,
+      1024,
+      1024,
+      640,
+      640,
+      1024,
+      768,
+      1024,
+      768,
+      896,
+      768,
+      1024,
+      1024,
+      896,
+      1024,
+      512,
+      512,
+      1024,
+      1024,
+      768,
+      384,
+      256,
+      1024,
+      768,
+      896,
+      256,
+      896,
+      1024,
+      1024,
+      896,
+      512,
+      1024,
+      640,
+      512,
+      1024,
+      768,
+      896,
+      768
+    ],
+    [
+      1024,
+      1024,
+      1024,
+      1024,
+      256,
+      896,
+      512,
+      1024,
+      896,
+      1024,
+      1024,
+      896,
+      1024,
+      896,
+      896,
+      1024,
+      1024,
+      768,
+      1024,
+      1024,
+      128,
+      1024,
+      1024,
+      1024,
+      1024,
+      1024,
+      896,
+      1024,
+      1024,
+      768,
+      1024,
+      512,
+      896,
+      640,
+      896,
+      384,
+      896,
+      896,
+      128,
+      1024,
+      1024,
+      896,
+      1024,
+      896,
+      768,
+      1024,
+      1024,
+      384,
+      896,
+      896,
+      1024,
+      896,
+      1024,
+      1024,
+      896,
+      896
+    ],
+    [
+      1024,
+      896,
+      768,
+      1024,
+      512,
+      896,
+      1024,
+      256,
+      1024,
+      1024,
+      1024,
+      896,
+      1024,
+      896,
+      1024,
+      128,
+      1024,
+      768,
+      512,
+      1024,
+      1024,
+      256,
+      640,
+      768,
+      896,
+      384,
+      128,
+      384,
+      768,
+      512,
+      896,
+      896,
+      1024,
+      1024,
+      896,
+      896,
+      1024,
+      256,
+      896,
+      256,
+      1024,
+      1024,
+      896,
+      896,
+      1024,
+      896,
+      1024,
+      1024,
+      1024,
+      1024,
+      768,
+      768,
+      896,
+      256,
+      256,
+      768,
+      1024,
+      896,
+      1024,
+      1024,
+      1024
+    ],
+    [
+      1024,
+      1024,
+      1024,
+      1024,
+      1024,
+      768,
+      640,
+      1024,
+      384,
+      1024,
+      768,
+      1024,
+      1024,
+      896,
+      896,
+      1024,
+      896,
+      896,
+      896,
+      1024,
+      256,
+      896,
+      1024,
+      768,
+      1024,
+      1024,
+      1024,
+      896,
+      1024,
+      896,
+      1024,
+      1024,
+      896,
+      1024,
+      896,
+      896,
+      1024,
+      1024,
+      896,
+      384,
+      256,
+      256,
+      1024,
+      1024,
+      1024,
+      256,
+      384,
+      512,
+      512,
+      896,
+      640,
+      1024,
+      1024,
+      384,
+      1024,
+      768,
+      896,
+      1024
+    ],
+    [
+      1024,
+      768,
+      640,
+      896,
+      384,
+      1024,
+      1024,
+      768,
+      1024,
+      1024,
+      896,
+      128,
+      1024,
+      896,
+      1024,
+      896,
+      512,
+      1024,
+      1024,
+      1024,
+      896,
+      768,
+      1024,
+      896,
+      768,
+      768,
+      256,
+      1024,
+      1024,
+      640,
+      1024,
+      768,
+      640,
+      384,
+      640,
+      896,
+      1024,
+      896,
+      896,
+      1024,
+      1024,
+      1024,
+      1024,
+      1024,
+      1024,
+      256,
+      256,
+      384,
+      1024,
+      1024,
+      512,
+      896,
+      1024,
+      896,
+      896,
+      896,
+      896,
+      896,
+      896
+    ],
+    [
+      1024,
+      1024,
+      896,
+      1024,
+      1024,
+      768,
+      1024,
+      768,
+      896,
+      384,
+      896,
+      256,
+      896,
+      896,
+      896,
+      896,
+      1024,
+      640,
+      256,
+      896,
+      768,
+      384,
+      768,
+      640,
+      896,
+      768,
+      384,
+      896,
+      1024,
+      896,
+      1024,
+      128,
+      768,
+      512,
+      1024,
+      256,
+      896,
+      1024,
+      768,
+      896,
+      896,
+      1024,
+      1024,
+      768,
+      1024,
+      896,
+      512,
+      768,
+      384,
+      1024,
+      1024,
+      512,
+      896,
+      640,
+      896,
+      1024,
+      768,
+      384,
+      768,
+      1024,
+      896,
+      896
+    ]
+  ],
+  "glean_metadata": {
+    "base_model": "allenai/OLMoE-1B-7B-0125-Instruct",
+    "block_size": 128,
+    "criterion": "reap",
+    "dead_experts": 82,
+    "keep_fraction": 0.75,
+    "min_width": 128,
+    "params": 5308381184,
+    "scores": "outputs/scores_general_mix/scores.pt"
+  },
+  "hidden_act": "silu",
+  "hidden_size": 2048,
+  "initializer_range": 0.02,
+  "intermediate_size": 1024,
+  "max_position_embeddings": 4096,
+  "model_type": "pruned_olmoe",
+  "norm_topk_prob": false,
+  "num_attention_heads": 16,
+  "num_experts": 64,
+  "num_experts_per_tok": 8,
+  "num_hidden_layers": 16,
+  "num_key_value_heads": 16,
+  "output_router_logits": false,
+  "pad_token_id": 1,
+  "rms_norm_eps": 1e-05,
+  "rope_scaling": null,
+  "rope_theta": 10000.0,
+  "router_aux_loss_coef": 0.01,
+  "tie_word_embeddings": false,
+  "transformers_version": "4.57.6",
+  "use_cache": false,
+  "vocab_size": 50304
+}

healed/calib/glean_keep75_s1224/step0200/configuration_pruned_olmoe.py

@@ -0,0 +1,27 @@
+"""Configuration for GLEAN-pruned OLMoE: variable-width, variable-count experts.
+"""
+
+from transformers.models.olmoe.configuration_olmoe import OlmoeConfig
+
+
+class PrunedOlmoeConfig(OlmoeConfig):
+    """OlmoeConfig plus a per-(layer, expert) width table.
+
+    ``expert_widths[l]`` lists the SwiGLU intermediate width of each surviving
+    expert in decoder layer ``l``, in expert order. Lists are ragged: layers
+    may keep different numbers of experts (deleted experts simply don't
+    appear — the router in layer ``l`` has ``len(expert_widths[l])`` rows),
+    and each width may differ (multiples of the GEMM block size, 128, for
+    variable-MegaBlocks execution). ``None`` means an unpruned model
+    (uniform ``num_experts`` × ``intermediate_size``).
+
+    The inherited ``num_experts`` / ``intermediate_size`` keep their ORIGINAL
+    (pre-pruning) values for provenance; the width table is authoritative for
+    the built architecture.
+    """
+
+    model_type = "pruned_olmoe"
+
+    def __init__(self, expert_widths: list[list[int]] | None = None, **kwargs):
+        super().__init__(**kwargs)
+        self.expert_widths = expert_widths

healed/calib/glean_keep75_s1224/step0200/generation_config.json

@@ -0,0 +1,6 @@
+{
+  "_from_model_config": true,
+  "eos_token_id": 50279,
+  "pad_token_id": 1,
+  "transformers_version": "4.57.6"
+}

healed/calib/glean_keep75_s1224/step0200/modeling_pruned_olmoe.py

@@ -0,0 +1,66 @@
+"""GLEAN-pruned OLMoE: HF-loadable model with ragged (variable-width) experts.
+
+Pattern follows hbfreed/variable-flex-olmo's PrunedFlexOlmoForCausalLM
+(docs/recon/prior-work-hbfreed.md), generalized from one scalar width to a
+per-(layer, expert) width table: ``super().__init__`` builds the uniform
+architecture from the config, then every MoE block is rebuilt to its pruned
+shape — surviving experts only, each at its own width, router sliced to
+match — so the state dict aligns exactly with what
+``glean.prune.prune_channels_global`` leaves behind.
+
+Caveat: ``output_router_logits=True`` (the load-balancing aux loss) assumes a
+uniform ``config.num_experts`` and is unsupported on ragged models.
+"""
+
+import torch.nn as nn
+from transformers.activations import ACT2FN
+from transformers.models.olmoe.modeling_olmoe import OlmoeForCausalLM
+
+from .configuration_pruned_olmoe import PrunedOlmoeConfig
+
+
+class RaggedOlmoeMLP(nn.Module):
+    """OlmoeMLP with an explicit intermediate width (SwiGLU, no biases)."""
+
+    def __init__(self, hidden_size: int, intermediate_size: int, hidden_act: str):
+        super().__init__()
+        self.hidden_size = hidden_size
+        self.intermediate_size = intermediate_size
+        self.gate_proj = nn.Linear(hidden_size, intermediate_size, bias=False)
+        self.up_proj = nn.Linear(hidden_size, intermediate_size, bias=False)
+        self.down_proj = nn.Linear(intermediate_size, hidden_size, bias=False)
+        self.act_fn = ACT2FN[hidden_act]
+
+    def forward(self, x):
+        return self.down_proj(self.act_fn(self.gate_proj(x)) * self.up_proj(x))
+
+
+class PrunedOlmoeForCausalLM(OlmoeForCausalLM):
+    """OLMoE with per-layer surviving-expert lists at per-expert widths."""
+
+    config_class = PrunedOlmoeConfig
+
+    def __init__(self, config: PrunedOlmoeConfig):
+        super().__init__(config)
+        widths_table = getattr(config, "expert_widths", None)
+        if widths_table is None:
+            return  # unpruned: plain OLMoE
+        if len(widths_table) != len(self.model.layers):
+            raise ValueError(
+                f"expert_widths has {len(widths_table)} rows but the model has "
+                f"{len(self.model.layers)} decoder layers"
+            )
+        for layer, widths in zip(self.model.layers, widths_table):
+            if any(w <= 0 for w in widths):
+                raise ValueError("expert_widths must list surviving experts only (>0)")
+            block = layer.mlp
+            if len(widths) < block.top_k:
+                raise ValueError(
+                    f"a layer keeps {len(widths)} experts < top_k={block.top_k}"
+                )
+            block.num_experts = len(widths)
+            block.gate = nn.Linear(config.hidden_size, len(widths), bias=False)
+            block.experts = nn.ModuleList(
+                RaggedOlmoeMLP(config.hidden_size, w, config.hidden_act)
+                for w in widths
+            )

healed/calib/glean_keep75_s1224/step0200/special_tokens_map.json

@@ -0,0 +1,23 @@
+{
+  "bos_token": {
+    "content": "|||IP_ADDRESS|||",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "|||IP_ADDRESS|||",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<pad>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

healed/calib/glean_keep75_s1224/step0200/tokenizer_config.json

@@ -0,0 +1,247 @@
+{
+  "add_bos_token": false,
+  "add_eos_token": false,
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<|padding|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "50254": {
+      "content": "                        ",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50255": {
+      "content": "                       ",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50256": {
+      "content": "                      ",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50257": {
+      "content": "                     ",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50258": {
+      "content": "                    ",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50259": {
+      "content": "                   ",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50260": {
+      "content": "                  ",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50261": {
+      "content": "                 ",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50262": {
+      "content": "                ",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50263": {
+      "content": "               ",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50264": {
+      "content": "              ",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50265": {
+      "content": "             ",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50266": {
+      "content": "            ",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50267": {
+      "content": "           ",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50268": {
+      "content": "          ",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50269": {
+      "content": "         ",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50270": {
+      "content": "        ",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50271": {
+      "content": "       ",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50272": {
+      "content": "      ",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50273": {
+      "content": "     ",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50274": {
+      "content": "    ",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50275": {
+      "content": "   ",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50276": {
+      "content": "  ",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50277": {
+      "content": "|||EMAIL_ADDRESS|||",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50278": {
+      "content": "|||PHONE_NUMBER|||",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50279": {
+      "content": "|||IP_ADDRESS|||",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "50280": {
+      "content": "<pad>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "|||IP_ADDRESS|||",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "|||IP_ADDRESS|||",
+  "extra_special_tokens": {},
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<pad>",
+  "tokenizer_class": "GPTNeoXTokenizer",
+  "unk_token": null
+}

healed/calib/glean_keep75_s1224/train_log.jsonl

@@ -0,0 +1,268 @@
+{"step": 1, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.13071662265943984, "tokens": 120000, "cumulative_loss_tokens": 120000, "grad_norm": 1.8671875, "lr": 6e-06, "finish_rate": 0.869, "comp_len": 463.3, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 84.6, "frames": {"chat": 259}, "mem_gb": 21.93}
+{"step": 2, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.1904543469694443, "tokens": 120000, "cumulative_loss_tokens": 240000, "grad_norm": 1.5859375, "lr": 9e-06, "finish_rate": 0.997, "comp_len": 344.8, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 69.7, "frames": {"chat": 348}, "mem_gb": 21.69}
+{"step": 3, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.17038091265981395, "tokens": 120000, "cumulative_loss_tokens": 360000, "grad_norm": 1.421875, "lr": 1.2e-05, "finish_rate": 0.997, "comp_len": 324.3, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 68.9, "frames": {"chat": 370}, "mem_gb": 21.84}
+{"step": 4, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.06029140768460929, "tokens": 120000, "cumulative_loss_tokens": 480000, "grad_norm": 0.87109375, "lr": 1.5e-05, "finish_rate": 0.852, "comp_len": 491.8, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 50.8, "frames": {"chat": 244}, "mem_gb": 22.07}
+{"step": 5, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.1347129997333667, "tokens": 120000, "cumulative_loss_tokens": 600000, "grad_norm": 0.94140625, "lr": 1.8e-05, "finish_rate": 0.953, "comp_len": 375.0, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 63.0, "frames": {"chat": 320}, "mem_gb": 21.99}
+{"step": 6, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.1584433459235588, "tokens": 120000, "cumulative_loss_tokens": 720000, "grad_norm": 0.91015625, "lr": 2.1e-05, "finish_rate": 0.997, "comp_len": 317.5, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 69.4, "frames": {"chat": 378}, "mem_gb": 21.59}
+{"step": 7, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.194529585396188, "tokens": 120000, "cumulative_loss_tokens": 840000, "grad_norm": 0.921875, "lr": 2.4e-05, "finish_rate": 0.997, "comp_len": 312.5, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 70.7, "frames": {"chat": 384}, "mem_gb": 22.08}
+{"step": 8, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.1841610342130841, "tokens": 120000, "cumulative_loss_tokens": 960000, "grad_norm": 0.9140625, "lr": 2.7000000000000002e-05, "finish_rate": 0.997, "comp_len": 370.4, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 65.7, "frames": {"chat": 324}, "mem_gb": 21.93}
+{"step": 9, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.06550765600646846, "tokens": 120000, "cumulative_loss_tokens": 1080000, "grad_norm": 0.90234375, "lr": 3e-05, "finish_rate": 0.891, "comp_len": 502.1, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 53.4, "frames": {"chat": 239}, "mem_gb": 22.09}
+{"step": 10, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.15740935421491353, "tokens": 120000, "cumulative_loss_tokens": 1200000, "grad_norm": 1.078125, "lr": 3e-05, "finish_rate": 0.961, "comp_len": 392.2, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 60.4, "frames": {"chat": 306}, "mem_gb": 21.83}
+{"step": 11, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.16742474506984775, "tokens": 120000, "cumulative_loss_tokens": 1320000, "grad_norm": 0.79296875, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 364.7, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 66.8, "frames": {"chat": 329}, "mem_gb": 21.87}
+{"step": 12, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.08396718050995841, "tokens": 120000, "cumulative_loss_tokens": 1440000, "grad_norm": 0.671875, "lr": 3e-05, "finish_rate": 0.882, "comp_len": 472.4, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 54.4, "frames": {"chat": 254}, "mem_gb": 22.1}
+{"step": 13, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.1496814745740189, "tokens": 120000, "cumulative_loss_tokens": 1560000, "grad_norm": 0.65234375, "lr": 3e-05, "finish_rate": 0.974, "comp_len": 439.6, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 58.8, "frames": {"chat": 273}, "mem_gb": 22.08}
+{"step": 14, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.04953998726032054, "tokens": 120000, "cumulative_loss_tokens": 1680000, "grad_norm": 0.423828125, "lr": 3e-05, "finish_rate": 0.873, "comp_len": 508.5, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 50.6, "frames": {"chat": 236}, "mem_gb": 22.0}
+{"step": 15, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.1870271451675333, "tokens": 120000, "cumulative_loss_tokens": 1800000, "grad_norm": 0.8046875, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 315.0, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 71.0, "frames": {"chat": 381}, "mem_gb": 21.73}
+{"step": 16, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.065772617749994, "tokens": 120000, "cumulative_loss_tokens": 1920000, "grad_norm": 0.5078125, "lr": 3e-05, "finish_rate": 0.925, "comp_len": 451.1, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 54.1, "frames": {"chat": 266}, "mem_gb": 21.93}
+{"step": 17, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.10379441037213741, "tokens": 120000, "cumulative_loss_tokens": 2040000, "grad_norm": 0.5234375, "lr": 3e-05, "finish_rate": 0.961, "comp_len": 389.6, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 62.6, "frames": {"chat": 308}, "mem_gb": 21.92}
+{"step": 18, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.09278505114205182, "tokens": 120000, "cumulative_loss_tokens": 2160000, "grad_norm": 0.484375, "lr": 3e-05, "finish_rate": 0.853, "comp_len": 439.6, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 56.1, "frames": {"chat": 273}, "mem_gb": 22.1}
+{"step": 19, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.15566241025024405, "tokens": 120000, "cumulative_loss_tokens": 2280000, "grad_norm": 0.6328125, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 326.1, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 68.0, "frames": {"chat": 368}, "mem_gb": 21.72}
+{"step": 20, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.2065904488075835, "tokens": 120000, "cumulative_loss_tokens": 2400000, "grad_norm": 0.66015625, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 354.0, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 67.6, "frames": {"chat": 339}, "mem_gb": 21.97}
+{"step": 21, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.09833553963121182, "tokens": 120000, "cumulative_loss_tokens": 2520000, "grad_norm": 0.48828125, "lr": 3e-05, "finish_rate": 0.963, "comp_len": 402.7, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 60.7, "frames": {"chat": 298}, "mem_gb": 21.89}
+{"step": 22, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.16023428318003813, "tokens": 120000, "cumulative_loss_tokens": 2640000, "grad_norm": 0.640625, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 342.9, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 66.8, "frames": {"chat": 350}, "mem_gb": 21.62}
+{"step": 23, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.08126048041522348, "tokens": 120000, "cumulative_loss_tokens": 2760000, "grad_norm": 0.462890625, "lr": 3e-05, "finish_rate": 0.858, "comp_len": 487.8, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 54.7, "frames": {"chat": 246}, "mem_gb": 22.06}
+{"step": 24, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.1679548702717526, "tokens": 120000, "cumulative_loss_tokens": 2880000, "grad_norm": 0.66796875, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 320.0, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 69.2, "frames": {"chat": 375}, "mem_gb": 21.79}
+{"step": 25, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.16708556045874332, "tokens": 120000, "cumulative_loss_tokens": 3000000, "grad_norm": 0.64453125, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 340.9, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 69.7, "frames": {"chat": 352}, "mem_gb": 21.85}
+{"step": 26, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.16134658801977833, "tokens": 120000, "cumulative_loss_tokens": 3120000, "grad_norm": 0.5546875, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 303.0, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 74.4, "frames": {"chat": 396}, "mem_gb": 21.8}
+{"step": 27, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.1342053831448313, "tokens": 120000, "cumulative_loss_tokens": 3240000, "grad_norm": 0.59765625, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 331.5, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 70.1, "frames": {"chat": 362}, "mem_gb": 21.66}
+{"step": 28, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.17713635508442918, "tokens": 120000, "cumulative_loss_tokens": 3360000, "grad_norm": 0.59375, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 325.2, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 71.9, "frames": {"chat": 369}, "mem_gb": 21.87}
+{"step": 29, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.1749166687601867, "tokens": 120000, "cumulative_loss_tokens": 3480000, "grad_norm": 0.609375, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 300.8, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 73.5, "frames": {"chat": 399}, "mem_gb": 21.72}
+{"step": 30, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.10207746837508554, "tokens": 120000, "cumulative_loss_tokens": 3600000, "grad_norm": 1.0859375, "lr": 3e-05, "finish_rate": 0.911, "comp_len": 485.8, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 55.3, "frames": {"chat": 247}, "mem_gb": 21.99}
+{"step": 31, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.03470097794899096, "tokens": 120000, "cumulative_loss_tokens": 3720000, "grad_norm": 0.48828125, "lr": 3e-05, "finish_rate": 0.891, "comp_len": 521.7, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 51.4, "frames": {"chat": 230}, "mem_gb": 21.93}
+{"step": 32, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.07190631164882022, "tokens": 120000, "cumulative_loss_tokens": 3840000, "grad_norm": 0.462890625, "lr": 3e-05, "finish_rate": 0.905, "comp_len": 456.3, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 58.8, "frames": {"chat": 263}, "mem_gb": 22.1}
+{"step": 33, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.08197221969100647, "tokens": 120000, "cumulative_loss_tokens": 3960000, "grad_norm": 0.58203125, "lr": 3e-05, "finish_rate": 0.848, "comp_len": 454.5, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 55.6, "frames": {"chat": 264}, "mem_gb": 22.09}
+{"step": 34, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.14606728826724769, "tokens": 120000, "cumulative_loss_tokens": 4080000, "grad_norm": 0.5859375, "lr": 3e-05, "finish_rate": 0.991, "comp_len": 365.9, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 69.9, "frames": {"chat": 328}, "mem_gb": 21.95}
+{"step": 35, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.07470543375237224, "tokens": 120000, "cumulative_loss_tokens": 4200000, "grad_norm": 0.57421875, "lr": 3e-05, "finish_rate": 0.925, "comp_len": 427.0, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 60.2, "frames": {"chat": 281}, "mem_gb": 22.02}
+{"step": 36, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.04788410694993411, "tokens": 120000, "cumulative_loss_tokens": 4320000, "grad_norm": 0.390625, "lr": 3e-05, "finish_rate": 0.861, "comp_len": 506.3, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 53.2, "frames": {"chat": 237}, "mem_gb": 22.01}
+{"step": 37, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.060693397747950316, "tokens": 120000, "cumulative_loss_tokens": 4440000, "grad_norm": 0.4453125, "lr": 3e-05, "finish_rate": 0.91, "comp_len": 431.7, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 56.8, "frames": {"chat": 278}, "mem_gb": 22.1}
+{"step": 38, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.13441723232055083, "tokens": 120000, "cumulative_loss_tokens": 4560000, "grad_norm": 0.5546875, "lr": 3e-05, "finish_rate": 0.955, "comp_len": 387.1, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 62.9, "frames": {"chat": 310}, "mem_gb": 21.97}
+{"step": 39, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.07622199243601256, "tokens": 120000, "cumulative_loss_tokens": 4680000, "grad_norm": 0.4296875, "lr": 3e-05, "finish_rate": 0.959, "comp_len": 406.8, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 59.5, "frames": {"chat": 295}, "mem_gb": 21.91}
+{"step": 40, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.09086993359592743, "tokens": 120000, "cumulative_loss_tokens": 4800000, "grad_norm": 0.470703125, "lr": 3e-05, "finish_rate": 0.957, "comp_len": 364.7, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 62.7, "frames": {"chat": 329}, "mem_gb": 21.92}
+{"step": 41, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.08632465346843625, "tokens": 120000, "cumulative_loss_tokens": 4920000, "grad_norm": 0.46484375, "lr": 3e-05, "finish_rate": 0.905, "comp_len": 456.3, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 59.3, "frames": {"chat": 263}, "mem_gb": 22.04}
+{"step": 42, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.1030513121418655, "tokens": 120000, "cumulative_loss_tokens": 5040000, "grad_norm": 0.458984375, "lr": 3e-05, "finish_rate": 0.981, "comp_len": 381.0, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 61.1, "frames": {"chat": 315}, "mem_gb": 21.92}
+{"step": 43, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.13386018192392463, "tokens": 120000, "cumulative_loss_tokens": 5160000, "grad_norm": 0.53515625, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 339.0, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 70.1, "frames": {"chat": 354}, "mem_gb": 21.87}
+{"step": 44, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.07069139333600179, "tokens": 120000, "cumulative_loss_tokens": 5280000, "grad_norm": 0.5625, "lr": 3e-05, "finish_rate": 0.927, "comp_len": 463.3, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 55.8, "frames": {"chat": 259}, "mem_gb": 22.04}
+{"step": 45, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.09212584427177596, "tokens": 120000, "cumulative_loss_tokens": 5400000, "grad_norm": 0.4765625, "lr": 3e-05, "finish_rate": 0.92, "comp_len": 436.4, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 60.0, "frames": {"chat": 275}, "mem_gb": 21.9}
+{"step": 46, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.12813016946287825, "tokens": 120000, "cumulative_loss_tokens": 5520000, "grad_norm": 0.5625, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 359.3, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 69.8, "frames": {"chat": 334}, "mem_gb": 21.77}
+{"step": 47, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.14056541419519733, "tokens": 120000, "cumulative_loss_tokens": 5640000, "grad_norm": 0.53125, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 320.9, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 72.6, "frames": {"chat": 374}, "mem_gb": 21.62}
+{"step": 48, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.12320457009705715, "tokens": 120000, "cumulative_loss_tokens": 5760000, "grad_norm": 0.490234375, "lr": 3e-05, "finish_rate": 0.974, "comp_len": 385.9, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 65.6, "frames": {"chat": 311}, "mem_gb": 21.79}
+{"step": 49, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.09483733469761597, "tokens": 120000, "cumulative_loss_tokens": 5880000, "grad_norm": 0.431640625, "lr": 3e-05, "finish_rate": 0.876, "comp_len": 463.3, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 56.4, "frames": {"chat": 259}, "mem_gb": 22.1}
+{"step": 50, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.1235223285190761, "tokens": 120000, "cumulative_loss_tokens": 6000000, "grad_norm": 0.494140625, "lr": 3e-05, "finish_rate": 0.944, "comp_len": 393.4, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 64.8, "frames": {"chat": 305}, "mem_gb": 22.08}
+{"step": 51, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.10593181157653841, "tokens": 120000, "cumulative_loss_tokens": 6120000, "grad_norm": 0.478515625, "lr": 3e-05, "finish_rate": 0.946, "comp_len": 402.7, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 64.2, "frames": {"chat": 298}, "mem_gb": 22.06}
+{"step": 52, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.16442105506554555, "tokens": 120000, "cumulative_loss_tokens": 6240000, "grad_norm": 0.58203125, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 329.7, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 71.2, "frames": {"chat": 364}, "mem_gb": 21.84}
+{"step": 53, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.07198908675844626, "tokens": 120000, "cumulative_loss_tokens": 6360000, "grad_norm": 0.3984375, "lr": 3e-05, "finish_rate": 0.96, "comp_len": 397.4, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 63.8, "frames": {"chat": 302}, "mem_gb": 21.94}
+{"step": 54, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.15115031077408542, "tokens": 120000, "cumulative_loss_tokens": 6480000, "grad_norm": 0.57421875, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 379.7, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 67.3, "frames": {"chat": 316}, "mem_gb": 21.7}
+{"step": 55, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.15825922077251597, "tokens": 120000, "cumulative_loss_tokens": 6600000, "grad_norm": 0.54296875, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 384.6, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 69.3, "frames": {"chat": 312}, "mem_gb": 21.8}
+{"step": 56, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.1291787802760179, "tokens": 120000, "cumulative_loss_tokens": 6720000, "grad_norm": 0.52734375, "lr": 3e-05, "finish_rate": 0.979, "comp_len": 363.6, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 64.5, "frames": {"chat": 330}, "mem_gb": 21.95}
+{"step": 57, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.1187889644394784, "tokens": 120000, "cumulative_loss_tokens": 6840000, "grad_norm": 0.4921875, "lr": 3e-05, "finish_rate": 0.966, "comp_len": 373.8, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 65.2, "frames": {"chat": 321}, "mem_gb": 21.95}
+{"step": 58, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.10625948149410543, "tokens": 120000, "cumulative_loss_tokens": 6960000, "grad_norm": 0.474609375, "lr": 3e-05, "finish_rate": 0.968, "comp_len": 348.8, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 62.1, "frames": {"chat": 344}, "mem_gb": 21.99}
+{"step": 59, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.13131812240947038, "tokens": 120000, "cumulative_loss_tokens": 7080000, "grad_norm": 0.498046875, "lr": 3e-05, "finish_rate": 0.992, "comp_len": 339.9, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 66.9, "frames": {"chat": 353}, "mem_gb": 21.66}
+{"step": 60, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.06957110121420895, "tokens": 120000, "cumulative_loss_tokens": 7200000, "grad_norm": 0.3984375, "lr": 3e-05, "finish_rate": 0.89, "comp_len": 441.2, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 55.0, "frames": {"chat": 272}, "mem_gb": 22.09}
+{"step": 61, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.14536656329245307, "tokens": 120000, "cumulative_loss_tokens": 7320000, "grad_norm": 0.51171875, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 378.5, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 60.1, "frames": {"chat": 317}, "mem_gb": 21.84}
+{"step": 62, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.057032847379900824, "tokens": 120000, "cumulative_loss_tokens": 7440000, "grad_norm": 0.380859375, "lr": 3e-05, "finish_rate": 0.949, "comp_len": 411.0, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 61.0, "frames": {"chat": 292}, "mem_gb": 21.84}
+{"step": 63, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.026177874839523187, "tokens": 120000, "cumulative_loss_tokens": 7560000, "grad_norm": 0.287109375, "lr": 3e-05, "finish_rate": 0.886, "comp_len": 524.0, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 52.9, "frames": {"chat": 229}, "mem_gb": 21.95}
+{"step": 64, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.14116634934321046, "tokens": 120000, "cumulative_loss_tokens": 7680000, "grad_norm": 0.52734375, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 334.3, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 74.2, "frames": {"chat": 359}, "mem_gb": 21.96}
+{"step": 65, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.12220887529390553, "tokens": 120000, "cumulative_loss_tokens": 7800000, "grad_norm": 0.498046875, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 335.2, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 72.1, "frames": {"chat": 358}, "mem_gb": 21.68}
+{"step": 66, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.14990485909172954, "tokens": 120000, "cumulative_loss_tokens": 7920000, "grad_norm": 0.52734375, "lr": 3e-05, "finish_rate": 0.975, "comp_len": 381.0, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 70.0, "frames": {"chat": 315}, "mem_gb": 22.06}
+{"step": 67, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.0956310116295858, "tokens": 120000, "cumulative_loss_tokens": 8040000, "grad_norm": 0.46484375, "lr": 3e-05, "finish_rate": 0.905, "comp_len": 454.5, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 61.2, "frames": {"chat": 264}, "mem_gb": 22.14}
+{"step": 68, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.13878010407060695, "tokens": 120000, "cumulative_loss_tokens": 8160000, "grad_norm": 0.51953125, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 356.1, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 70.5, "frames": {"chat": 337}, "mem_gb": 21.81}
+{"step": 69, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.07037615146461564, "tokens": 120000, "cumulative_loss_tokens": 8280000, "grad_norm": 0.45703125, "lr": 3e-05, "finish_rate": 0.873, "comp_len": 478.1, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 58.0, "frames": {"chat": 251}, "mem_gb": 22.1}
+{"step": 70, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.11031303176231838, "tokens": 120000, "cumulative_loss_tokens": 8400000, "grad_norm": 0.458984375, "lr": 3e-05, "finish_rate": 0.973, "comp_len": 402.7, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 64.0, "frames": {"chat": 298}, "mem_gb": 21.83}
+{"step": 71, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.08863228327104201, "tokens": 120000, "cumulative_loss_tokens": 8520000, "grad_norm": 0.39453125, "lr": 3e-05, "finish_rate": 0.917, "comp_len": 454.5, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 56.7, "frames": {"chat": 264}, "mem_gb": 22.04}
+{"step": 72, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.08974066026785876, "tokens": 120000, "cumulative_loss_tokens": 8640000, "grad_norm": 0.421875, "lr": 3e-05, "finish_rate": 0.948, "comp_len": 393.4, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 61.8, "frames": {"chat": 305}, "mem_gb": 22.06}
+{"step": 73, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.07843934982779902, "tokens": 120000, "cumulative_loss_tokens": 8760000, "grad_norm": 0.400390625, "lr": 3e-05, "finish_rate": 0.953, "comp_len": 372.7, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 64.0, "frames": {"chat": 322}, "mem_gb": 21.83}
+{"step": 74, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.1308825180749409, "tokens": 120000, "cumulative_loss_tokens": 8880000, "grad_norm": 0.546875, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 332.4, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 71.5, "frames": {"chat": 361}, "mem_gb": 21.82}
+{"step": 75, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.09685897526154294, "tokens": 120000, "cumulative_loss_tokens": 9000000, "grad_norm": 0.427734375, "lr": 3e-05, "finish_rate": 0.972, "comp_len": 373.8, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 68.4, "frames": {"chat": 321}, "mem_gb": 22.0}
+{"step": 76, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.10711560201556422, "tokens": 120000, "cumulative_loss_tokens": 9120000, "grad_norm": 0.4765625, "lr": 3e-05, "finish_rate": 0.967, "comp_len": 390.9, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 65.5, "frames": {"chat": 307}, "mem_gb": 21.89}
+{"step": 77, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.13811318850273577, "tokens": 120000, "cumulative_loss_tokens": 9240000, "grad_norm": 0.54296875, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 367.0, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 70.8, "frames": {"chat": 327}, "mem_gb": 21.99}
+{"step": 78, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.15955485395295546, "tokens": 120000, "cumulative_loss_tokens": 9360000, "grad_norm": 0.58203125, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 337.1, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 70.8, "frames": {"chat": 356}, "mem_gb": 21.61}
+{"step": 79, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.07491358114203128, "tokens": 120000, "cumulative_loss_tokens": 9480000, "grad_norm": 0.43359375, "lr": 3e-05, "finish_rate": 0.915, "comp_len": 424.0, "t_data_s": 0.1, "t_rollout_s": 0.0, "t_step_s": 59.9, "frames": {"chat": 283}, "mem_gb": 21.95}
+{"step": 80, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.13741200089437577, "tokens": 120000, "cumulative_loss_tokens": 9600000, "grad_norm": 0.5, "lr": 3e-05, "finish_rate": 0.998, "comp_len": 299.3, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 76.6, "frames": {"chat": 401}, "mem_gb": 21.7}
+{"step": 81, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.08265723406509497, "tokens": 120000, "cumulative_loss_tokens": 9720000, "grad_norm": 0.400390625, "lr": 3e-05, "finish_rate": 0.965, "comp_len": 416.7, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 61.9, "frames": {"chat": 288}, "mem_gb": 21.9}
+{"step": 82, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.11680444544868854, "tokens": 120000, "cumulative_loss_tokens": 9840000, "grad_norm": 0.48828125, "lr": 3e-05, "finish_rate": 0.981, "comp_len": 371.5, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 67.8, "frames": {"chat": 323}, "mem_gb": 21.94}
+{"step": 83, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.1520622681313505, "tokens": 120000, "cumulative_loss_tokens": 9960000, "grad_norm": 0.55859375, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 357.1, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 67.9, "frames": {"chat": 336}, "mem_gb": 21.83}
+{"step": 84, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.1286243218161476, "tokens": 120000, "cumulative_loss_tokens": 10080000, "grad_norm": 0.515625, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 364.7, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 65.5, "frames": {"chat": 329}, "mem_gb": 21.74}
+{"step": 85, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.12255671736062504, "tokens": 120000, "cumulative_loss_tokens": 10200000, "grad_norm": 0.5, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 349.9, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 73.8, "frames": {"chat": 343}, "mem_gb": 21.72}
+{"step": 86, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.13805025024986825, "tokens": 120000, "cumulative_loss_tokens": 10320000, "grad_norm": 0.51953125, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 357.1, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 70.8, "frames": {"chat": 336}, "mem_gb": 21.84}
+{"step": 87, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.12031711872095863, "tokens": 120000, "cumulative_loss_tokens": 10440000, "grad_norm": 0.5078125, "lr": 3e-05, "finish_rate": 0.984, "comp_len": 382.2, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 65.1, "frames": {"chat": 314}, "mem_gb": 22.08}
+{"step": 88, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.07509244097399059, "tokens": 120000, "cumulative_loss_tokens": 10560000, "grad_norm": 0.37890625, "lr": 3e-05, "finish_rate": 0.968, "comp_len": 427.0, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 58.0, "frames": {"chat": 281}, "mem_gb": 21.91}
+{"step": 89, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.09532498086250077, "tokens": 120000, "cumulative_loss_tokens": 10680000, "grad_norm": 0.4453125, "lr": 3e-05, "finish_rate": 0.977, "comp_len": 401.3, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 62.5, "frames": {"chat": 299}, "mem_gb": 21.79}
+{"step": 90, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.12706806688119346, "tokens": 120000, "cumulative_loss_tokens": 10800000, "grad_norm": 0.466796875, "lr": 3e-05, "finish_rate": 0.966, "comp_len": 365.9, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 66.1, "frames": {"chat": 328}, "mem_gb": 21.98}
+{"step": 91, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.08623081075329489, "tokens": 120000, "cumulative_loss_tokens": 10920000, "grad_norm": 0.419921875, "lr": 3e-05, "finish_rate": 0.919, "comp_len": 442.8, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 54.0, "frames": {"chat": 271}, "mem_gb": 22.08}
+{"step": 92, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.09138209423874505, "tokens": 120000, "cumulative_loss_tokens": 11040000, "grad_norm": 0.431640625, "lr": 3e-05, "finish_rate": 0.874, "comp_len": 446.1, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 56.2, "frames": {"chat": 269}, "mem_gb": 22.18}
+{"step": 93, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.1642836628535685, "tokens": 120000, "cumulative_loss_tokens": 11160000, "grad_norm": 0.578125, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 324.3, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 69.3, "frames": {"chat": 370}, "mem_gb": 21.72}
+{"step": 94, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.17387420818852262, "tokens": 120000, "cumulative_loss_tokens": 11280000, "grad_norm": 0.546875, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 340.9, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 64.2, "frames": {"chat": 352}, "mem_gb": 21.78}
+{"step": 95, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.13778432187635142, "tokens": 120000, "cumulative_loss_tokens": 11400000, "grad_norm": 0.50390625, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 363.6, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 64.2, "frames": {"chat": 330}, "mem_gb": 21.99}
+{"step": 96, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.10431248163531806, "tokens": 120000, "cumulative_loss_tokens": 11520000, "grad_norm": 0.470703125, "lr": 3e-05, "finish_rate": 0.991, "comp_len": 359.3, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 62.3, "frames": {"chat": 334}, "mem_gb": 21.71}
+{"step": 97, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.1091810320273197, "tokens": 120000, "cumulative_loss_tokens": 11640000, "grad_norm": 0.484375, "lr": 3e-05, "finish_rate": 0.967, "comp_len": 357.1, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 58.5, "frames": {"chat": 336}, "mem_gb": 21.85}
+{"step": 98, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.0820784143627466, "tokens": 120000, "cumulative_loss_tokens": 11760000, "grad_norm": 0.443359375, "lr": 3e-05, "finish_rate": 0.964, "comp_len": 388.3, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 56.5, "frames": {"chat": 309}, "mem_gb": 21.93}
+{"step": 99, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.1385175480405645, "tokens": 120000, "cumulative_loss_tokens": 11880000, "grad_norm": 0.5234375, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 306.9, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 68.9, "frames": {"chat": 391}, "mem_gb": 21.82}
+{"step": 100, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.07162473855441591, "tokens": 120000, "cumulative_loss_tokens": 12000000, "grad_norm": 0.3671875, "lr": 3e-05, "finish_rate": 0.83, "comp_len": 454.5, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 57.5, "frames": {"chat": 264}, "mem_gb": 22.22}
+{"step": 101, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.12676862803295566, "tokens": 120000, "cumulative_loss_tokens": 12120000, "grad_norm": 0.5, "lr": 3e-05, "finish_rate": 0.994, "comp_len": 343.8, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 73.8, "frames": {"chat": 349}, "mem_gb": 21.61}
+{"step": 102, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.12114182180125888, "tokens": 120000, "cumulative_loss_tokens": 12240000, "grad_norm": 0.51171875, "lr": 3e-05, "finish_rate": 0.968, "comp_len": 384.6, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 58.3, "frames": {"chat": 312}, "mem_gb": 21.82}
+{"step": 103, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.03901445078571172, "tokens": 120000, "cumulative_loss_tokens": 12360000, "grad_norm": 0.314453125, "lr": 3e-05, "finish_rate": 0.817, "comp_len": 524.0, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 48.7, "frames": {"chat": 229}, "mem_gb": 22.09}
+{"step": 104, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.1029369749694597, "tokens": 120000, "cumulative_loss_tokens": 12480000, "grad_norm": 0.4453125, "lr": 3e-05, "finish_rate": 0.938, "comp_len": 416.7, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 57.0, "frames": {"chat": 288}, "mem_gb": 22.09}
+{"step": 105, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.14577210398303966, "tokens": 120000, "cumulative_loss_tokens": 12600000, "grad_norm": 0.515625, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 341.9, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 64.0, "frames": {"chat": 351}, "mem_gb": 21.75}
+{"step": 106, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.042434302429893676, "tokens": 120000, "cumulative_loss_tokens": 12720000, "grad_norm": 0.341796875, "lr": 3e-05, "finish_rate": 0.812, "comp_len": 538.1, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 46.0, "frames": {"chat": 223}, "mem_gb": 22.12}
+{"step": 107, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.023880430516523, "tokens": 120000, "cumulative_loss_tokens": 12840000, "grad_norm": 0.30078125, "lr": 3e-05, "finish_rate": 0.793, "comp_len": 563.4, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 45.2, "frames": {"chat": 213}, "mem_gb": 22.08}
+{"step": 108, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.1253991394669128, "tokens": 120000, "cumulative_loss_tokens": 12960000, "grad_norm": 0.51171875, "lr": 3e-05, "finish_rate": 0.995, "comp_len": 315.8, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 68.9, "frames": {"chat": 380}, "mem_gb": 21.93}
+{"step": 109, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.07316874562174393, "tokens": 120000, "cumulative_loss_tokens": 13080000, "grad_norm": 0.4140625, "lr": 3e-05, "finish_rate": 0.893, "comp_len": 442.8, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 50.6, "frames": {"chat": 271}, "mem_gb": 22.11}
+{"step": 110, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.15606389061900167, "tokens": 120000, "cumulative_loss_tokens": 13200000, "grad_norm": 0.50390625, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 352.9, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 62.9, "frames": {"chat": 340}, "mem_gb": 22.01}
+{"step": 111, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.14502378453067505, "tokens": 120000, "cumulative_loss_tokens": 13320000, "grad_norm": 0.5234375, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 327.9, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 67.8, "frames": {"chat": 366}, "mem_gb": 21.7}
+{"step": 112, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.044772777596488594, "tokens": 120000, "cumulative_loss_tokens": 13440000, "grad_norm": 0.326171875, "lr": 3e-05, "finish_rate": 0.854, "comp_len": 515.0, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 47.5, "frames": {"chat": 233}, "mem_gb": 22.09}
+{"step": 113, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.09647136719362268, "tokens": 120000, "cumulative_loss_tokens": 13560000, "grad_norm": 0.45703125, "lr": 3e-05, "finish_rate": 0.968, "comp_len": 384.6, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 58.7, "frames": {"chat": 312}, "mem_gb": 21.96}
+{"step": 114, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.14180562431244956, "tokens": 120000, "cumulative_loss_tokens": 13680000, "grad_norm": 0.55859375, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 330.6, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 64.9, "frames": {"chat": 363}, "mem_gb": 21.67}
+{"step": 115, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.14334856677559826, "tokens": 120000, "cumulative_loss_tokens": 13800000, "grad_norm": 0.51171875, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 371.5, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 60.1, "frames": {"chat": 323}, "mem_gb": 21.94}
+{"step": 116, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.14154685637778602, "tokens": 120000, "cumulative_loss_tokens": 13920000, "grad_norm": 0.55859375, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 346.8, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 64.4, "frames": {"chat": 346}, "mem_gb": 21.69}
+{"step": 117, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.1396051885420767, "tokens": 120000, "cumulative_loss_tokens": 14040000, "grad_norm": 0.5, "lr": 3e-05, "finish_rate": 0.975, "comp_len": 377.4, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 58.0, "frames": {"chat": 318}, "mem_gb": 21.83}
+{"step": 118, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.11092782069956884, "tokens": 120000, "cumulative_loss_tokens": 14160000, "grad_norm": 0.455078125, "lr": 3e-05, "finish_rate": 0.935, "comp_len": 436.4, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 58.1, "frames": {"chat": 275}, "mem_gb": 22.02}
+{"step": 119, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.12665537766687726, "tokens": 120000, "cumulative_loss_tokens": 14280000, "grad_norm": 0.4921875, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 347.8, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 64.8, "frames": {"chat": 345}, "mem_gb": 21.76}
+{"step": 120, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.120160753988754, "tokens": 120000, "cumulative_loss_tokens": 14400000, "grad_norm": 0.53515625, "lr": 3e-05, "finish_rate": 0.992, "comp_len": 318.3, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 66.6, "frames": {"chat": 377}, "mem_gb": 21.91}
+{"step": 121, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.09895123931532726, "tokens": 120000, "cumulative_loss_tokens": 14520000, "grad_norm": 0.43359375, "lr": 3e-05, "finish_rate": 0.956, "comp_len": 408.2, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 57.6, "frames": {"chat": 294}, "mem_gb": 22.14}
+{"step": 122, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.1409571954037218, "tokens": 120000, "cumulative_loss_tokens": 14640000, "grad_norm": 0.51171875, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 355.0, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 63.2, "frames": {"chat": 338}, "mem_gb": 21.83}
+{"step": 123, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.15211505075778503, "tokens": 120000, "cumulative_loss_tokens": 14760000, "grad_norm": 0.51953125, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 340.9, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 66.0, "frames": {"chat": 352}, "mem_gb": 21.73}
+{"step": 124, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.1332457024556352, "tokens": 120000, "cumulative_loss_tokens": 14880000, "grad_norm": 0.49609375, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 345.8, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 67.9, "frames": {"chat": 347}, "mem_gb": 21.78}
+{"step": 125, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.08373471758464972, "tokens": 120000, "cumulative_loss_tokens": 15000000, "grad_norm": 0.404296875, "lr": 3e-05, "finish_rate": 0.968, "comp_len": 387.1, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 57.8, "frames": {"chat": 310}, "mem_gb": 21.76}
+{"step": 126, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.15742334490390494, "tokens": 120000, "cumulative_loss_tokens": 15120000, "grad_norm": 0.578125, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 335.2, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 66.3, "frames": {"chat": 358}, "mem_gb": 21.78}
+{"step": 127, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.07713352295395452, "tokens": 120000, "cumulative_loss_tokens": 15240000, "grad_norm": 0.427734375, "lr": 3e-05, "finish_rate": 0.947, "comp_len": 376.2, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 59.3, "frames": {"chat": 319}, "mem_gb": 22.06}
+{"step": 128, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.052739765353671586, "tokens": 120000, "cumulative_loss_tokens": 15360000, "grad_norm": 0.3515625, "lr": 3e-05, "finish_rate": 0.9, "comp_len": 446.1, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 54.5, "frames": {"chat": 269}, "mem_gb": 22.09}
+{"step": 129, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.10360154020698586, "tokens": 120000, "cumulative_loss_tokens": 15480000, "grad_norm": 0.466796875, "lr": 3e-05, "finish_rate": 0.974, "comp_len": 387.1, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 63.2, "frames": {"chat": 310}, "mem_gb": 21.94}
+{"step": 130, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.06664360370513363, "tokens": 120000, "cumulative_loss_tokens": 15600000, "grad_norm": 0.41796875, "lr": 3e-05, "finish_rate": 0.834, "comp_len": 463.3, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 55.2, "frames": {"chat": 259}, "mem_gb": 22.12}
+{"step": 131, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.06523915852874827, "tokens": 120000, "cumulative_loss_tokens": 15720000, "grad_norm": 0.41796875, "lr": 3e-05, "finish_rate": 0.868, "comp_len": 452.8, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 53.8, "frames": {"chat": 265}, "mem_gb": 22.1}
+{"step": 132, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.1503767799893394, "tokens": 120000, "cumulative_loss_tokens": 15840000, "grad_norm": 0.546875, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 356.1, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 66.1, "frames": {"chat": 337}, "mem_gb": 21.65}
+{"step": 133, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.07360061740261348, "tokens": 120000, "cumulative_loss_tokens": 15960000, "grad_norm": 0.419921875, "lr": 3e-05, "finish_rate": 0.972, "comp_len": 413.8, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 58.9, "frames": {"chat": 290}, "mem_gb": 21.91}
+{"step": 134, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.12301088459033054, "tokens": 120000, "cumulative_loss_tokens": 16080000, "grad_norm": 0.50390625, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 343.8, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 66.8, "frames": {"chat": 349}, "mem_gb": 21.84}
+{"step": 135, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.07363978966639066, "tokens": 120000, "cumulative_loss_tokens": 16200000, "grad_norm": 0.404296875, "lr": 3e-05, "finish_rate": 0.928, "comp_len": 412.4, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 57.8, "frames": {"chat": 291}, "mem_gb": 21.94}
+{"step": 136, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.12913179965493424, "tokens": 120000, "cumulative_loss_tokens": 16320000, "grad_norm": 0.515625, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 350.9, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 66.5, "frames": {"chat": 342}, "mem_gb": 21.8}
+{"step": 137, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.17013440595478752, "tokens": 120000, "cumulative_loss_tokens": 16440000, "grad_norm": 0.55078125, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 355.0, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 64.0, "frames": {"chat": 338}, "mem_gb": 21.75}
+{"step": 138, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.14697195186861792, "tokens": 120000, "cumulative_loss_tokens": 16560000, "grad_norm": 0.5234375, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 337.1, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 70.2, "frames": {"chat": 356}, "mem_gb": 21.71}
+{"step": 139, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.06590260987370275, "tokens": 120000, "cumulative_loss_tokens": 16680000, "grad_norm": 0.39453125, "lr": 3e-05, "finish_rate": 0.86, "comp_len": 452.8, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 59.6, "frames": {"chat": 265}, "mem_gb": 22.09}
+{"step": 140, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.11447063266391246, "tokens": 120000, "cumulative_loss_tokens": 16800000, "grad_norm": 0.498046875, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 339.9, "t_data_s": 0.1, "t_rollout_s": 0.0, "t_step_s": 68.9, "frames": {"chat": 353}, "mem_gb": 21.81}
+{"step": 141, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.11048259019283578, "tokens": 120000, "cumulative_loss_tokens": 16920000, "grad_norm": 0.498046875, "lr": 3e-05, "finish_rate": 0.949, "comp_len": 379.7, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 62.9, "frames": {"chat": 316}, "mem_gb": 21.99}
+{"step": 142, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.09986399798836404, "tokens": 120000, "cumulative_loss_tokens": 17040000, "grad_norm": 0.455078125, "lr": 3e-05, "finish_rate": 0.945, "comp_len": 415.2, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 60.7, "frames": {"chat": 289}, "mem_gb": 22.14}
+{"step": 143, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.0992997664115702, "tokens": 120000, "cumulative_loss_tokens": 17160000, "grad_norm": 0.46484375, "lr": 3e-05, "finish_rate": 0.982, "comp_len": 362.5, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 65.8, "frames": {"chat": 331}, "mem_gb": 21.78}
+{"step": 144, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.09031849862008046, "tokens": 120000, "cumulative_loss_tokens": 17280000, "grad_norm": 0.42578125, "lr": 3e-05, "finish_rate": 0.972, "comp_len": 375.0, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 61.6, "frames": {"chat": 320}, "mem_gb": 22.1}
+{"step": 145, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.08688670156883696, "tokens": 120000, "cumulative_loss_tokens": 17400000, "grad_norm": 0.380859375, "lr": 3e-05, "finish_rate": 0.941, "comp_len": 441.2, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 53.3, "frames": {"chat": 272}, "mem_gb": 21.83}
+{"step": 146, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.17230489805215815, "tokens": 120000, "cumulative_loss_tokens": 17520000, "grad_norm": 0.56640625, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 326.1, "t_data_s": 0.1, "t_rollout_s": 0.0, "t_step_s": 71.0, "frames": {"chat": 368}, "mem_gb": 21.75}
+{"step": 147, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.12812672173778217, "tokens": 120000, "cumulative_loss_tokens": 17640000, "grad_norm": 0.478515625, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 365.9, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 65.5, "frames": {"chat": 328}, "mem_gb": 21.78}
+{"step": 148, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.12114051378740308, "tokens": 120000, "cumulative_loss_tokens": 17760000, "grad_norm": 0.51953125, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 368.1, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 64.5, "frames": {"chat": 326}, "mem_gb": 21.99}
+{"step": 149, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.1461482815605433, "tokens": 120000, "cumulative_loss_tokens": 17880000, "grad_norm": 0.5, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 329.7, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 72.8, "frames": {"chat": 364}, "mem_gb": 21.93}
+{"step": 150, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.09341100072071422, "tokens": 120000, "cumulative_loss_tokens": 18000000, "grad_norm": 0.412109375, "lr": 3e-05, "finish_rate": 0.97, "comp_len": 405.4, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 60.8, "frames": {"chat": 296}, "mem_gb": 21.98}
+{"step": 151, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.13002619124348275, "tokens": 120000, "cumulative_loss_tokens": 18120000, "grad_norm": 0.5078125, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 317.5, "t_data_s": 0.1, "t_rollout_s": 0.0, "t_step_s": 73.4, "frames": {"chat": 378}, "mem_gb": 21.83}
+{"step": 152, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.1715810147296327, "tokens": 120000, "cumulative_loss_tokens": 18240000, "grad_norm": 0.5703125, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 326.1, "t_data_s": 0.1, "t_rollout_s": 0.0, "t_step_s": 72.4, "frames": {"chat": 368}, "mem_gb": 21.93}
+{"step": 153, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.07034443499888682, "tokens": 120000, "cumulative_loss_tokens": 18360000, "grad_norm": 0.390625, "lr": 3e-05, "finish_rate": 0.898, "comp_len": 468.8, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 56.5, "frames": {"chat": 256}, "mem_gb": 22.03}
+{"step": 154, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.08062464235329535, "tokens": 120000, "cumulative_loss_tokens": 18480000, "grad_norm": 0.423828125, "lr": 3e-05, "finish_rate": 0.93, "comp_len": 381.0, "t_data_s": 0.1, "t_rollout_s": 0.0, "t_step_s": 62.1, "frames": {"chat": 315}, "mem_gb": 21.98}
+{"step": 155, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.07058507993586827, "tokens": 120000, "cumulative_loss_tokens": 18600000, "grad_norm": 0.384765625, "lr": 3e-05, "finish_rate": 0.974, "comp_len": 384.6, "t_data_s": 0.1, "t_rollout_s": 0.0, "t_step_s": 62.6, "frames": {"chat": 312}, "mem_gb": 21.91}
+{"step": 156, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.11778950736102803, "tokens": 120000, "cumulative_loss_tokens": 18720000, "grad_norm": 0.462890625, "lr": 3e-05, "finish_rate": 0.97, "comp_len": 363.6, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 65.6, "frames": {"chat": 330}, "mem_gb": 22.09}
+{"step": 157, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.09653343385270952, "tokens": 120000, "cumulative_loss_tokens": 18840000, "grad_norm": 0.435546875, "lr": 3e-05, "finish_rate": 0.954, "comp_len": 367.0, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 67.3, "frames": {"chat": 327}, "mem_gb": 22.09}
+{"step": 158, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.13204492282280697, "tokens": 120000, "cumulative_loss_tokens": 18960000, "grad_norm": 0.48046875, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 341.9, "t_data_s": 0.1, "t_rollout_s": 0.0, "t_step_s": 72.3, "frames": {"chat": 351}, "mem_gb": 21.75}
+{"step": 159, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.07468217376978913, "tokens": 120000, "cumulative_loss_tokens": 19080000, "grad_norm": 0.396484375, "lr": 3e-05, "finish_rate": 0.973, "comp_len": 357.1, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 66.4, "frames": {"chat": 336}, "mem_gb": 21.8}
+{"step": 160, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.13069434770564547, "tokens": 120000, "cumulative_loss_tokens": 19200000, "grad_norm": 0.51171875, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 332.4, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 74.0, "frames": {"chat": 361}, "mem_gb": 21.78}
+{"step": 161, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.14887345913428193, "tokens": 120000, "cumulative_loss_tokens": 19320000, "grad_norm": 0.486328125, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 328.8, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 75.2, "frames": {"chat": 365}, "mem_gb": 21.72}
+{"step": 162, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.13350603818915163, "tokens": 120000, "cumulative_loss_tokens": 19440000, "grad_norm": 0.486328125, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 364.7, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 64.7, "frames": {"chat": 329}, "mem_gb": 21.78}
+{"step": 163, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.09072161397730621, "tokens": 120000, "cumulative_loss_tokens": 19560000, "grad_norm": 0.42578125, "lr": 3e-05, "finish_rate": 0.931, "comp_len": 416.7, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 59.5, "frames": {"chat": 288}, "mem_gb": 21.99}
+{"step": 164, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.06492729764790274, "tokens": 120000, "cumulative_loss_tokens": 19680000, "grad_norm": 0.41796875, "lr": 3e-05, "finish_rate": 0.868, "comp_len": 451.1, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 59.0, "frames": {"chat": 266}, "mem_gb": 22.14}
+{"step": 165, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.11315758886872791, "tokens": 120000, "cumulative_loss_tokens": 19800000, "grad_norm": 0.4921875, "lr": 3e-05, "finish_rate": 0.988, "comp_len": 372.7, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 58.5, "frames": {"chat": 322}, "mem_gb": 22.05}
+{"step": 166, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.1163916651834889, "tokens": 120000, "cumulative_loss_tokens": 19920000, "grad_norm": 0.47265625, "lr": 3e-05, "finish_rate": 0.989, "comp_len": 334.3, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 62.8, "frames": {"chat": 359}, "mem_gb": 21.75}
+{"step": 167, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.09991620500825811, "tokens": 120000, "cumulative_loss_tokens": 20040000, "grad_norm": 0.447265625, "lr": 3e-05, "finish_rate": 0.965, "comp_len": 385.9, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 58.9, "frames": {"chat": 311}, "mem_gb": 21.88}
+{"step": 168, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.08487513831759182, "tokens": 120000, "cumulative_loss_tokens": 20160000, "grad_norm": 0.408203125, "lr": 3e-05, "finish_rate": 0.95, "comp_len": 396.0, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 55.4, "frames": {"chat": 303}, "mem_gb": 21.93}
+{"step": 169, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.15273571721582363, "tokens": 120000, "cumulative_loss_tokens": 20280000, "grad_norm": 0.52734375, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 351.9, "t_data_s": 0.1, "t_rollout_s": 0.0, "t_step_s": 68.0, "frames": {"chat": 341}, "mem_gb": 21.94}
+{"step": 170, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.0354515667994041, "tokens": 120000, "cumulative_loss_tokens": 20400000, "grad_norm": 0.357421875, "lr": 3e-05, "finish_rate": 0.843, "comp_len": 524.0, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 47.7, "frames": {"chat": 229}, "mem_gb": 22.0}
+{"step": 171, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.0888629938938267, "tokens": 120000, "cumulative_loss_tokens": 20520000, "grad_norm": 0.423828125, "lr": 3e-05, "finish_rate": 0.964, "comp_len": 389.6, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 60.9, "frames": {"chat": 308}, "mem_gb": 22.16}
+{"step": 172, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.1270647643733304, "tokens": 120000, "cumulative_loss_tokens": 20640000, "grad_norm": 0.5078125, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 331.5, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 66.9, "frames": {"chat": 362}, "mem_gb": 21.65}
+{"step": 173, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.1342932996113319, "tokens": 120000, "cumulative_loss_tokens": 20760000, "grad_norm": 0.5234375, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 315.8, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 73.5, "frames": {"chat": 380}, "mem_gb": 21.9}
+{"step": 174, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.12381919350024934, "tokens": 120000, "cumulative_loss_tokens": 20880000, "grad_norm": 0.51953125, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 336.1, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 66.0, "frames": {"chat": 357}, "mem_gb": 21.68}
+{"step": 175, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.06805174825843424, "tokens": 120000, "cumulative_loss_tokens": 21000000, "grad_norm": 0.40625, "lr": 3e-05, "finish_rate": 0.835, "comp_len": 472.4, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 52.3, "frames": {"chat": 254}, "mem_gb": 22.16}
+{"step": 176, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.05205729084465808, "tokens": 120000, "cumulative_loss_tokens": 21120000, "grad_norm": 0.34765625, "lr": 3e-05, "finish_rate": 0.874, "comp_len": 459.8, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 52.0, "frames": {"chat": 261}, "mem_gb": 22.12}
+{"step": 177, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.1126549622541992, "tokens": 120000, "cumulative_loss_tokens": 21240000, "grad_norm": 0.4765625, "lr": 3e-05, "finish_rate": 0.982, "comp_len": 431.7, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 54.5, "frames": {"chat": 278}, "mem_gb": 21.91}
+{"step": 178, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.12549313363223336, "tokens": 120000, "cumulative_loss_tokens": 21360000, "grad_norm": 0.474609375, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 360.4, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 63.5, "frames": {"chat": 333}, "mem_gb": 22.05}
+{"step": 179, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.066560423706177, "tokens": 120000, "cumulative_loss_tokens": 21480000, "grad_norm": 0.447265625, "lr": 3e-05, "finish_rate": 0.888, "comp_len": 480.0, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 51.7, "frames": {"chat": 250}, "mem_gb": 22.02}
+{"step": 180, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.1541350697507616, "tokens": 120000, "cumulative_loss_tokens": 21600000, "grad_norm": 0.55078125, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 336.1, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 66.5, "frames": {"chat": 357}, "mem_gb": 21.79}
+{"step": 181, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.1540505784819601, "tokens": 120000, "cumulative_loss_tokens": 21720000, "grad_norm": 0.5390625, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 363.6, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 61.7, "frames": {"chat": 330}, "mem_gb": 21.76}
+{"step": 182, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.12838531766707698, "tokens": 120000, "cumulative_loss_tokens": 21840000, "grad_norm": 0.51171875, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 341.9, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 63.0, "frames": {"chat": 351}, "mem_gb": 21.8}
+{"step": 183, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.08405779912746511, "tokens": 120000, "cumulative_loss_tokens": 21960000, "grad_norm": 0.455078125, "lr": 3e-05, "finish_rate": 0.959, "comp_len": 413.8, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 58.7, "frames": {"chat": 290}, "mem_gb": 21.93}
+{"step": 184, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.1232834550240698, "tokens": 120000, "cumulative_loss_tokens": 22080000, "grad_norm": 0.486328125, "lr": 3e-05, "finish_rate": 0.969, "comp_len": 372.7, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 65.7, "frames": {"chat": 322}, "mem_gb": 21.87}
+{"step": 185, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.056777896666216354, "tokens": 120000, "cumulative_loss_tokens": 22200000, "grad_norm": 0.37890625, "lr": 3e-05, "finish_rate": 0.827, "comp_len": 451.1, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 61.1, "frames": {"chat": 266}, "mem_gb": 22.1}
+{"step": 186, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.09375398674480773, "tokens": 120000, "cumulative_loss_tokens": 22320000, "grad_norm": 0.46875, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 362.5, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 68.1, "frames": {"chat": 331}, "mem_gb": 21.96}
+{"step": 187, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.08996357367694534, "tokens": 120000, "cumulative_loss_tokens": 22440000, "grad_norm": 0.4453125, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 350.9, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 68.9, "frames": {"chat": 342}, "mem_gb": 21.72}
+{"step": 188, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.09844830975172421, "tokens": 120000, "cumulative_loss_tokens": 22560000, "grad_norm": 0.55078125, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 335.2, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 70.0, "frames": {"chat": 358}, "mem_gb": 21.75}
+{"step": 189, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.0910161497346436, "tokens": 120000, "cumulative_loss_tokens": 22680000, "grad_norm": 0.443359375, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 348.8, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 72.4, "frames": {"chat": 344}, "mem_gb": 21.84}
+{"step": 190, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.04903289433984707, "tokens": 120000, "cumulative_loss_tokens": 22800000, "grad_norm": 0.375, "lr": 3e-05, "finish_rate": 0.861, "comp_len": 449.4, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 61.5, "frames": {"chat": 267}, "mem_gb": 22.15}
+{"step": 191, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.02093672791371743, "tokens": 120000, "cumulative_loss_tokens": 22920000, "grad_norm": 0.216796875, "lr": 3e-05, "finish_rate": 0.835, "comp_len": 483.9, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 54.5, "frames": {"chat": 248}, "mem_gb": 21.99}
+{"step": 192, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.07855782948022631, "tokens": 120000, "cumulative_loss_tokens": 23040000, "grad_norm": 0.42578125, "lr": 3e-05, "finish_rate": 0.979, "comp_len": 364.7, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 66.9, "frames": {"chat": 329}, "mem_gb": 21.98}
+{"step": 193, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.08327089551646884, "tokens": 120000, "cumulative_loss_tokens": 23160000, "grad_norm": 0.41796875, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 354.0, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 70.5, "frames": {"chat": 339}, "mem_gb": 21.94}
+{"step": 194, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.09215007556722654, "tokens": 120000, "cumulative_loss_tokens": 23280000, "grad_norm": 0.419921875, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 339.0, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 71.8, "frames": {"chat": 354}, "mem_gb": 21.94}
+{"step": 195, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.05571042603639653, "tokens": 120000, "cumulative_loss_tokens": 23400000, "grad_norm": 0.3359375, "lr": 3e-05, "finish_rate": 0.943, "comp_len": 430.1, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 58.1, "frames": {"chat": 279}, "mem_gb": 22.07}
+{"step": 196, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.06696861219073956, "tokens": 120000, "cumulative_loss_tokens": 23520000, "grad_norm": 0.439453125, "lr": 3e-05, "finish_rate": 0.959, "comp_len": 377.4, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 66.2, "frames": {"chat": 318}, "mem_gb": 22.0}
+{"step": 197, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.048480764170961144, "tokens": 120000, "cumulative_loss_tokens": 23640000, "grad_norm": 0.345703125, "lr": 3e-05, "finish_rate": 0.875, "comp_len": 452.8, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 58.6, "frames": {"chat": 265}, "mem_gb": 22.09}
+{"step": 198, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.043221544861785754, "tokens": 120000, "cumulative_loss_tokens": 23760000, "grad_norm": 0.30078125, "lr": 3e-05, "finish_rate": 0.85, "comp_len": 472.4, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 55.3, "frames": {"chat": 254}, "mem_gb": 22.03}
+{"step": 199, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.06468877917430751, "tokens": 120000, "cumulative_loss_tokens": 23880000, "grad_norm": 0.384765625, "lr": 3e-05, "finish_rate": 0.962, "comp_len": 381.0, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 66.1, "frames": {"chat": 315}, "mem_gb": 22.06}
+{"step": 200, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.08913087736829184, "tokens": 120000, "cumulative_loss_tokens": 24000000, "grad_norm": 0.435546875, "lr": 3e-05, "finish_rate": 0.977, "comp_len": 400.0, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 61.7, "frames": {"chat": 300}, "mem_gb": 21.85}
+{"step": 201, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.06372527353525317, "tokens": 120000, "cumulative_loss_tokens": 24120000, "grad_norm": 0.373046875, "lr": 3e-05, "finish_rate": 0.924, "comp_len": 434.8, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 67.5, "frames": {"chat": 276}, "mem_gb": 21.9}
+{"step": 202, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.10188659057969537, "tokens": 120000, "cumulative_loss_tokens": 24240000, "grad_norm": 0.474609375, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 337.1, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 67.9, "frames": {"chat": 356}, "mem_gb": 22.0}
+{"step": 203, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.08538098230928493, "tokens": 120000, "cumulative_loss_tokens": 24360000, "grad_norm": 0.443359375, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 394.7, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 56.9, "frames": {"chat": 304}, "mem_gb": 21.81}
+{"step": 204, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.08974360643521262, "tokens": 120000, "cumulative_loss_tokens": 24480000, "grad_norm": 0.4140625, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 320.0, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 68.2, "frames": {"chat": 375}, "mem_gb": 21.73}
+{"step": 205, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.04663258421965099, "tokens": 120000, "cumulative_loss_tokens": 24600000, "grad_norm": 0.33984375, "lr": 3e-05, "finish_rate": 0.939, "comp_len": 408.2, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 56.1, "frames": {"chat": 294}, "mem_gb": 22.09}
+{"step": 206, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.08282181139490567, "tokens": 120000, "cumulative_loss_tokens": 24720000, "grad_norm": 0.37890625, "lr": 3e-05, "finish_rate": 0.98, "comp_len": 401.3, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 60.9, "frames": {"chat": 299}, "mem_gb": 22.1}
+{"step": 207, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.06012468190827252, "tokens": 120000, "cumulative_loss_tokens": 24840000, "grad_norm": 0.341796875, "lr": 3e-05, "finish_rate": 0.98, "comp_len": 393.4, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 59.3, "frames": {"chat": 305}, "mem_gb": 21.88}
+{"step": 208, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.04234457039027087, "tokens": 120000, "cumulative_loss_tokens": 24960000, "grad_norm": 0.279296875, "lr": 3e-05, "finish_rate": 0.931, "comp_len": 458.0, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 50.8, "frames": {"chat": 262}, "mem_gb": 21.93}
+{"step": 209, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.05611148730162143, "tokens": 120000, "cumulative_loss_tokens": 25080000, "grad_norm": 0.390625, "lr": 3e-05, "finish_rate": 0.95, "comp_len": 396.0, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 53.9, "frames": {"chat": 303}, "mem_gb": 21.99}
+{"step": 210, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.09454328215078761, "tokens": 120000, "cumulative_loss_tokens": 25200000, "grad_norm": 0.46484375, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 315.8, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 66.2, "frames": {"chat": 380}, "mem_gb": 21.87}
+{"step": 211, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.037079838351839374, "tokens": 120000, "cumulative_loss_tokens": 25320000, "grad_norm": 0.2734375, "lr": 3e-05, "finish_rate": 0.95, "comp_len": 425.5, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 53.7, "frames": {"chat": 282}, "mem_gb": 21.83}
+{"step": 212, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.07862069786739691, "tokens": 120000, "cumulative_loss_tokens": 25440000, "grad_norm": 0.365234375, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 331.5, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 67.9, "frames": {"chat": 362}, "mem_gb": 21.84}
+{"step": 213, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.07989904291653074, "tokens": 120000, "cumulative_loss_tokens": 25560000, "grad_norm": 0.408203125, "lr": 3e-05, "finish_rate": 0.967, "comp_len": 390.9, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 60.1, "frames": {"chat": 307}, "mem_gb": 22.08}
+{"step": 214, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.0554238200636503, "tokens": 120000, "cumulative_loss_tokens": 25680000, "grad_norm": 0.322265625, "lr": 3e-05, "finish_rate": 0.926, "comp_len": 425.5, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 54.9, "frames": {"chat": 282}, "mem_gb": 21.95}
+{"step": 215, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.08592917260187678, "tokens": 120000, "cumulative_loss_tokens": 25800000, "grad_norm": 0.427734375, "lr": 3e-05, "finish_rate": 0.979, "comp_len": 356.1, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 62.8, "frames": {"chat": 337}, "mem_gb": 21.98}
+{"step": 216, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.048842965997181215, "tokens": 120000, "cumulative_loss_tokens": 25920000, "grad_norm": 0.3125, "lr": 3e-05, "finish_rate": 0.891, "comp_len": 436.4, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 60.4, "frames": {"chat": 275}, "mem_gb": 22.03}
+{"step": 217, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.08592547073025877, "tokens": 120000, "cumulative_loss_tokens": 26040000, "grad_norm": 0.392578125, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 377.4, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 65.9, "frames": {"chat": 318}, "mem_gb": 21.85}
+{"step": 218, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.10503062903866327, "tokens": 120000, "cumulative_loss_tokens": 26160000, "grad_norm": 0.427734375, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 311.7, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 73.8, "frames": {"chat": 385}, "mem_gb": 21.87}
+{"step": 219, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.09659696672237478, "tokens": 120000, "cumulative_loss_tokens": 26280000, "grad_norm": 0.435546875, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 379.7, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 66.1, "frames": {"chat": 316}, "mem_gb": 21.82}
+{"step": 220, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.03816706270910023, "tokens": 120000, "cumulative_loss_tokens": 26400000, "grad_norm": 0.294921875, "lr": 3e-05, "finish_rate": 0.887, "comp_len": 451.1, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 61.5, "frames": {"chat": 266}, "mem_gb": 22.11}
+{"step": 221, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.09084978325705355, "tokens": 120000, "cumulative_loss_tokens": 26520000, "grad_norm": 0.439453125, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 323.5, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 75.8, "frames": {"chat": 371}, "mem_gb": 21.85}
+{"step": 222, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.043425673709320836, "tokens": 120000, "cumulative_loss_tokens": 26640000, "grad_norm": 0.291015625, "lr": 3e-05, "finish_rate": 0.878, "comp_len": 458.0, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 59.9, "frames": {"chat": 262}, "mem_gb": 22.05}
+{"step": 223, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.10171279898905196, "tokens": 120000, "cumulative_loss_tokens": 26760000, "grad_norm": 0.50390625, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 344.8, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 71.9, "frames": {"chat": 348}, "mem_gb": 21.88}
+{"step": 224, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.018827865882070426, "tokens": 120000, "cumulative_loss_tokens": 26880000, "grad_norm": 0.255859375, "lr": 3e-05, "finish_rate": 0.811, "comp_len": 540.5, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 51.3, "frames": {"chat": 222}, "mem_gb": 22.12}
+{"step": 225, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.08322445326047795, "tokens": 120000, "cumulative_loss_tokens": 27000000, "grad_norm": 0.46875, "lr": 3e-05, "finish_rate": 0.99, "comp_len": 294.1, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 76.3, "frames": {"chat": 408}, "mem_gb": 21.68}
+{"step": 226, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.08017959230930234, "tokens": 120000, "cumulative_loss_tokens": 27120000, "grad_norm": 0.375, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 359.3, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 66.9, "frames": {"chat": 334}, "mem_gb": 21.75}
+{"step": 227, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.03997079523146385, "tokens": 120000, "cumulative_loss_tokens": 27240000, "grad_norm": 0.27734375, "lr": 3e-05, "finish_rate": 0.968, "comp_len": 384.6, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 61.4, "frames": {"chat": 312}, "mem_gb": 21.79}
+{"step": 228, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.08290009895348921, "tokens": 120000, "cumulative_loss_tokens": 27360000, "grad_norm": 0.392578125, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 340.9, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 70.8, "frames": {"chat": 352}, "mem_gb": 21.78}
+{"step": 229, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.0663638661310232, "tokens": 120000, "cumulative_loss_tokens": 27480000, "grad_norm": 0.39453125, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 323.5, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 70.9, "frames": {"chat": 371}, "mem_gb": 21.59}
+{"step": 230, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.046608242106852896, "tokens": 120000, "cumulative_loss_tokens": 27600000, "grad_norm": 0.33984375, "lr": 3e-05, "finish_rate": 0.898, "comp_len": 422.5, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 62.0, "frames": {"chat": 284}, "mem_gb": 22.04}
+{"step": 231, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.040455189670513696, "tokens": 120000, "cumulative_loss_tokens": 27720000, "grad_norm": 0.3359375, "lr": 3e-05, "finish_rate": 0.923, "comp_len": 418.1, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 57.0, "frames": {"chat": 287}, "mem_gb": 22.1}
+{"step": 232, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.03750597964787157, "tokens": 120000, "cumulative_loss_tokens": 27840000, "grad_norm": 0.318359375, "lr": 3e-05, "finish_rate": 0.916, "comp_len": 421.1, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 57.6, "frames": {"chat": 285}, "mem_gb": 22.0}
+{"step": 233, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.021991027936533403, "tokens": 120000, "cumulative_loss_tokens": 27960000, "grad_norm": 0.2373046875, "lr": 3e-05, "finish_rate": 0.873, "comp_len": 476.2, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 53.6, "frames": {"chat": 252}, "mem_gb": 22.09}
+{"step": 234, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.08295298451466175, "tokens": 120000, "cumulative_loss_tokens": 28080000, "grad_norm": 0.4375, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 333.3, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 69.9, "frames": {"chat": 360}, "mem_gb": 21.88}
+{"step": 235, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.09032735534195478, "tokens": 120000, "cumulative_loss_tokens": 28200000, "grad_norm": 0.44921875, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 334.3, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 70.1, "frames": {"chat": 359}, "mem_gb": 21.87}
+{"step": 236, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.08670118659605117, "tokens": 120000, "cumulative_loss_tokens": 28320000, "grad_norm": 0.451171875, "lr": 3e-05, "finish_rate": 1.0, "comp_len": 330.6, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 70.8, "frames": {"chat": 363}, "mem_gb": 21.63}
+{"step": 237, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.07076076977982496, "tokens": 120000, "cumulative_loss_tokens": 28440000, "grad_norm": 0.365234375, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 332.4, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 70.6, "frames": {"chat": 361}, "mem_gb": 21.72}
+{"step": 238, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.04752216829869042, "tokens": 120000, "cumulative_loss_tokens": 28560000, "grad_norm": 0.3125, "lr": 3e-05, "finish_rate": 0.944, "comp_len": 419.6, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 61.2, "frames": {"chat": 286}, "mem_gb": 21.91}
+{"step": 239, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.0892048023262682, "tokens": 120000, "cumulative_loss_tokens": 28680000, "grad_norm": 0.447265625, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 327.0, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 75.2, "frames": {"chat": 367}, "mem_gb": 21.76}
+{"step": 240, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.08939793932631146, "tokens": 120000, "cumulative_loss_tokens": 28800000, "grad_norm": 0.390625, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 340.9, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 71.4, "frames": {"chat": 352}, "mem_gb": 21.79}
+{"step": 241, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.05280678555501315, "tokens": 120000, "cumulative_loss_tokens": 28920000, "grad_norm": 0.32421875, "lr": 3e-05, "finish_rate": 0.962, "comp_len": 382.2, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 63.7, "frames": {"chat": 314}, "mem_gb": 21.74}
+{"step": 242, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.08378173946699438, "tokens": 120000, "cumulative_loss_tokens": 29040000, "grad_norm": 0.45703125, "lr": 3e-05, "finish_rate": 0.995, "comp_len": 322.6, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 72.3, "frames": {"chat": 372}, "mem_gb": 21.71}
+{"step": 243, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.06951046424278369, "tokens": 120000, "cumulative_loss_tokens": 29160000, "grad_norm": 0.380859375, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 359.3, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 71.4, "frames": {"chat": 334}, "mem_gb": 21.86}
+{"step": 244, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.09616741288111079, "tokens": 120000, "cumulative_loss_tokens": 29280000, "grad_norm": 0.408203125, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 326.1, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 72.0, "frames": {"chat": 368}, "mem_gb": 21.8}
+{"step": 245, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.07002581880175664, "tokens": 120000, "cumulative_loss_tokens": 29400000, "grad_norm": 0.369140625, "lr": 3e-05, "finish_rate": 0.994, "comp_len": 339.0, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 71.7, "frames": {"chat": 354}, "mem_gb": 21.95}
+{"step": 246, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.03909036792227998, "tokens": 120000, "cumulative_loss_tokens": 29520000, "grad_norm": 0.333984375, "lr": 3e-05, "finish_rate": 0.973, "comp_len": 409.6, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 59.6, "frames": {"chat": 293}, "mem_gb": 21.84}
+{"step": 247, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.025099564243853092, "tokens": 120000, "cumulative_loss_tokens": 29640000, "grad_norm": 0.318359375, "lr": 3e-05, "finish_rate": 0.835, "comp_len": 519.5, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 54.6, "frames": {"chat": 231}, "mem_gb": 22.0}
+{"step": 248, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.0694033814901098, "tokens": 120000, "cumulative_loss_tokens": 29760000, "grad_norm": 0.470703125, "lr": 3e-05, "finish_rate": 0.964, "comp_len": 397.4, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 62.2, "frames": {"chat": 302}, "mem_gb": 21.97}
+{"step": 249, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.04470528662351426, "tokens": 120000, "cumulative_loss_tokens": 29880000, "grad_norm": 0.306640625, "lr": 3e-05, "finish_rate": 0.878, "comp_len": 456.3, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 61.0, "frames": {"chat": 263}, "mem_gb": 22.11}
+{"step": 250, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.045194222508526095, "tokens": 120000, "cumulative_loss_tokens": 30000000, "grad_norm": 0.310546875, "lr": 3e-05, "finish_rate": 0.889, "comp_len": 476.2, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 56.9, "frames": {"chat": 252}, "mem_gb": 22.04}
+{"step": 251, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.07064214115955401, "tokens": 120000, "cumulative_loss_tokens": 30120000, "grad_norm": 0.400390625, "lr": 3e-05, "finish_rate": 0.968, "comp_len": 348.8, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 66.9, "frames": {"chat": 344}, "mem_gb": 22.07}
+{"step": 252, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.06479810920925423, "tokens": 120000, "cumulative_loss_tokens": 30240000, "grad_norm": 0.361328125, "lr": 3e-05, "finish_rate": 0.981, "comp_len": 383.4, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 66.2, "frames": {"chat": 313}, "mem_gb": 21.99}
+{"step": 253, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.07711766247903773, "tokens": 120000, "cumulative_loss_tokens": 30360000, "grad_norm": 0.37890625, "lr": 3e-05, "finish_rate": 0.992, "comp_len": 335.2, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 68.1, "frames": {"chat": 358}, "mem_gb": 21.7}
+{"step": 254, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.03752173047271402, "tokens": 120000, "cumulative_loss_tokens": 30480000, "grad_norm": 0.28125, "lr": 3e-05, "finish_rate": 0.951, "comp_len": 449.4, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 55.3, "frames": {"chat": 267}, "mem_gb": 21.92}
+{"step": 255, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.06976982899419963, "tokens": 120000, "cumulative_loss_tokens": 30600000, "grad_norm": 0.369140625, "lr": 3e-05, "finish_rate": 0.98, "comp_len": 400.0, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 61.8, "frames": {"chat": 300}, "mem_gb": 21.79}
+{"step": 256, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.06250632524039053, "tokens": 120000, "cumulative_loss_tokens": 30720000, "grad_norm": 0.373046875, "lr": 3e-05, "finish_rate": 0.994, "comp_len": 344.8, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 67.0, "frames": {"chat": 348}, "mem_gb": 21.72}
+{"step": 257, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.08240252913138053, "tokens": 120000, "cumulative_loss_tokens": 30840000, "grad_norm": 0.408203125, "lr": 3e-05, "finish_rate": 0.988, "comp_len": 364.7, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 66.5, "frames": {"chat": 329}, "mem_gb": 21.81}
+{"step": 258, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.043369864273940524, "tokens": 120000, "cumulative_loss_tokens": 30960000, "grad_norm": 0.310546875, "lr": 3e-05, "finish_rate": 0.948, "comp_len": 446.1, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 58.2, "frames": {"chat": 269}, "mem_gb": 21.98}
+{"step": 259, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.08294476295701073, "tokens": 120000, "cumulative_loss_tokens": 31080000, "grad_norm": 0.42578125, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 335.2, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 69.8, "frames": {"chat": 358}, "mem_gb": 21.65}
+{"step": 260, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.05206059708436175, "tokens": 120000, "cumulative_loss_tokens": 31200000, "grad_norm": 0.345703125, "lr": 3e-05, "finish_rate": 0.96, "comp_len": 397.4, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 61.7, "frames": {"chat": 302}, "mem_gb": 21.9}
+{"step": 261, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.08964639542956526, "tokens": 120000, "cumulative_loss_tokens": 31320000, "grad_norm": 0.419921875, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 315.0, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 71.3, "frames": {"chat": 381}, "mem_gb": 21.92}
+{"step": 262, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.0973753253737154, "tokens": 120000, "cumulative_loss_tokens": 31440000, "grad_norm": 0.40625, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 316.6, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 76.1, "frames": {"chat": 379}, "mem_gb": 21.87}
+{"step": 263, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.09017735037549088, "tokens": 120000, "cumulative_loss_tokens": 31560000, "grad_norm": 0.44921875, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 341.9, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 70.6, "frames": {"chat": 351}, "mem_gb": 21.69}
+{"step": 264, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.07600556313122313, "tokens": 120000, "cumulative_loss_tokens": 31680000, "grad_norm": 0.369140625, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 363.6, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 62.1, "frames": {"chat": 330}, "mem_gb": 21.89}
+{"step": 265, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.07593888448358824, "tokens": 120000, "cumulative_loss_tokens": 31800000, "grad_norm": 0.365234375, "lr": 3e-05, "finish_rate": 0.991, "comp_len": 379.7, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 58.9, "frames": {"chat": 316}, "mem_gb": 21.92}
+{"step": 266, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.05233188844580048, "tokens": 120000, "cumulative_loss_tokens": 31920000, "grad_norm": 0.404296875, "lr": 3e-05, "finish_rate": 0.891, "comp_len": 434.8, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 57.3, "frames": {"chat": 276}, "mem_gb": 22.1}
+{"step": 267, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.045060560752583356, "tokens": 120000, "cumulative_loss_tokens": 32040000, "grad_norm": 0.30078125, "lr": 3e-05, "finish_rate": 0.833, "comp_len": 500.0, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 52.2, "frames": {"chat": 240}, "mem_gb": 22.04}
+{"step": 268, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.04613023779313856, "tokens": 120000, "cumulative_loss_tokens": 32160000, "grad_norm": 0.314453125, "lr": 3e-05, "finish_rate": 0.948, "comp_len": 416.7, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 55.5, "frames": {"chat": 288}, "mem_gb": 21.95}

healed/calib/glean_keep75_s1224/wandb/debug-internal.log

@@ -0,0 +1,599 @@
+{"time":"2026-07-17T23:24:40.756478928-07:00","level":"INFO","msg":"wandb-core"}
+{"time":"2026-07-17T23:24:40.757035796-07:00","level":"INFO","msg":"stream: starting","core version":"0.28.0"}
+{"time":"2026-07-17T23:24:40.989511132-07:00","level":"INFO","msg":"stream: created new stream","id":"ql8bd3ph"}
+{"time":"2026-07-17T23:24:40.989603625-07:00","level":"INFO","msg":"handler: started"}
+{"time":"2026-07-17T23:24:40.989724879-07:00","level":"INFO","msg":"stream: started"}
+{"time":"2026-07-17T23:24:40.989793651-07:00","level":"INFO","msg":"sender: started"}
+{"time":"2026-07-17T23:24:40.989788771-07:00","level":"INFO","msg":"writer: started","stream_id":"ql8bd3ph"}
+{"time":"2026-07-17T23:24:42.474211675-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"console_offset":0,"console_lines":1}
+{"time":"2026-07-17T23:24:42.675210395-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:24:57.474484195-07:00","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":0,"events_lines":2,"console_offset":0,"console_lines":3,"uploaded_len":2}
+{"time":"2026-07-17T23:24:57.730340611-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:25:12.474471825-07:00","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":2,"events_lines":2,"console_offset":3,"console_lines":1}
+{"time":"2026-07-17T23:25:12.742491872-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:25:27.474997141-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":4,"events_lines":2}
+{"time":"2026-07-17T23:25:27.623704772-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:25:42.475106488-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":6,"events_lines":2}
+{"time":"2026-07-17T23:25:42.80885941-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:25:57.47446318-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":8,"events_lines":2}
+{"time":"2026-07-17T23:25:57.616024971-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:26:12.474441213-07:00","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":0,"history_lines":1,"events_offset":10,"events_lines":2,"console_offset":4,"console_lines":1}
+{"time":"2026-07-17T23:26:12.637305449-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:26:27.474731297-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":12,"events_lines":2}
+{"time":"2026-07-17T23:26:27.674412538-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:26:42.474257344-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":14,"events_lines":2}
+{"time":"2026-07-17T23:26:42.682627788-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:26:57.475054015-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":16,"events_lines":2}
+{"time":"2026-07-17T23:26:57.722773789-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:27:12.474158209-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":18,"events_lines":2}
+{"time":"2026-07-17T23:27:12.692026211-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:27:27.474274066-07:00","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":1,"history_lines":1,"events_offset":20,"events_lines":2,"console_offset":5,"console_lines":1}
+{"time":"2026-07-17T23:27:27.793134058-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:27:42.474496907-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":22,"events_lines":2}
+{"time":"2026-07-17T23:27:42.663396637-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:27:57.475096531-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":24,"events_lines":2}
+{"time":"2026-07-17T23:27:57.68065895-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:28:12.474337779-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":26,"events_lines":2}
+{"time":"2026-07-17T23:28:12.694809219-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:28:27.475187332-07:00","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":2,"history_lines":1,"events_offset":28,"events_lines":2,"console_offset":6,"console_lines":1}
+{"time":"2026-07-17T23:28:27.693990025-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:28:42.474304926-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":30,"events_lines":2}
+{"time":"2026-07-17T23:28:42.653145508-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:28:57.474761005-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":32,"events_lines":2}
+{"time":"2026-07-17T23:28:57.635386166-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:29:12.47511698-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":34,"events_lines":2}
+{"time":"2026-07-17T23:29:12.655526675-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:29:27.475170686-07:00","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":3,"history_lines":1,"events_offset":36,"events_lines":2,"console_offset":7,"console_lines":1}
+{"time":"2026-07-17T23:29:27.623773313-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:29:42.475066386-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":38,"events_lines":2}
+{"time":"2026-07-17T23:29:42.792937675-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:29:57.475106041-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":40,"events_lines":2}
+{"time":"2026-07-17T23:29:57.690224231-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:30:12.474499875-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":42,"events_lines":2}
+{"time":"2026-07-17T23:30:12.630281063-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:30:27.47425411-07:00","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":4,"history_lines":1,"events_offset":44,"events_lines":2,"console_offset":8,"console_lines":1}
+{"time":"2026-07-17T23:30:27.638461521-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:30:42.474716779-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":46,"events_lines":2}
+{"time":"2026-07-17T23:30:42.686856989-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:30:57.474679342-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":48,"events_lines":2}
+{"time":"2026-07-17T23:30:57.624894553-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:31:12.474843391-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":50,"events_lines":2}
+{"time":"2026-07-17T23:31:12.649042137-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:31:27.474534164-07:00","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":5,"history_lines":1,"events_offset":52,"events_lines":2,"console_offset":9,"console_lines":1}
+{"time":"2026-07-17T23:31:27.687290985-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:31:42.475028215-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":54,"events_lines":2}
+{"time":"2026-07-17T23:31:42.691605203-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:31:57.474578613-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":56,"events_lines":2}
+{"time":"2026-07-17T23:31:57.719773573-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:32:12.474762843-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":58,"events_lines":2}
+{"time":"2026-07-17T23:32:12.621934593-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:32:27.474290981-07:00","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":6,"history_lines":1,"events_offset":60,"events_lines":2,"console_offset":10,"console_lines":1}
+{"time":"2026-07-17T23:32:27.71409825-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:32:42.474474911-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":62,"events_lines":2}
+{"time":"2026-07-17T23:32:42.624207236-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:32:57.474594789-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":64,"events_lines":2}
+{"time":"2026-07-17T23:32:57.781507049-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:33:12.474440617-07:00","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":7,"history_lines":1,"events_offset":66,"events_lines":2,"console_offset":11,"console_lines":1}
+{"time":"2026-07-17T23:33:12.649639467-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:33:27.475170965-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":68,"events_lines":2}
+{"time":"2026-07-17T23:33:27.817841986-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:33:42.475127778-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":70,"events_lines":2}
+{"time":"2026-07-17T23:33:42.746172144-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:33:57.474529164-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":72,"events_lines":2}
+{"time":"2026-07-17T23:33:57.693271986-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:34:12.474335651-07:00","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":8,"history_lines":1,"events_offset":74,"events_lines":2,"console_offset":12,"console_lines":1}
+{"time":"2026-07-17T23:34:12.72546577-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:34:27.474969175-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":76,"events_lines":2}
+{"time":"2026-07-17T23:34:27.630690522-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:34:42.474673918-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":78,"events_lines":2}
+{"time":"2026-07-17T23:34:42.684739472-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:34:57.474495076-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":80,"events_lines":2}
+{"time":"2026-07-17T23:34:57.727058802-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:35:12.474569861-07:00","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":9,"history_lines":1,"events_offset":82,"events_lines":2,"console_offset":13,"console_lines":2}
+{"time":"2026-07-17T23:35:12.727156129-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:35:27.474767631-07:00","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":84,"events_lines":2,"console_offset":15,"console_lines":1}
+{"time":"2026-07-17T23:35:27.728545118-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:35:42.475178978-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":86,"events_lines":2}
+{"time":"2026-07-17T23:35:42.722691467-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:35:57.474776158-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":88,"events_lines":2}
+{"time":"2026-07-17T23:35:57.679909158-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:36:12.474336197-07:00","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":10,"history_lines":1,"events_offset":90,"events_lines":2,"console_offset":16,"console_lines":1}
+{"time":"2026-07-17T23:36:12.823080071-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:36:27.474607459-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":92,"events_lines":2}
+{"time":"2026-07-17T23:36:27.717142532-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:36:42.474603822-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":94,"events_lines":2}
+{"time":"2026-07-17T23:36:42.609372482-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:36:57.474518432-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":96,"events_lines":2}
+{"time":"2026-07-17T23:36:57.651760455-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:37:12.474164293-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":98,"events_lines":2}
+{"time":"2026-07-17T23:37:12.749822968-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:37:27.475056636-07:00","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":11,"history_lines":1,"events_offset":100,"events_lines":2,"console_offset":17,"console_lines":1}
+{"time":"2026-07-17T23:37:27.719963138-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:37:42.474313195-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":102,"events_lines":2}
+{"time":"2026-07-17T23:37:42.655153057-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:37:57.474541676-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":104,"events_lines":2}
+{"time":"2026-07-17T23:37:57.621344296-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:38:12.474336792-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":106,"events_lines":2}
+{"time":"2026-07-17T23:38:12.634515477-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:38:27.474436258-07:00","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":12,"history_lines":1,"events_offset":108,"events_lines":2,"console_offset":18,"console_lines":1}
+{"time":"2026-07-17T23:38:27.824712064-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:38:42.474965249-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":110,"events_lines":2}
+{"time":"2026-07-17T23:38:42.625892877-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:38:57.475135438-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":112,"events_lines":2}
+{"time":"2026-07-17T23:38:57.633096299-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:39:12.475200233-07:00","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":13,"history_lines":1,"events_offset":114,"events_lines":2,"console_offset":19,"console_lines":1}
+{"time":"2026-07-17T23:39:12.633215126-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:39:27.474519564-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":116,"events_lines":2}
+{"time":"2026-07-17T23:39:27.813347088-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:39:42.474127304-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":118,"events_lines":2}
+{"time":"2026-07-17T23:39:42.647633112-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:39:57.474559351-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":120,"events_lines":2}
+{"time":"2026-07-17T23:39:57.637820909-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:40:12.474281375-07:00","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":14,"history_lines":1,"events_offset":122,"events_lines":2,"console_offset":20,"console_lines":1}
+{"time":"2026-07-17T23:40:12.689046115-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:40:27.474706033-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":124,"events_lines":2}
+{"time":"2026-07-17T23:40:27.689198714-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:40:42.474166898-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":126,"events_lines":2}
+{"time":"2026-07-17T23:40:42.593496525-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:40:57.474687588-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":128,"events_lines":2}
+{"time":"2026-07-17T23:40:57.674611745-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:41:12.474596149-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":130,"events_lines":2}
+{"time":"2026-07-17T23:41:12.620880472-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:41:27.474343984-07:00","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":15,"history_lines":1,"events_offset":132,"events_lines":2,"console_offset":21,"console_lines":1}
+{"time":"2026-07-17T23:41:27.745002422-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:41:42.474339117-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":134,"events_lines":2}
+{"time":"2026-07-17T23:41:42.661186649-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:41:57.475109256-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":136,"events_lines":2}
+{"time":"2026-07-17T23:41:57.718253859-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:42:12.47515003-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":138,"events_lines":2}
+{"time":"2026-07-17T23:42:12.659557361-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:42:27.474423609-07:00","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":16,"history_lines":1,"events_offset":140,"events_lines":2,"console_offset":22,"console_lines":1}
+{"time":"2026-07-17T23:42:27.676826818-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:42:42.474639541-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":142,"events_lines":2}
+{"time":"2026-07-17T23:42:42.737945368-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:42:57.474934539-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":144,"events_lines":2}
+{"time":"2026-07-17T23:42:57.777113497-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:43:12.474237905-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":146,"events_lines":2}
+{"time":"2026-07-17T23:43:12.668327465-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:43:27.474324406-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":148,"events_lines":2}
+{"time":"2026-07-17T23:43:27.715608933-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:43:42.474557702-07:00","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":17,"history_lines":1,"events_offset":150,"events_lines":2,"console_offset":23,"console_lines":1}
+{"time":"2026-07-17T23:43:42.764686739-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:43:57.474206259-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":152,"events_lines":2}
+{"time":"2026-07-17T23:43:57.742910925-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:44:12.474738275-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":154,"events_lines":2}
+{"time":"2026-07-17T23:44:12.703091813-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:44:27.47493708-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":156,"events_lines":2}
+{"time":"2026-07-17T23:44:27.744030029-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:44:42.474509454-07:00","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":18,"history_lines":1,"events_offset":158,"events_lines":2,"console_offset":24,"console_lines":1}
+{"time":"2026-07-17T23:44:42.788514084-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:44:57.47413948-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":160,"events_lines":2}
+{"time":"2026-07-17T23:44:57.751711781-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:45:12.475095621-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":162,"events_lines":2}
+{"time":"2026-07-17T23:45:12.703828101-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:45:27.474326544-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":164,"events_lines":2}
+{"time":"2026-07-17T23:45:27.697943354-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:45:42.474325962-07:00","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":19,"history_lines":1,"events_offset":166,"events_lines":2,"console_offset":25,"console_lines":2}
+{"time":"2026-07-17T23:45:42.61711388-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:45:57.474574749-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":168,"events_lines":2}
+{"time":"2026-07-17T23:45:57.680417929-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:46:12.47435375-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":170,"events_lines":2}
+{"time":"2026-07-17T23:46:12.732562758-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:46:27.475076732-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":172,"events_lines":2}
+{"time":"2026-07-17T23:46:27.623767926-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:46:42.474792321-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":174,"events_lines":2}
+{"time":"2026-07-17T23:46:42.761007169-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:46:57.475063779-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":176,"events_lines":2}
+{"time":"2026-07-17T23:46:57.670133812-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:47:12.474758887-07:00","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":20,"history_lines":1,"events_offset":178,"events_lines":2,"console_offset":27,"console_lines":1}
+{"time":"2026-07-17T23:47:12.689291386-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:47:27.474341311-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":180,"events_lines":2}
+{"time":"2026-07-17T23:47:27.73950374-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:47:42.474448463-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":182,"events_lines":2}
+{"time":"2026-07-17T23:47:42.714691575-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:47:57.474872586-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":184,"events_lines":2}
+{"time":"2026-07-17T23:47:57.628954369-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:48:12.47506226-07:00","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":21,"history_lines":1,"events_offset":186,"events_lines":2,"console_offset":28,"console_lines":1}
+{"time":"2026-07-17T23:48:12.617032611-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:48:27.474752348-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":188,"events_lines":2}
+{"time":"2026-07-17T23:48:27.722313773-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:48:42.474941633-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":190,"events_lines":2}
+{"time":"2026-07-17T23:48:42.756480555-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:48:57.475116727-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":192,"events_lines":2}
+{"time":"2026-07-17T23:48:57.719552118-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:49:12.474372691-07:00","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":22,"history_lines":1,"events_offset":194,"events_lines":2,"console_offset":29,"console_lines":1}
+{"time":"2026-07-17T23:49:12.716758584-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:49:27.474849805-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":196,"events_lines":2}
+{"time":"2026-07-17T23:49:27.64496999-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:49:42.474578894-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":198,"events_lines":2}
+{"time":"2026-07-17T23:49:42.753166599-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:49:57.474815191-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":200,"events_lines":2}
+{"time":"2026-07-17T23:49:57.660331737-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:50:12.474530239-07:00","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":23,"history_lines":1,"events_offset":202,"events_lines":2,"console_offset":30,"console_lines":1}
+{"time":"2026-07-17T23:50:12.770543362-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:50:27.474211137-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":204,"events_lines":2}
+{"time":"2026-07-17T23:50:27.653812427-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:50:42.474815306-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":206,"events_lines":2}
+{"time":"2026-07-17T23:50:42.762975558-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:50:57.474352039-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":208,"events_lines":2}
+{"time":"2026-07-17T23:50:57.626124935-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:51:12.474518667-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":210,"events_lines":2}
+{"time":"2026-07-17T23:51:12.673289099-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:51:27.474275099-07:00","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":24,"history_lines":1,"events_offset":212,"events_lines":2,"console_offset":31,"console_lines":1}
+{"time":"2026-07-17T23:51:27.66150294-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:51:42.474399064-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":214,"events_lines":2}
+{"time":"2026-07-17T23:51:42.703675695-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:51:57.474307631-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":216,"events_lines":2}
+{"time":"2026-07-17T23:51:57.654856611-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:52:12.47456297-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":218,"events_lines":2}
+{"time":"2026-07-17T23:52:12.789027117-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:52:27.474785128-07:00","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":25,"history_lines":1,"events_offset":220,"events_lines":2,"console_offset":32,"console_lines":1}
+{"time":"2026-07-17T23:52:27.731214606-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:52:42.474467897-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":222,"events_lines":2}
+{"time":"2026-07-17T23:52:42.619384549-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:52:57.475057947-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":224,"events_lines":2}
+{"time":"2026-07-17T23:52:57.705588179-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:53:12.475120359-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":226,"events_lines":2}
+{"time":"2026-07-17T23:53:12.676736616-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:53:27.475203453-07:00","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":26,"history_lines":1,"events_offset":228,"events_lines":2,"console_offset":33,"console_lines":1}
+{"time":"2026-07-17T23:53:27.716968836-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:53:42.474844901-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":230,"events_lines":2}
+{"time":"2026-07-17T23:53:42.661053458-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:53:57.475008607-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":232,"events_lines":2}
+{"time":"2026-07-17T23:53:57.786268758-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:54:12.474974516-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":234,"events_lines":2}
+{"time":"2026-07-17T23:54:12.726516643-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:54:27.475157692-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":236,"events_lines":2}
+{"time":"2026-07-17T23:54:27.702793029-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:54:42.475257166-07:00","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":27,"history_lines":1,"events_offset":238,"events_lines":2,"console_offset":34,"console_lines":1}
+{"time":"2026-07-17T23:54:42.698913471-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:54:57.474115168-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":240,"events_lines":2}
+{"time":"2026-07-17T23:54:57.829084965-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:55:12.474194121-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":242,"events_lines":2}
+{"time":"2026-07-17T23:55:12.670279405-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:55:27.474494871-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":244,"events_lines":2}
+{"time":"2026-07-17T23:55:27.723503545-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:55:42.474957857-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":246,"events_lines":2}
+{"time":"2026-07-17T23:55:42.844493835-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:55:57.474963468-07:00","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":28,"history_lines":1,"events_offset":248,"events_lines":2,"console_offset":35,"console_lines":1}
+{"time":"2026-07-17T23:55:57.876701551-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:56:12.475133874-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":250,"events_lines":2}
+{"time":"2026-07-17T23:56:12.741976786-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:56:27.474697839-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":252,"events_lines":2}
+{"time":"2026-07-17T23:56:27.736213326-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:56:42.474306757-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":254,"events_lines":2}
+{"time":"2026-07-17T23:56:42.760294803-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:56:57.475278619-07:00","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":29,"history_lines":1,"events_offset":256,"events_lines":2,"console_offset":36,"console_lines":2}
+{"time":"2026-07-17T23:56:57.873555148-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:57:12.474621998-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":258,"events_lines":2}
+{"time":"2026-07-17T23:57:12.655783798-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:57:27.47469616-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":260,"events_lines":2}
+{"time":"2026-07-17T23:57:27.767943666-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:57:42.474479564-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":262,"events_lines":2}
+{"time":"2026-07-17T23:57:42.780068008-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:57:57.474273247-07:00","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":30,"history_lines":1,"events_offset":264,"events_lines":2,"console_offset":38,"console_lines":1}
+{"time":"2026-07-17T23:57:57.654260568-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:58:12.474927569-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":266,"events_lines":2}
+{"time":"2026-07-17T23:58:12.706476245-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:58:27.474609659-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":268,"events_lines":2}
+{"time":"2026-07-17T23:58:27.788745145-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:58:42.47494199-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":270,"events_lines":2}
+{"time":"2026-07-17T23:58:42.738876567-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:58:57.474378362-07:00","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":31,"history_lines":1,"events_offset":272,"events_lines":2,"console_offset":39,"console_lines":1}
+{"time":"2026-07-17T23:58:57.616023915-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:59:12.47431415-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":274,"events_lines":2}
+{"time":"2026-07-17T23:59:12.665180181-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:59:27.474766415-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":276,"events_lines":2}
+{"time":"2026-07-17T23:59:27.749458698-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:59:42.474274269-07:00","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":32,"history_lines":1,"events_offset":278,"events_lines":2,"console_offset":40,"console_lines":1}
+{"time":"2026-07-17T23:59:42.653543047-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-17T23:59:57.475099175-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":280,"events_lines":2}
+{"time":"2026-07-17T23:59:57.602756403-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:00:12.47467708-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":282,"events_lines":2}
+{"time":"2026-07-18T00:00:12.724880828-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:00:27.474593567-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":284,"events_lines":2}
+{"time":"2026-07-18T00:00:27.692160249-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:00:42.474576616-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":286,"events_lines":2}
+{"time":"2026-07-18T00:00:42.659295555-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:00:57.474756962-07:00","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":33,"history_lines":1,"events_offset":288,"events_lines":2,"console_offset":41,"console_lines":1}
+{"time":"2026-07-18T00:00:57.716448299-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:01:12.474950988-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":290,"events_lines":2}
+{"time":"2026-07-18T00:01:12.711694502-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:01:27.474988969-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":292,"events_lines":2}
+{"time":"2026-07-18T00:01:27.689859661-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:01:42.474664768-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":294,"events_lines":2}
+{"time":"2026-07-18T00:01:42.790079214-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:01:57.474997408-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":296,"events_lines":2}
+{"time":"2026-07-18T00:01:57.732121673-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:02:12.474307655-07:00","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":34,"history_lines":1,"events_offset":298,"events_lines":2,"console_offset":42,"console_lines":1}
+{"time":"2026-07-18T00:02:12.698461133-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:02:27.474461819-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":300,"events_lines":2}
+{"time":"2026-07-18T00:02:27.767640343-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:02:42.474333935-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":302,"events_lines":2}
+{"time":"2026-07-18T00:02:42.732780764-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:02:57.474361075-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":304,"events_lines":2}
+{"time":"2026-07-18T00:02:57.769949978-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:03:12.474864051-07:00","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":35,"history_lines":1,"events_offset":306,"events_lines":2,"console_offset":43,"console_lines":1}
+{"time":"2026-07-18T00:03:12.682107592-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:03:27.474598601-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":308,"events_lines":2}
+{"time":"2026-07-18T00:03:27.748317324-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:03:42.47518401-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":310,"events_lines":2}
+{"time":"2026-07-18T00:03:42.827582986-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:03:57.47458443-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":312,"events_lines":2}
+{"time":"2026-07-18T00:03:57.674701096-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:04:12.474819187-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":314,"events_lines":2}
+{"time":"2026-07-18T00:04:12.79496622-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:04:27.47460277-07:00","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":36,"history_lines":1,"events_offset":316,"events_lines":2,"console_offset":44,"console_lines":1}
+{"time":"2026-07-18T00:04:27.803147009-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:04:42.47491797-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":318,"events_lines":2}
+{"time":"2026-07-18T00:04:42.694301501-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:04:57.475001262-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":320,"events_lines":2}
+{"time":"2026-07-18T00:04:57.798402212-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:05:12.475169217-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":322,"events_lines":2}
+{"time":"2026-07-18T00:05:12.774669199-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:05:27.474422712-07:00","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":37,"history_lines":1,"events_offset":324,"events_lines":2,"console_offset":45,"console_lines":1}
+{"time":"2026-07-18T00:05:27.766991785-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:05:42.47438598-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":326,"events_lines":2}
+{"time":"2026-07-18T00:05:42.799044341-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:05:57.475063632-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":328,"events_lines":2}
+{"time":"2026-07-18T00:05:57.701206076-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:06:12.474935117-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":330,"events_lines":2}
+{"time":"2026-07-18T00:06:12.819430762-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:06:27.474493032-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":332,"events_lines":2}
+{"time":"2026-07-18T00:06:27.689548081-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:06:42.475223646-07:00","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":38,"history_lines":1,"events_offset":334,"events_lines":2,"console_offset":46,"console_lines":1}
+{"time":"2026-07-18T00:06:42.795621417-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:06:57.474789901-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":336,"events_lines":2}
+{"time":"2026-07-18T00:06:57.792913497-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:07:12.474198721-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":338,"events_lines":2}
+{"time":"2026-07-18T00:07:12.715151183-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:07:27.474380497-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":340,"events_lines":2}
+{"time":"2026-07-18T00:07:27.680323955-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:07:42.474604004-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":342,"events_lines":2}
+{"time":"2026-07-18T00:07:42.764639604-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:07:57.4745158-07:00","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":39,"history_lines":1,"events_offset":344,"events_lines":2,"console_offset":47,"console_lines":2}
+{"time":"2026-07-18T00:07:57.723737915-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:08:12.475035557-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":346,"events_lines":2}
+{"time":"2026-07-18T00:08:12.724769779-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:08:27.474218969-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":348,"events_lines":2}
+{"time":"2026-07-18T00:08:27.74790723-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:08:42.474363051-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":350,"events_lines":2}
+{"time":"2026-07-18T00:08:42.792219198-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:08:57.475034-07:00","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":40,"history_lines":1,"events_offset":352,"events_lines":2,"console_offset":49,"console_lines":1}
+{"time":"2026-07-18T00:08:57.642427518-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:09:12.474181989-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":354,"events_lines":2}
+{"time":"2026-07-18T00:09:12.717630274-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:09:27.474912581-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":356,"events_lines":2}
+{"time":"2026-07-18T00:09:27.762679836-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:09:42.474139512-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":358,"events_lines":2}
+{"time":"2026-07-18T00:09:42.711953701-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:09:57.474346206-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":360,"events_lines":2}
+{"time":"2026-07-18T00:09:57.688149144-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:10:12.474317203-07:00","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":41,"history_lines":1,"events_offset":362,"events_lines":2,"console_offset":50,"console_lines":1}
+{"time":"2026-07-18T00:10:12.711358446-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:10:27.474711163-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":364,"events_lines":2}
+{"time":"2026-07-18T00:10:27.69548324-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:10:42.474842454-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":366,"events_lines":2}
+{"time":"2026-07-18T00:10:42.688667442-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:10:57.475012737-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":368,"events_lines":2}
+{"time":"2026-07-18T00:10:57.673884042-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:11:12.474354853-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":370,"events_lines":2}
+{"time":"2026-07-18T00:11:12.745077316-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:11:27.474825145-07:00","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":42,"history_lines":1,"events_offset":372,"events_lines":2,"console_offset":51,"console_lines":1}
+{"time":"2026-07-18T00:11:27.760210472-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:11:42.474348186-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":374,"events_lines":2}
+{"time":"2026-07-18T00:11:42.704321736-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:11:57.474177928-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":376,"events_lines":2}
+{"time":"2026-07-18T00:11:57.758596693-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:12:12.475049384-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":378,"events_lines":2}
+{"time":"2026-07-18T00:12:12.789709455-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:12:27.47502411-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":380,"events_lines":2}
+{"time":"2026-07-18T00:12:27.712946102-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:12:42.474935334-07:00","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":43,"history_lines":1,"events_offset":382,"events_lines":2,"console_offset":52,"console_lines":1}
+{"time":"2026-07-18T00:12:42.727132747-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:12:57.474547838-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":384,"events_lines":2}
+{"time":"2026-07-18T00:12:57.721310042-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:13:12.474310138-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":386,"events_lines":2}
+{"time":"2026-07-18T00:13:12.77350326-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:13:27.475148213-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":388,"events_lines":2}
+{"time":"2026-07-18T00:13:27.728625728-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:13:42.474462088-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":390,"events_lines":2}
+{"time":"2026-07-18T00:13:42.719927239-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:13:57.475145848-07:00","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":44,"history_lines":1,"events_offset":392,"events_lines":2,"console_offset":53,"console_lines":1}
+{"time":"2026-07-18T00:13:57.754115823-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:14:12.475137015-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":394,"events_lines":2}
+{"time":"2026-07-18T00:14:12.715324022-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:14:27.475138622-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":396,"events_lines":2}
+{"time":"2026-07-18T00:14:27.703355854-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:14:42.474515078-07:00","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":45,"history_lines":1,"events_offset":398,"events_lines":2,"console_offset":54,"console_lines":1}
+{"time":"2026-07-18T00:14:42.693884089-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:14:57.474204845-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":400,"events_lines":2}
+{"time":"2026-07-18T00:14:57.681811238-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:15:12.475102752-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":402,"events_lines":2}
+{"time":"2026-07-18T00:15:12.877004109-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:15:27.474455498-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":404,"events_lines":2}
+{"time":"2026-07-18T00:15:27.736199325-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:15:42.47520358-07:00","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":46,"history_lines":1,"events_offset":406,"events_lines":2,"console_offset":55,"console_lines":1}
+{"time":"2026-07-18T00:15:42.785347973-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:15:57.474478453-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":408,"events_lines":2}
+{"time":"2026-07-18T00:15:57.70556475-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:16:12.474804311-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":410,"events_lines":2}
+{"time":"2026-07-18T00:16:12.718753182-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:16:27.474356233-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":412,"events_lines":2}
+{"time":"2026-07-18T00:16:27.63900909-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:16:42.4743567-07:00","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":47,"history_lines":1,"events_offset":414,"events_lines":2,"console_offset":56,"console_lines":1}
+{"time":"2026-07-18T00:16:42.687141434-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:16:57.47413603-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":416,"events_lines":2}
+{"time":"2026-07-18T00:16:57.687286306-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:17:12.474447958-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":418,"events_lines":2}
+{"time":"2026-07-18T00:17:12.670480279-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:17:27.474273819-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":420,"events_lines":2}
+{"time":"2026-07-18T00:17:27.72671012-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:17:42.474429001-07:00","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":48,"history_lines":1,"events_offset":422,"events_lines":2,"console_offset":57,"console_lines":1}
+{"time":"2026-07-18T00:17:42.618928064-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:17:57.474342085-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":424,"events_lines":2}
+{"time":"2026-07-18T00:17:57.714093938-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:18:12.474594091-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":426,"events_lines":2}
+{"time":"2026-07-18T00:18:12.685239844-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:18:27.474760843-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":428,"events_lines":2}
+{"time":"2026-07-18T00:18:27.671419025-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:18:42.475132512-07:00","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":49,"history_lines":1,"events_offset":430,"events_lines":2,"console_offset":58,"console_lines":2}
+{"time":"2026-07-18T00:18:42.797643093-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:18:57.47515124-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":432,"events_lines":2}
+{"time":"2026-07-18T00:18:57.652844537-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:19:12.475101965-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":434,"events_lines":2}
+{"time":"2026-07-18T00:19:12.721032112-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:19:27.474603066-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":436,"events_lines":2}
+{"time":"2026-07-18T00:19:27.642211161-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:19:42.474160889-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":438,"events_lines":2}
+{"time":"2026-07-18T00:19:42.748336426-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:19:57.475124477-07:00","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":50,"history_lines":1,"events_offset":440,"events_lines":2,"console_offset":60,"console_lines":1}
+{"time":"2026-07-18T00:19:57.631493721-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:20:12.475197446-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":442,"events_lines":2}
+{"time":"2026-07-18T00:20:12.701754574-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:20:27.475138612-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":444,"events_lines":2}
+{"time":"2026-07-18T00:20:27.677924316-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:20:42.474321882-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":446,"events_lines":2}
+{"time":"2026-07-18T00:20:42.741160477-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:20:57.474573457-07:00","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":51,"history_lines":1,"events_offset":448,"events_lines":2,"console_offset":61,"console_lines":1}
+{"time":"2026-07-18T00:20:57.730324137-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:21:12.474659497-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":450,"events_lines":2}
+{"time":"2026-07-18T00:21:12.686464698-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:21:27.474917222-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":452,"events_lines":2}
+{"time":"2026-07-18T00:21:27.711753758-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:21:42.475045473-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":454,"events_lines":2}
+{"time":"2026-07-18T00:21:42.745951553-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:21:57.474899245-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":456,"events_lines":2}
+{"time":"2026-07-18T00:21:57.62916944-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:22:12.475263254-07:00","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":52,"history_lines":1,"events_offset":458,"events_lines":2,"console_offset":62,"console_lines":1}
+{"time":"2026-07-18T00:22:12.715164352-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:22:27.474351241-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":460,"events_lines":2}
+{"time":"2026-07-18T00:22:27.738520349-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:22:42.475043431-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":462,"events_lines":2}
+{"time":"2026-07-18T00:22:42.643591797-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:22:57.475043138-07:00","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":53,"history_lines":1,"events_offset":464,"events_lines":2,"console_offset":63,"console_lines":1}
+{"time":"2026-07-18T00:22:57.73388995-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:23:12.474993163-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":466,"events_lines":2}
+{"time":"2026-07-18T00:23:12.700928391-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:23:27.474619648-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":468,"events_lines":2}
+{"time":"2026-07-18T00:23:27.697198975-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:23:42.474778971-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":470,"events_lines":2}
+{"time":"2026-07-18T00:23:42.680397698-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:23:57.475033136-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":472,"events_lines":2}
+{"time":"2026-07-18T00:23:57.706551947-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:24:12.474795345-07:00","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":54,"history_lines":1,"events_offset":474,"events_lines":2,"console_offset":64,"console_lines":1}
+{"time":"2026-07-18T00:24:12.698818909-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:24:27.474781862-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":476,"events_lines":2}
+{"time":"2026-07-18T00:24:27.67194731-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:24:42.474786409-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":478,"events_lines":2}
+{"time":"2026-07-18T00:24:42.748383147-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:24:57.474369552-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":480,"events_lines":2}
+{"time":"2026-07-18T00:24:57.756406038-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:25:12.474673519-07:00","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":55,"history_lines":1,"events_offset":482,"events_lines":2,"console_offset":65,"console_lines":1}
+{"time":"2026-07-18T00:25:12.662464649-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:25:27.474631215-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":484,"events_lines":2}
+{"time":"2026-07-18T00:25:27.673705187-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:25:42.47456519-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":486,"events_lines":2}
+{"time":"2026-07-18T00:25:42.67785268-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:25:57.474645463-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":488,"events_lines":2}
+{"time":"2026-07-18T00:25:57.747034698-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:26:12.474316284-07:00","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":56,"history_lines":1,"events_offset":490,"events_lines":2,"console_offset":66,"console_lines":1}
+{"time":"2026-07-18T00:26:12.781204091-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:26:27.474298765-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":492,"events_lines":2}
+{"time":"2026-07-18T00:26:27.734381756-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:26:42.474259556-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":494,"events_lines":2}
+{"time":"2026-07-18T00:26:42.683696993-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:26:57.474804375-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":496,"events_lines":2}
+{"time":"2026-07-18T00:26:57.802899369-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:27:12.475157998-07:00","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":57,"history_lines":1,"events_offset":498,"events_lines":2,"console_offset":67,"console_lines":1}
+{"time":"2026-07-18T00:27:12.716023768-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:27:27.474452747-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":500,"events_lines":2}
+{"time":"2026-07-18T00:27:27.77025763-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:27:42.475140471-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":502,"events_lines":2}
+{"time":"2026-07-18T00:27:42.684435653-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:27:57.474220922-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":504,"events_lines":2}
+{"time":"2026-07-18T00:27:57.801604873-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:28:12.474815703-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":506,"events_lines":2}
+{"time":"2026-07-18T00:28:12.780839232-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:28:27.474566156-07:00","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":58,"history_lines":1,"events_offset":508,"events_lines":2,"console_offset":68,"console_lines":1}
+{"time":"2026-07-18T00:28:27.70491438-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:28:42.474197076-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":510,"events_lines":2}
+{"time":"2026-07-18T00:28:42.700168226-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:28:57.474157586-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":512,"events_lines":2}
+{"time":"2026-07-18T00:28:57.745283159-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:29:12.474339144-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":514,"events_lines":2}
+{"time":"2026-07-18T00:29:12.699567289-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:29:27.475019377-07:00","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":59,"history_lines":1,"events_offset":516,"events_lines":2,"console_offset":69,"console_lines":2}
+{"time":"2026-07-18T00:29:27.79669791-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:29:42.474974038-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":518,"events_lines":2}
+{"time":"2026-07-18T00:29:42.752854612-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:29:57.47438466-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":520,"events_lines":2}
+{"time":"2026-07-18T00:29:57.617989775-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:30:12.474180435-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":522,"events_lines":2}
+{"time":"2026-07-18T00:30:12.703198825-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:30:27.47429996-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":524,"events_lines":2}
+{"time":"2026-07-18T00:30:27.671394182-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:30:42.474930833-07:00","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":60,"history_lines":1,"events_offset":526,"events_lines":2,"console_offset":71,"console_lines":1}
+{"time":"2026-07-18T00:30:42.728663744-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:30:57.474976566-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":528,"events_lines":2}
+{"time":"2026-07-18T00:30:57.70471727-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:31:12.475088861-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":530,"events_lines":2}
+{"time":"2026-07-18T00:31:12.785955888-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:31:27.474531894-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":532,"events_lines":2}
+{"time":"2026-07-18T00:31:27.704164524-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:31:42.475111014-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":534,"events_lines":2}
+{"time":"2026-07-18T00:31:42.777352579-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:31:57.474737064-07:00","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":61,"history_lines":1,"events_offset":536,"events_lines":2,"console_offset":72,"console_lines":1}
+{"time":"2026-07-18T00:31:57.69450054-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:32:12.474829199-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":538,"events_lines":2}
+{"time":"2026-07-18T00:32:12.736674047-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:32:27.475181852-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":540,"events_lines":2}
+{"time":"2026-07-18T00:32:27.724883121-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:32:42.474579904-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":542,"events_lines":2}
+{"time":"2026-07-18T00:32:42.73606113-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:32:57.474493262-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":544,"events_lines":2}
+{"time":"2026-07-18T00:32:57.726316881-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:33:12.474422041-07:00","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":62,"history_lines":1,"events_offset":546,"events_lines":2,"console_offset":73,"console_lines":1}
+{"time":"2026-07-18T00:33:12.731437501-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:33:27.475003472-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":548,"events_lines":2}
+{"time":"2026-07-18T00:33:27.76761835-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:33:42.4749004-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":550,"events_lines":2}
+{"time":"2026-07-18T00:33:42.744812083-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:33:57.4745581-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":552,"events_lines":2}
+{"time":"2026-07-18T00:33:57.779973859-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:34:12.475260454-07:00","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":63,"history_lines":1,"events_offset":554,"events_lines":2,"console_offset":74,"console_lines":1}
+{"time":"2026-07-18T00:34:12.703136807-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:34:27.474821021-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":556,"events_lines":2}
+{"time":"2026-07-18T00:34:27.681398685-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:34:42.474263015-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":558,"events_lines":2}
+{"time":"2026-07-18T00:34:42.752582484-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:34:57.474140052-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":560,"events_lines":2}
+{"time":"2026-07-18T00:34:57.765620523-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:35:12.474803535-07:00","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":64,"history_lines":1,"events_offset":562,"events_lines":2,"console_offset":75,"console_lines":1}
+{"time":"2026-07-18T00:35:12.744796831-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:35:27.474663952-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":564,"events_lines":2}
+{"time":"2026-07-18T00:35:27.775085167-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:35:42.474528009-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":566,"events_lines":2}
+{"time":"2026-07-18T00:35:42.708155021-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:35:57.474166589-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":568,"events_lines":2}
+{"time":"2026-07-18T00:35:57.738406146-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:36:12.475028209-07:00","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":65,"history_lines":1,"events_offset":570,"events_lines":2,"console_offset":76,"console_lines":1}
+{"time":"2026-07-18T00:36:12.798671156-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:36:27.474216754-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":572,"events_lines":2}
+{"time":"2026-07-18T00:36:27.722814647-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:36:42.474159513-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":574,"events_lines":2}
+{"time":"2026-07-18T00:36:42.757034482-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:36:57.474282719-07:00","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":66,"history_lines":1,"events_offset":576,"events_lines":2,"console_offset":77,"console_lines":1}
+{"time":"2026-07-18T00:36:57.735218467-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:37:12.474455396-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":578,"events_lines":2}
+{"time":"2026-07-18T00:37:12.783356129-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:37:27.474914653-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":580,"events_lines":2}
+{"time":"2026-07-18T00:37:27.737584048-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:37:42.474742099-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":582,"events_lines":2}
+{"time":"2026-07-18T00:37:42.760691838-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:37:57.474641797-07:00","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":67,"history_lines":1,"events_offset":584,"events_lines":2,"console_offset":78,"console_lines":1}
+{"time":"2026-07-18T00:37:57.713880123-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:38:12.474166423-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":586,"events_lines":2}
+{"time":"2026-07-18T00:38:12.724125961-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-07-18T00:38:27.474963761-07:00","level":"INFO","msg":"filestream: sending request","total_files":1,"events_offset":588,"events_lines":2}
+{"time":"2026-07-18T00:38:27.744276654-07:00","level":"INFO","msg":"filestream: request sent","status":"200 OK"}

healed/calib/glean_keep75_s1224/wandb/debug.log

@@ -0,0 +1,19 @@
+2026-07-17 23:24:40,754 INFO    MainThread:416246 [wandb_setup.py:_flush():81] Current SDK version is 0.28.0
+2026-07-17 23:24:40,754 INFO    MainThread:416246 [wandb_setup.py:_flush():81] Configure stats pid to 416246
+2026-07-17 23:24:40,754 INFO    MainThread:416246 [wandb_setup.py:_flush():81] Loading settings from environment variables
+2026-07-17 23:24:40,754 INFO    MainThread:416246 [wandb_init.py:setup_run_log_directory():725] Logging user logs to outputs/healed/calib/glean_keep75_s1224/wandb/run-20260717_232440-ql8bd3ph/logs/debug.log
+2026-07-17 23:24:40,754 INFO    MainThread:416246 [wandb_init.py:setup_run_log_directory():726] Logging internal logs to outputs/healed/calib/glean_keep75_s1224/wandb/run-20260717_232440-ql8bd3ph/logs/debug-internal.log
+2026-07-17 23:24:40,754 INFO    MainThread:416246 [wandb_init.py:init():768] calling init triggers
+2026-07-17 23:24:40,754 INFO    MainThread:416246 [wandb_init.py:init():773] wandb.init called with sweep_config: {}
+config: {'student': 'outputs/pruned/glean-0125inst-general-keep75', 'teacher': 'allenai/OLMoE-1B-7B-0125-Instruct', 'training_mode': 'off-policy', 'kl_direction': 'forward', 'dataset': 'allenai/RLVR-MATH', 'dataset_sources': None, 'max_difficulty': None, 'trajectories': 'outputs/teacher_trajectories/dolci_math_curated.jsonl', 'trajectory_dataset': 'allenai/Dolci-Instruct-RL', 'off_policy_frames': 'chat', 'off_policy_max_seq_len': 2048, 'topk_targets': 'outputs/teacher_trajectories/dolci_combined_top128', 'max_loss_tokens': None, 'loss_tokens_per_step': 120000, 'teacher_device': 'cuda:0', 'student_device': 'cuda:0', 'lr': 3e-05, 'optimizer': 'adamw8bit', 'weight_decay': 0.1, 'epochs': 9, 'prompts_per_step': 256, 'group_size': 1, 'rollout_batch': 64, 'micro_batch': 3, 'max_new_tokens': 256, 'max_prompt_len': 1024, 'warmup_steps': 10, 'max_grad_norm': 1.0, 'eval_every': 10, 'gsm8k_every': 0, 'gsm8k_n': 256, 'gsm8k_batch': 16, 'gsm8k_max_new_tokens': 512, 'gsm8k_frames': 'chat', 'save_every': 300, 'out_dir': 'outputs/healed/calib/glean_keep75_s1224', 'sweep': 300, 'wandb': True, 'wandb_project': 'glean-calib', 'wandb_run_name': 'calib_keep75_300', 'wandb_run_id': None, 'wandb_resume': None, 'wandb_mode': 'online', 'no_wandb_sync': False, 'debug': False, 'resume_from': 'outputs/healed/calib/glean_keep75_s1224/step0200', 'start_step': 0, 'no_grad_checkpointing': False, 'seed': 1224, 'no_teacher_overlap': False, 'sync_checkpoints': False, 'rollout_engine': 'hf', 'vllm_gpu': None, 'vllm_port': 8377, 'vllm_refresh_every': 5, 'vllm_serve_bin': 'vllm-plugin/.venv/bin/python', 'vllm_gpu_mem_util': 0.85, 'vllm_refresh_mode': 'reload', 'vllm_live_dir': None, 'resolved_kl_direction': 'forward', '_wandb': {}}
+2026-07-17 23:24:40,754 INFO    MainThread:416246 [wandb_init.py:init():816] starting backend
+2026-07-17 23:24:40,754 INFO    MainThread:416246 [wandb_init.py:init():831] sending inform_init request
+2026-07-17 23:24:40,990 INFO    MainThread:416246 [wandb_init.py:init():836] backend started and connected
+2026-07-17 23:24:40,992 INFO    MainThread:416246 [wandb_init.py:init():906] updated telemetry
+2026-07-17 23:24:40,999 INFO    MainThread:416246 [wandb_init.py:init():929] communicating run to backend with 90.0 second timeout
+2026-07-17 23:24:41,225 INFO    MainThread:416246 [wandb_init.py:init():974] starting run threads in backend
+2026-07-17 23:24:41,359 INFO    MainThread:416246 [wandb_run.py:_console_start():2523] atexit reg
+2026-07-17 23:24:41,359 INFO    MainThread:416246 [wandb_run.py:_redirect():2373] redirect: wrap_raw
+2026-07-17 23:24:41,359 INFO    MainThread:416246 [wandb_run.py:_redirect():2442] Wrapping output streams.
+2026-07-17 23:24:41,359 INFO    MainThread:416246 [wandb_run.py:_redirect():2465] Redirects installed.
+2026-07-17 23:24:41,361 INFO    MainThread:416246 [wandb_init.py:init():1012] run started, returning control to user process

healed/calib/glean_keep75_s1224/wandb/run-20260717_194734-j20ncveb/files/config.yaml

@@ -0,0 +1,244 @@
+_wandb:
+    value:
+        cli_version: 0.28.0
+        e:
+            kfcng9b2jza5h90z4aoi5kb4e7j8ouke:
+                args:
+                    - --training-mode
+                    - off-policy
+                    - --kl-direction
+                    - forward
+                    - --topk-targets
+                    - outputs/teacher_trajectories/dolci_combined_top128
+                    - --off-policy-frames
+                    - chat
+                    - --student-device
+                    - cuda:0
+                    - --student
+                    - outputs/pruned/glean-0125inst-general-keep75
+                    - --lr
+                    - "3e-5"
+                    - --optimizer
+                    - adamw8bit
+                    - --epochs
+                    - "9"
+                    - --sweep
+                    - "100"
+                    - --micro-batch
+                    - "3"
+                    - --loss-tokens-per-step
+                    - "120000"
+                    - --gsm8k-every
+                    - "0"
+                    - --save-every
+                    - "100"
+                    - --seed
+                    - "1224"
+                    - --out-dir
+                    - outputs/healed/calib/glean_keep75_s1224
+                    - --wandb
+                    - --wandb-mode
+                    - online
+                    - --wandb-project
+                    - glean-calib
+                    - --wandb-run-name
+                    - calib_keep75_100
+                codePath: scripts/11_distill_on_policy.py
+                codePathLocal: scripts/11_distill_on_policy.py
+                cpu_count: 24
+                cpu_count_logical: 48
+                cudaVersion: "13.0"
+                disk:
+                    /:
+                        total: "1958315118592"
+                        used: "1624201732096"
+                email: hbfreed@protonmail.com
+                executable: /home/henry/Documents/PythonProjects/variable-reap/.venv/bin/python3
+                git:
+                    commit: 247c7f0113d7a058b1db8044deddfd4456cc998d
+                    remote: https://github.com/hbfreed/variable-reap.git
+                gpu: NVIDIA GeForce RTX 3090
+                gpu_count: 3
+                gpu_nvidia:
+                    - architecture: Ampere
+                      cudaCores: 10496
+                      memoryTotal: "25769803776"
+                      name: NVIDIA GeForce RTX 3090
+                      uuid: GPU-8ca70870-ddf2-d274-bcc5-182c2075bced
+                    - architecture: Ampere
+                      cudaCores: 10496
+                      memoryTotal: "25769803776"
+                      name: NVIDIA GeForce RTX 3090
+                      uuid: GPU-a6acf07f-31f5-618f-a5d0-c0017e7e2e27
+                    - architecture: Ampere
+                      cudaCores: 10496
+                      memoryTotal: "25769803776"
+                      name: NVIDIA GeForce RTX 3090
+                      uuid: GPU-864c54df-0130-7780-e271-8a5551d1733f
+                host: pop-os
+                memory:
+                    total: "134900756480"
+                os: Linux-7.0.11-76070011-generic-x86_64-with-glibc2.35
+                program: /home/henry/Documents/PythonProjects/variable-reap/scripts/11_distill_on_policy.py
+                python: CPython 3.12.12
+                root: outputs/healed/calib/glean_keep75_s1224
+                startedAt: "2026-07-18T02:47:34.113455Z"
+                writerId: kfcng9b2jza5h90z4aoi5kb4e7j8ouke
+        m:
+            - "1": step
+              "6":
+                - 3
+              "7": []
+            - "2": '*'
+              "5": 1
+              "6":
+                - 1
+              "7": []
+        python_version: 3.12.12
+        t:
+            "1":
+                - 1
+                - 5
+                - 11
+                - 49
+                - 51
+                - 53
+                - 71
+            "2":
+                - 1
+                - 5
+                - 11
+                - 49
+                - 51
+                - 53
+                - 71
+            "3":
+                - 2
+                - 7
+                - 13
+                - 16
+            "4": 3.12.12
+            "5": 0.28.0
+            "6": 4.57.6
+            "12": 0.28.0
+            "13": linux-x86_64
+dataset:
+    value: allenai/RLVR-MATH
+dataset_sources:
+    value: null
+debug:
+    value: false
+epochs:
+    value: 9
+eval_every:
+    value: 10
+group_size:
+    value: 1
+gsm8k_batch:
+    value: 16
+gsm8k_every:
+    value: 0
+gsm8k_frames:
+    value: chat
+gsm8k_max_new_tokens:
+    value: 512
+gsm8k_n:
+    value: 256
+kl_direction:
+    value: forward
+loss_tokens_per_step:
+    value: 120000
+lr:
+    value: 3e-05
+max_difficulty:
+    value: null
+max_grad_norm:
+    value: 1
+max_loss_tokens:
+    value: null
+max_new_tokens:
+    value: 256
+max_prompt_len:
+    value: 1024
+micro_batch:
+    value: 3
+no_grad_checkpointing:
+    value: false
+no_teacher_overlap:
+    value: false
+no_wandb_sync:
+    value: false
+off_policy_frames:
+    value: chat
+off_policy_max_seq_len:
+    value: 2048
+optimizer:
+    value: adamw8bit
+out_dir:
+    value: outputs/healed/calib/glean_keep75_s1224
+prompts_per_step:
+    value: 256
+resolved_kl_direction:
+    value: forward
+resume_from:
+    value: null
+rollout_batch:
+    value: 64
+rollout_engine:
+    value: hf
+save_every:
+    value: 100
+seed:
+    value: 1224
+start_step:
+    value: 0
+student:
+    value: outputs/pruned/glean-0125inst-general-keep75
+student_device:
+    value: cuda:0
+sweep:
+    value: 100
+sync_checkpoints:
+    value: false
+teacher:
+    value: allenai/OLMoE-1B-7B-0125-Instruct
+teacher_device:
+    value: cuda:0
+topk_targets:
+    value: outputs/teacher_trajectories/dolci_combined_top128
+training_mode:
+    value: off-policy
+trajectories:
+    value: outputs/teacher_trajectories/dolci_math_curated.jsonl
+trajectory_dataset:
+    value: allenai/Dolci-Instruct-RL
+vllm_gpu:
+    value: null
+vllm_gpu_mem_util:
+    value: 0.85
+vllm_live_dir:
+    value: null
+vllm_port:
+    value: 8377
+vllm_refresh_every:
+    value: 5
+vllm_refresh_mode:
+    value: reload
+vllm_serve_bin:
+    value: vllm-plugin/.venv/bin/python
+wandb:
+    value: true
+wandb_mode:
+    value: online
+wandb_project:
+    value: glean-calib
+wandb_resume:
+    value: null
+wandb_run_id:
+    value: null
+wandb_run_name:
+    value: calib_keep75_100
+warmup_steps:
+    value: 10
+weight_decay:
+    value: 0.1

healed/calib/glean_keep75_s1224/wandb/run-20260717_194734-j20ncveb/files/output.log

@@ -0,0 +1,115 @@
+Loading checkpoint shards: 100%|██████████| 3/3 [00:02<00:00,  1.04it/s]
+58360 cached top-128 chat trajectories / 22,295,631 unique tokens | 185 steps/epoch | 100 total steps | student params 5.31B | teacher overlap=False
+{"step": 1, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.13071662265943984, "tokens": 120000, "cumulative_loss_tokens": 120000, "grad_norm": 1.8671875, "lr": 6e-06, "finish_rate": 0.869, "comp_len": 463.3, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 84.6, "frames": {"chat": 259}, "mem_gb": 21.93}
+The attention mask is not set and cannot be inferred from input because pad token is same as eos token. As a consequence, you may observe unexpected behavior. Please pass your input's `attention_mask` to obtain reliable results.
+[eval step 1] sample: "To compute the rank of a matrix, we need to determine the maximum number of linearly independent rows or columns in the matrix. Here's how we can do it step-by-step:\n\n1. **Write down the matrix:**\n   "
+{"step": 2, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.1904543469694443, "tokens": 120000, "cumulative_loss_tokens": 240000, "grad_norm": 1.5859375, "lr": 9e-06, "finish_rate": 0.997, "comp_len": 344.8, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 69.7, "frames": {"chat": 348}, "mem_gb": 21.69}
+{"step": 3, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.17038091265981395, "tokens": 120000, "cumulative_loss_tokens": 360000, "grad_norm": 1.421875, "lr": 1.2e-05, "finish_rate": 0.997, "comp_len": 324.3, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 68.9, "frames": {"chat": 370}, "mem_gb": 21.84}
+{"step": 4, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.06029140768460929, "tokens": 120000, "cumulative_loss_tokens": 480000, "grad_norm": 0.87109375, "lr": 1.5e-05, "finish_rate": 0.852, "comp_len": 491.8, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 50.8, "frames": {"chat": 244}, "mem_gb": 22.07}
+{"step": 5, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.1347129997333667, "tokens": 120000, "cumulative_loss_tokens": 600000, "grad_norm": 0.94140625, "lr": 1.8e-05, "finish_rate": 0.953, "comp_len": 375.0, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 63.0, "frames": {"chat": 320}, "mem_gb": 21.99}
+{"step": 6, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.1584433459235588, "tokens": 120000, "cumulative_loss_tokens": 720000, "grad_norm": 0.91015625, "lr": 2.1e-05, "finish_rate": 0.997, "comp_len": 317.5, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 69.4, "frames": {"chat": 378}, "mem_gb": 21.59}
+{"step": 7, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.194529585396188, "tokens": 120000, "cumulative_loss_tokens": 840000, "grad_norm": 0.921875, "lr": 2.4e-05, "finish_rate": 0.997, "comp_len": 312.5, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 70.7, "frames": {"chat": 384}, "mem_gb": 22.08}
+{"step": 8, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.1841610342130841, "tokens": 120000, "cumulative_loss_tokens": 960000, "grad_norm": 0.9140625, "lr": 2.7000000000000002e-05, "finish_rate": 0.997, "comp_len": 370.4, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 65.7, "frames": {"chat": 324}, "mem_gb": 21.93}
+{"step": 9, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.06550765600646846, "tokens": 120000, "cumulative_loss_tokens": 1080000, "grad_norm": 0.90234375, "lr": 3e-05, "finish_rate": 0.891, "comp_len": 502.1, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 53.4, "frames": {"chat": 239}, "mem_gb": 22.09}
+{"step": 10, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.15740935421491353, "tokens": 120000, "cumulative_loss_tokens": 1200000, "grad_norm": 1.078125, "lr": 3e-05, "finish_rate": 0.961, "comp_len": 392.2, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 60.4, "frames": {"chat": 306}, "mem_gb": 21.83}
+[eval step 10] sample: 'To compute the rank of a matrix, we need to determine the maximum number of linearly independent rows or columns. A matrix is said to be full rank if all its rows (or columns) are linearly independent'
+{"step": 11, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.16742474506984775, "tokens": 120000, "cumulative_loss_tokens": 1320000, "grad_norm": 0.79296875, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 364.7, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 66.8, "frames": {"chat": 329}, "mem_gb": 21.87}
+{"step": 12, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.08396718050995841, "tokens": 120000, "cumulative_loss_tokens": 1440000, "grad_norm": 0.671875, "lr": 3e-05, "finish_rate": 0.882, "comp_len": 472.4, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 54.4, "frames": {"chat": 254}, "mem_gb": 22.1}
+{"step": 13, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.1496814745740189, "tokens": 120000, "cumulative_loss_tokens": 1560000, "grad_norm": 0.65234375, "lr": 3e-05, "finish_rate": 0.974, "comp_len": 439.6, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 58.8, "frames": {"chat": 273}, "mem_gb": 22.08}
+{"step": 14, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.04953998726032054, "tokens": 120000, "cumulative_loss_tokens": 1680000, "grad_norm": 0.423828125, "lr": 3e-05, "finish_rate": 0.873, "comp_len": 508.5, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 50.6, "frames": {"chat": 236}, "mem_gb": 22.0}
+{"step": 15, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.1870271451675333, "tokens": 120000, "cumulative_loss_tokens": 1800000, "grad_norm": 0.8046875, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 315.0, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 71.0, "frames": {"chat": 381}, "mem_gb": 21.73}
+{"step": 16, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.065772617749994, "tokens": 120000, "cumulative_loss_tokens": 1920000, "grad_norm": 0.5078125, "lr": 3e-05, "finish_rate": 0.925, "comp_len": 451.1, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 54.1, "frames": {"chat": 266}, "mem_gb": 21.93}
+{"step": 17, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.10379441037213741, "tokens": 120000, "cumulative_loss_tokens": 2040000, "grad_norm": 0.5234375, "lr": 3e-05, "finish_rate": 0.961, "comp_len": 389.6, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 62.6, "frames": {"chat": 308}, "mem_gb": 21.92}
+{"step": 18, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.09278505114205182, "tokens": 120000, "cumulative_loss_tokens": 2160000, "grad_norm": 0.484375, "lr": 3e-05, "finish_rate": 0.853, "comp_len": 439.6, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 56.1, "frames": {"chat": 273}, "mem_gb": 22.1}
+{"step": 19, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.15566241025024405, "tokens": 120000, "cumulative_loss_tokens": 2280000, "grad_norm": 0.6328125, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 326.1, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 68.0, "frames": {"chat": 368}, "mem_gb": 21.72}
+{"step": 20, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.2065904488075835, "tokens": 120000, "cumulative_loss_tokens": 2400000, "grad_norm": 0.66015625, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 354.0, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 67.6, "frames": {"chat": 339}, "mem_gb": 21.97}
+[eval step 20] sample: 'To find the rank of a matrix, we need to determine the linearly independent rows or columns. We can do this by performing row operations to transform the matrix into row-echelon form (REF).\n\nThe given'
+{"step": 21, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.09833553963121182, "tokens": 120000, "cumulative_loss_tokens": 2520000, "grad_norm": 0.48828125, "lr": 3e-05, "finish_rate": 0.963, "comp_len": 402.7, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 60.7, "frames": {"chat": 298}, "mem_gb": 21.89}
+{"step": 22, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.16023428318003813, "tokens": 120000, "cumulative_loss_tokens": 2640000, "grad_norm": 0.640625, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 342.9, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 66.8, "frames": {"chat": 350}, "mem_gb": 21.62}
+{"step": 23, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.08126048041522348, "tokens": 120000, "cumulative_loss_tokens": 2760000, "grad_norm": 0.462890625, "lr": 3e-05, "finish_rate": 0.858, "comp_len": 487.8, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 54.7, "frames": {"chat": 246}, "mem_gb": 22.06}
+{"step": 24, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.1679548702717526, "tokens": 120000, "cumulative_loss_tokens": 2880000, "grad_norm": 0.66796875, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 320.0, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 69.2, "frames": {"chat": 375}, "mem_gb": 21.79}
+{"step": 25, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.16708556045874332, "tokens": 120000, "cumulative_loss_tokens": 3000000, "grad_norm": 0.64453125, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 340.9, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 69.7, "frames": {"chat": 352}, "mem_gb": 21.85}
+{"step": 26, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.16134658801977833, "tokens": 120000, "cumulative_loss_tokens": 3120000, "grad_norm": 0.5546875, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 303.0, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 74.4, "frames": {"chat": 396}, "mem_gb": 21.8}
+{"step": 27, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.1342053831448313, "tokens": 120000, "cumulative_loss_tokens": 3240000, "grad_norm": 0.59765625, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 331.5, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 70.1, "frames": {"chat": 362}, "mem_gb": 21.66}
+{"step": 28, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.17713635508442918, "tokens": 120000, "cumulative_loss_tokens": 3360000, "grad_norm": 0.59375, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 325.2, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 71.9, "frames": {"chat": 369}, "mem_gb": 21.87}
+{"step": 29, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.1749166687601867, "tokens": 120000, "cumulative_loss_tokens": 3480000, "grad_norm": 0.609375, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 300.8, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 73.5, "frames": {"chat": 399}, "mem_gb": 21.72}
+{"step": 30, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.10207746837508554, "tokens": 120000, "cumulative_loss_tokens": 3600000, "grad_norm": 1.0859375, "lr": 3e-05, "finish_rate": 0.911, "comp_len": 485.8, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 55.3, "frames": {"chat": 247}, "mem_gb": 21.99}
+[eval step 30] sample: "To find the rank of a matrix, we need to determine the maximum number of linearly independent rows or columns. \n\nLet's start by examining the rows:\n\n1. **Row 1:** \\(12, -16, 4, 16\\)\n2. **Row 2:** \\( -"
+{"step": 31, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.03470097794899096, "tokens": 120000, "cumulative_loss_tokens": 3720000, "grad_norm": 0.48828125, "lr": 3e-05, "finish_rate": 0.891, "comp_len": 521.7, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 51.4, "frames": {"chat": 230}, "mem_gb": 21.93}
+{"step": 32, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.07190631164882022, "tokens": 120000, "cumulative_loss_tokens": 3840000, "grad_norm": 0.462890625, "lr": 3e-05, "finish_rate": 0.905, "comp_len": 456.3, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 58.8, "frames": {"chat": 263}, "mem_gb": 22.1}
+{"step": 33, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.08197221969100647, "tokens": 120000, "cumulative_loss_tokens": 3960000, "grad_norm": 0.58203125, "lr": 3e-05, "finish_rate": 0.848, "comp_len": 454.5, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 55.6, "frames": {"chat": 264}, "mem_gb": 22.09}
+{"step": 34, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.14606728826724769, "tokens": 120000, "cumulative_loss_tokens": 4080000, "grad_norm": 0.5859375, "lr": 3e-05, "finish_rate": 0.991, "comp_len": 365.9, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 69.9, "frames": {"chat": 328}, "mem_gb": 21.95}
+{"step": 35, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.07470543375237224, "tokens": 120000, "cumulative_loss_tokens": 4200000, "grad_norm": 0.57421875, "lr": 3e-05, "finish_rate": 0.925, "comp_len": 427.0, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 60.2, "frames": {"chat": 281}, "mem_gb": 22.02}
+{"step": 36, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.04788410694993411, "tokens": 120000, "cumulative_loss_tokens": 4320000, "grad_norm": 0.390625, "lr": 3e-05, "finish_rate": 0.861, "comp_len": 506.3, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 53.2, "frames": {"chat": 237}, "mem_gb": 22.01}
+{"step": 37, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.060693397747950316, "tokens": 120000, "cumulative_loss_tokens": 4440000, "grad_norm": 0.4453125, "lr": 3e-05, "finish_rate": 0.91, "comp_len": 431.7, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 56.8, "frames": {"chat": 278}, "mem_gb": 22.1}
+{"step": 38, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.13441723232055083, "tokens": 120000, "cumulative_loss_tokens": 4560000, "grad_norm": 0.5546875, "lr": 3e-05, "finish_rate": 0.955, "comp_len": 387.1, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 62.9, "frames": {"chat": 310}, "mem_gb": 21.97}
+{"step": 39, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.07622199243601256, "tokens": 120000, "cumulative_loss_tokens": 4680000, "grad_norm": 0.4296875, "lr": 3e-05, "finish_rate": 0.959, "comp_len": 406.8, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 59.5, "frames": {"chat": 295}, "mem_gb": 21.91}
+{"step": 40, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.09086993359592743, "tokens": 120000, "cumulative_loss_tokens": 4800000, "grad_norm": 0.470703125, "lr": 3e-05, "finish_rate": 0.957, "comp_len": 364.7, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 62.7, "frames": {"chat": 329}, "mem_gb": 21.92}
+[eval step 40] sample: 'To compute the rank of a matrix, we need to determine the maximum number of linearly independent rows or columns in the matrix. We can do this by using Python and the `numpy` library, which provides a'
+{"step": 41, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.08632465346843625, "tokens": 120000, "cumulative_loss_tokens": 4920000, "grad_norm": 0.46484375, "lr": 3e-05, "finish_rate": 0.905, "comp_len": 456.3, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 59.3, "frames": {"chat": 263}, "mem_gb": 22.04}
+{"step": 42, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.1030513121418655, "tokens": 120000, "cumulative_loss_tokens": 5040000, "grad_norm": 0.458984375, "lr": 3e-05, "finish_rate": 0.981, "comp_len": 381.0, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 61.1, "frames": {"chat": 315}, "mem_gb": 21.92}
+{"step": 43, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.13386018192392463, "tokens": 120000, "cumulative_loss_tokens": 5160000, "grad_norm": 0.53515625, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 339.0, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 70.1, "frames": {"chat": 354}, "mem_gb": 21.87}
+{"step": 44, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.07069139333600179, "tokens": 120000, "cumulative_loss_tokens": 5280000, "grad_norm": 0.5625, "lr": 3e-05, "finish_rate": 0.927, "comp_len": 463.3, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 55.8, "frames": {"chat": 259}, "mem_gb": 22.04}
+{"step": 45, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.09212584427177596, "tokens": 120000, "cumulative_loss_tokens": 5400000, "grad_norm": 0.4765625, "lr": 3e-05, "finish_rate": 0.92, "comp_len": 436.4, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 60.0, "frames": {"chat": 275}, "mem_gb": 21.9}
+{"step": 46, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.12813016946287825, "tokens": 120000, "cumulative_loss_tokens": 5520000, "grad_norm": 0.5625, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 359.3, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 69.8, "frames": {"chat": 334}, "mem_gb": 21.77}
+{"step": 47, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.14056541419519733, "tokens": 120000, "cumulative_loss_tokens": 5640000, "grad_norm": 0.53125, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 320.9, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 72.6, "frames": {"chat": 374}, "mem_gb": 21.62}
+{"step": 48, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.12320457009705715, "tokens": 120000, "cumulative_loss_tokens": 5760000, "grad_norm": 0.490234375, "lr": 3e-05, "finish_rate": 0.974, "comp_len": 385.9, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 65.6, "frames": {"chat": 311}, "mem_gb": 21.79}
+{"step": 49, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.09483733469761597, "tokens": 120000, "cumulative_loss_tokens": 5880000, "grad_norm": 0.431640625, "lr": 3e-05, "finish_rate": 0.876, "comp_len": 463.3, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 56.4, "frames": {"chat": 259}, "mem_gb": 22.1}
+{"step": 50, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.1235223285190761, "tokens": 120000, "cumulative_loss_tokens": 6000000, "grad_norm": 0.494140625, "lr": 3e-05, "finish_rate": 0.944, "comp_len": 393.4, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 64.8, "frames": {"chat": 305}, "mem_gb": 22.08}
+[eval step 50] sample: "To compute the rank of a matrix, we need to determine the maximum number of linearly independent rows or columns in the matrix. Let's break down the problem step-by-step:\n\n1. **Understand the Matrix:*"
+{"step": 51, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.10593181157653841, "tokens": 120000, "cumulative_loss_tokens": 6120000, "grad_norm": 0.478515625, "lr": 3e-05, "finish_rate": 0.946, "comp_len": 402.7, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 64.2, "frames": {"chat": 298}, "mem_gb": 22.06}
+{"step": 52, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.16442105506554555, "tokens": 120000, "cumulative_loss_tokens": 6240000, "grad_norm": 0.58203125, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 329.7, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 71.2, "frames": {"chat": 364}, "mem_gb": 21.84}
+{"step": 53, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.07198908675844626, "tokens": 120000, "cumulative_loss_tokens": 6360000, "grad_norm": 0.3984375, "lr": 3e-05, "finish_rate": 0.96, "comp_len": 397.4, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 63.8, "frames": {"chat": 302}, "mem_gb": 21.94}
+{"step": 54, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.15115031077408542, "tokens": 120000, "cumulative_loss_tokens": 6480000, "grad_norm": 0.57421875, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 379.7, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 67.3, "frames": {"chat": 316}, "mem_gb": 21.7}
+{"step": 55, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.15825922077251597, "tokens": 120000, "cumulative_loss_tokens": 6600000, "grad_norm": 0.54296875, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 384.6, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 69.3, "frames": {"chat": 312}, "mem_gb": 21.8}
+{"step": 56, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.1291787802760179, "tokens": 120000, "cumulative_loss_tokens": 6720000, "grad_norm": 0.52734375, "lr": 3e-05, "finish_rate": 0.979, "comp_len": 363.6, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 64.5, "frames": {"chat": 330}, "mem_gb": 21.95}
+{"step": 57, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.1187889644394784, "tokens": 120000, "cumulative_loss_tokens": 6840000, "grad_norm": 0.4921875, "lr": 3e-05, "finish_rate": 0.966, "comp_len": 373.8, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 65.2, "frames": {"chat": 321}, "mem_gb": 21.95}
+{"step": 58, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.10625948149410543, "tokens": 120000, "cumulative_loss_tokens": 6960000, "grad_norm": 0.474609375, "lr": 3e-05, "finish_rate": 0.968, "comp_len": 348.8, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 62.1, "frames": {"chat": 344}, "mem_gb": 21.99}
+{"step": 59, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.13131812240947038, "tokens": 120000, "cumulative_loss_tokens": 7080000, "grad_norm": 0.498046875, "lr": 3e-05, "finish_rate": 0.992, "comp_len": 339.9, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 66.9, "frames": {"chat": 353}, "mem_gb": 21.66}
+{"step": 60, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.06957110121420895, "tokens": 120000, "cumulative_loss_tokens": 7200000, "grad_norm": 0.3984375, "lr": 3e-05, "finish_rate": 0.89, "comp_len": 441.2, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 55.0, "frames": {"chat": 272}, "mem_gb": 22.09}
+[eval step 60] sample: "To compute the rank of a matrix, we need to determine the maximum number of linearly independent rows or columns in the matrix.\n\nLet's represent the given 4x4 matrix:\n\n\\[\n\\begin{bmatrix}\n12 & -16 & 4 "
+{"step": 61, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.14536656329245307, "tokens": 120000, "cumulative_loss_tokens": 7320000, "grad_norm": 0.51171875, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 378.5, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 60.1, "frames": {"chat": 317}, "mem_gb": 21.84}
+{"step": 62, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.057032847379900824, "tokens": 120000, "cumulative_loss_tokens": 7440000, "grad_norm": 0.380859375, "lr": 3e-05, "finish_rate": 0.949, "comp_len": 411.0, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 61.0, "frames": {"chat": 292}, "mem_gb": 21.84}
+{"step": 63, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.026177874839523187, "tokens": 120000, "cumulative_loss_tokens": 7560000, "grad_norm": 0.287109375, "lr": 3e-05, "finish_rate": 0.886, "comp_len": 524.0, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 52.9, "frames": {"chat": 229}, "mem_gb": 21.95}
+{"step": 64, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.14116634934321046, "tokens": 120000, "cumulative_loss_tokens": 7680000, "grad_norm": 0.52734375, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 334.3, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 74.2, "frames": {"chat": 359}, "mem_gb": 21.96}
+{"step": 65, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.12220887529390553, "tokens": 120000, "cumulative_loss_tokens": 7800000, "grad_norm": 0.498046875, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 335.2, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 72.1, "frames": {"chat": 358}, "mem_gb": 21.68}
+{"step": 66, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.14990485909172954, "tokens": 120000, "cumulative_loss_tokens": 7920000, "grad_norm": 0.52734375, "lr": 3e-05, "finish_rate": 0.975, "comp_len": 381.0, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 70.0, "frames": {"chat": 315}, "mem_gb": 22.06}
+{"step": 67, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.0956310116295858, "tokens": 120000, "cumulative_loss_tokens": 8040000, "grad_norm": 0.46484375, "lr": 3e-05, "finish_rate": 0.905, "comp_len": 454.5, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 61.2, "frames": {"chat": 264}, "mem_gb": 22.14}
+{"step": 68, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.13878010407060695, "tokens": 120000, "cumulative_loss_tokens": 8160000, "grad_norm": 0.51953125, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 356.1, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 70.5, "frames": {"chat": 337}, "mem_gb": 21.81}
+{"step": 69, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.07037615146461564, "tokens": 120000, "cumulative_loss_tokens": 8280000, "grad_norm": 0.45703125, "lr": 3e-05, "finish_rate": 0.873, "comp_len": 478.1, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 58.0, "frames": {"chat": 251}, "mem_gb": 22.1}
+{"step": 70, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.11031303176231838, "tokens": 120000, "cumulative_loss_tokens": 8400000, "grad_norm": 0.458984375, "lr": 3e-05, "finish_rate": 0.973, "comp_len": 402.7, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 64.0, "frames": {"chat": 298}, "mem_gb": 21.83}
+[eval step 70] sample: 'To find the rank of a matrix, we need to determine the maximum number of linearly independent rows or columns. A matrix is rank-deficient if it has fewer linearly independent rows or columns than its '
+{"step": 71, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.08863228327104201, "tokens": 120000, "cumulative_loss_tokens": 8520000, "grad_norm": 0.39453125, "lr": 3e-05, "finish_rate": 0.917, "comp_len": 454.5, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 56.7, "frames": {"chat": 264}, "mem_gb": 22.04}
+{"step": 72, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.08974066026785876, "tokens": 120000, "cumulative_loss_tokens": 8640000, "grad_norm": 0.421875, "lr": 3e-05, "finish_rate": 0.948, "comp_len": 393.4, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 61.8, "frames": {"chat": 305}, "mem_gb": 22.06}
+{"step": 73, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.07843934982779902, "tokens": 120000, "cumulative_loss_tokens": 8760000, "grad_norm": 0.400390625, "lr": 3e-05, "finish_rate": 0.953, "comp_len": 372.7, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 64.0, "frames": {"chat": 322}, "mem_gb": 21.83}
+{"step": 74, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.1308825180749409, "tokens": 120000, "cumulative_loss_tokens": 8880000, "grad_norm": 0.546875, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 332.4, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 71.5, "frames": {"chat": 361}, "mem_gb": 21.82}
+{"step": 75, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.09685897526154294, "tokens": 120000, "cumulative_loss_tokens": 9000000, "grad_norm": 0.427734375, "lr": 3e-05, "finish_rate": 0.972, "comp_len": 373.8, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 68.4, "frames": {"chat": 321}, "mem_gb": 22.0}
+{"step": 76, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.10711560201556422, "tokens": 120000, "cumulative_loss_tokens": 9120000, "grad_norm": 0.4765625, "lr": 3e-05, "finish_rate": 0.967, "comp_len": 390.9, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 65.5, "frames": {"chat": 307}, "mem_gb": 21.89}
+{"step": 77, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.13811318850273577, "tokens": 120000, "cumulative_loss_tokens": 9240000, "grad_norm": 0.54296875, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 367.0, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 70.8, "frames": {"chat": 327}, "mem_gb": 21.99}
+{"step": 78, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.15955485395295546, "tokens": 120000, "cumulative_loss_tokens": 9360000, "grad_norm": 0.58203125, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 337.1, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 70.8, "frames": {"chat": 356}, "mem_gb": 21.61}
+{"step": 79, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.07491358114203128, "tokens": 120000, "cumulative_loss_tokens": 9480000, "grad_norm": 0.43359375, "lr": 3e-05, "finish_rate": 0.915, "comp_len": 424.0, "t_data_s": 0.1, "t_rollout_s": 0.0, "t_step_s": 59.9, "frames": {"chat": 283}, "mem_gb": 21.95}
+{"step": 80, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.13741200089437577, "tokens": 120000, "cumulative_loss_tokens": 9600000, "grad_norm": 0.5, "lr": 3e-05, "finish_rate": 0.998, "comp_len": 299.3, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 76.6, "frames": {"chat": 401}, "mem_gb": 21.7}
+[eval step 80] sample: 'To find the rank of a matrix, we need to determine the maximum number of linearly independent rows or columns in the matrix. A matrix is said to be of rank \\( r \\) if it has \\( r \\) linearly independe'
+{"step": 81, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.08265723406509497, "tokens": 120000, "cumulative_loss_tokens": 9720000, "grad_norm": 0.400390625, "lr": 3e-05, "finish_rate": 0.965, "comp_len": 416.7, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 61.9, "frames": {"chat": 288}, "mem_gb": 21.9}
+{"step": 82, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.11680444544868854, "tokens": 120000, "cumulative_loss_tokens": 9840000, "grad_norm": 0.48828125, "lr": 3e-05, "finish_rate": 0.981, "comp_len": 371.5, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 67.8, "frames": {"chat": 323}, "mem_gb": 21.94}
+{"step": 83, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.1520622681313505, "tokens": 120000, "cumulative_loss_tokens": 9960000, "grad_norm": 0.55859375, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 357.1, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 67.9, "frames": {"chat": 336}, "mem_gb": 21.83}
+{"step": 84, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.1286243218161476, "tokens": 120000, "cumulative_loss_tokens": 10080000, "grad_norm": 0.515625, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 364.7, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 65.5, "frames": {"chat": 329}, "mem_gb": 21.74}
+{"step": 85, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.12255671736062504, "tokens": 120000, "cumulative_loss_tokens": 10200000, "grad_norm": 0.5, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 349.9, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 73.8, "frames": {"chat": 343}, "mem_gb": 21.72}
+{"step": 86, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.13805025024986825, "tokens": 120000, "cumulative_loss_tokens": 10320000, "grad_norm": 0.51953125, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 357.1, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 70.8, "frames": {"chat": 336}, "mem_gb": 21.84}
+{"step": 87, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.12031711872095863, "tokens": 120000, "cumulative_loss_tokens": 10440000, "grad_norm": 0.5078125, "lr": 3e-05, "finish_rate": 0.984, "comp_len": 382.2, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 65.1, "frames": {"chat": 314}, "mem_gb": 22.08}
+{"step": 88, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.07509244097399059, "tokens": 120000, "cumulative_loss_tokens": 10560000, "grad_norm": 0.37890625, "lr": 3e-05, "finish_rate": 0.968, "comp_len": 427.0, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 58.0, "frames": {"chat": 281}, "mem_gb": 21.91}
+{"step": 89, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.09532498086250077, "tokens": 120000, "cumulative_loss_tokens": 10680000, "grad_norm": 0.4453125, "lr": 3e-05, "finish_rate": 0.977, "comp_len": 401.3, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 62.5, "frames": {"chat": 299}, "mem_gb": 21.79}
+{"step": 90, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.12706806688119346, "tokens": 120000, "cumulative_loss_tokens": 10800000, "grad_norm": 0.466796875, "lr": 3e-05, "finish_rate": 0.966, "comp_len": 365.9, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 66.1, "frames": {"chat": 328}, "mem_gb": 21.98}
+[eval step 90] sample: 'To find the rank of a matrix, we need to determine the maximum number of linearly independent rows or columns in the matrix. A matrix is said to be of rank \\( r \\) if it has \\( r \\) linearly independe'
+{"step": 91, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.08623081075329489, "tokens": 120000, "cumulative_loss_tokens": 10920000, "grad_norm": 0.419921875, "lr": 3e-05, "finish_rate": 0.919, "comp_len": 442.8, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 54.0, "frames": {"chat": 271}, "mem_gb": 22.08}
+{"step": 92, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.09138209423874505, "tokens": 120000, "cumulative_loss_tokens": 11040000, "grad_norm": 0.431640625, "lr": 3e-05, "finish_rate": 0.874, "comp_len": 446.1, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 56.2, "frames": {"chat": 269}, "mem_gb": 22.18}
+{"step": 93, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.1642836628535685, "tokens": 120000, "cumulative_loss_tokens": 11160000, "grad_norm": 0.578125, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 324.3, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 69.3, "frames": {"chat": 370}, "mem_gb": 21.72}
+{"step": 94, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.17387420818852262, "tokens": 120000, "cumulative_loss_tokens": 11280000, "grad_norm": 0.546875, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 340.9, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 64.2, "frames": {"chat": 352}, "mem_gb": 21.78}
+{"step": 95, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.13778432187635142, "tokens": 120000, "cumulative_loss_tokens": 11400000, "grad_norm": 0.50390625, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 363.6, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 64.2, "frames": {"chat": 330}, "mem_gb": 21.99}
+{"step": 96, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.10431248163531806, "tokens": 120000, "cumulative_loss_tokens": 11520000, "grad_norm": 0.470703125, "lr": 3e-05, "finish_rate": 0.991, "comp_len": 359.3, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 62.3, "frames": {"chat": 334}, "mem_gb": 21.71}
+{"step": 97, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.1091810320273197, "tokens": 120000, "cumulative_loss_tokens": 11640000, "grad_norm": 0.484375, "lr": 3e-05, "finish_rate": 0.967, "comp_len": 357.1, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 58.5, "frames": {"chat": 336}, "mem_gb": 21.85}
+{"step": 98, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.0820784143627466, "tokens": 120000, "cumulative_loss_tokens": 11760000, "grad_norm": 0.443359375, "lr": 3e-05, "finish_rate": 0.964, "comp_len": 388.3, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 56.5, "frames": {"chat": 309}, "mem_gb": 21.93}
+{"step": 99, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.1385175480405645, "tokens": 120000, "cumulative_loss_tokens": 11880000, "grad_norm": 0.5234375, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 306.9, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 68.9, "frames": {"chat": 391}, "mem_gb": 21.82}
+{"step": 100, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.07162473855441591, "tokens": 120000, "cumulative_loss_tokens": 12000000, "grad_norm": 0.3671875, "lr": 3e-05, "finish_rate": 0.83, "comp_len": 454.5, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 57.5, "frames": {"chat": 264}, "mem_gb": 22.22}
+[eval step 100] sample: "To find the rank of a matrix, we need to determine the maximum number of linearly independent rows or columns in the matrix. \n\nLet's represent the given matrix:\n\n\\[\n\\begin{bmatrix}\n12 & -16 & 4 & 16 \\"
+checkpoint snapshot queued -> outputs/healed/calib/glean_keep75_s1224/step0100

healed/calib/glean_keep75_s1224/wandb/run-20260717_194734-j20ncveb/files/requirements.txt

@@ -0,0 +1,129 @@
+certifi==2026.6.17
+fonttools==4.63.0
+urllib3==2.7.0
+requests==2.34.2
+nvidia-cublas-cu12==12.6.4.1
+packaging==26.2
+regex==2026.6.28
+portalocker==3.2.0
+safetensors==0.8.0
+datasets==5.0.0
+sqlitedict==2.1.0
+narwhals==2.23.0
+latex2sympy2_extended==1.11.0
+smmap==5.0.3
+colorama==0.4.6
+anyio==4.14.1
+cycler==0.12.1
+pytablewriter==1.2.1
+pydantic_core==2.46.4
+huggingface_hub==0.36.2
+charset-normalizer==3.4.8
+nvidia-cusolver-cu12==11.7.1.2
+joblib==1.5.3
+typing-inspection==0.4.2
+wandb==0.28.0
+bitsandbytes==0.49.2
+kiwisolver==1.5.0
+cuda-toolkit==12.6.3
+propcache==0.5.2
+Jinja2==3.1.6
+cuda-bindings==12.9.7
+pluggy==1.6.0
+pytest==9.1.1
+fsspec==2026.4.0
+tqdm==4.68.3
+psutil==7.2.2
+pydantic==2.13.4
+typepy==1.3.5
+xxhash==3.8.1
+DataProperty==1.1.1
+aiosignal==1.4.0
+annotated-types==0.7.0
+Pygments==2.20.0
+zstandard==0.25.0
+threadpoolctl==3.6.0
+nvidia-cudnn-cu12==9.10.2.21
+evaluate==0.4.6
+pillow==12.3.0
+GitPython==3.1.50
+httpx==0.28.1
+lm_eval==0.4.12
+pytz==2026.2
+mbstrdecoder==1.1.5
+immutabledict==4.3.1
+dill==0.4.1
+mpmath==1.3.0
+yarl==1.24.2
+word2number==1.1
+networkx==3.6.1
+einops==0.8.2
+tabulate==0.10.0
+pathvalidate==3.3.1
+tenacity==9.1.4
+sentry-sdk==2.64.0
+nvidia-nvshmem-cu12==3.4.5
+hf-xet==1.5.1
+six==1.17.0
+pyarrow==24.0.0
+nvidia-curand-cu12==10.3.7.77
+cuda-pathfinder==1.5.6
+sacrebleu==2.6.0
+triton==3.7.1
+frozenlist==1.8.0
+antlr4-python3-runtime==4.11.0
+tabledata==1.3.5
+h11==0.16.0
+nvidia-cusparselt-cu12==0.7.1
+nvidia-nvjitlink-cu12==12.6.85
+scipy==1.18.0
+rouge_score==0.1.2
+transformers==4.57.6
+multiprocess==0.70.19
+absl-py==2.5.0
+accelerate==1.14.0
+click==8.4.2
+lxml==6.1.1
+defusedxml==0.7.1
+setuptools==81.0.0
+nvidia-cuda-cupti-cu12==12.6.80
+httpcore==1.0.9
+langdetect==1.0.9
+nltk==3.10.0
+iniconfig==2.3.0
+nvidia-cuda-runtime-cu12==12.6.77
+nvidia-cusparse-cu12==12.5.4.2
+matplotlib==3.11.0
+protobuf==7.35.1
+nvidia-cufft-cu12==11.3.0.4
+pyparsing==3.3.2
+chardet==6.0.0.post1
+typing_extensions==4.16.0
+stanford-stk==0.7.1
+aiohappyeyeballs==2.7.1
+MarkupSafe==3.0.3
+nvidia-cufile-cu12==1.11.1.6
+tcolorpy==0.1.7
+scikit-learn==1.9.0
+tokenizers==0.22.2
+attrs==26.1.0
+gitdb==4.0.12
+aiohttp==3.14.1
+glean==0.0.1
+platformdirs==4.10.0
+megablocks==0.11.0.dev0
+nvidia-cuda-nvrtc-cu12==12.6.85
+pandas==3.0.3
+multidict==6.7.1
+torch==2.12.1+cu126
+contourpy==1.3.3
+nvidia-nvtx-cu12==12.6.77
+math-verify==0.9.0
+filelock==3.29.5
+idna==3.18
+nvidia-nccl-cu12==2.29.3
+python-dateutil==2.9.0.post0
+more-itertools==11.1.0
+numpy==2.0.2
+sympy==1.14.0
+PyYAML==6.0.3

healed/calib/glean_keep75_s1224/wandb/run-20260717_194734-j20ncveb/files/wandb-metadata.json

@@ -0,0 +1,95 @@
+{
+  "os": "Linux-7.0.11-76070011-generic-x86_64-with-glibc2.35",
+  "python": "CPython 3.12.12",
+  "startedAt": "2026-07-18T02:47:34.113455Z",
+  "args": [
+    "--training-mode",
+    "off-policy",
+    "--kl-direction",
+    "forward",
+    "--topk-targets",
+    "outputs/teacher_trajectories/dolci_combined_top128",
+    "--off-policy-frames",
+    "chat",
+    "--student-device",
+    "cuda:0",
+    "--student",
+    "outputs/pruned/glean-0125inst-general-keep75",
+    "--lr",
+    "3e-5",
+    "--optimizer",
+    "adamw8bit",
+    "--epochs",
+    "9",
+    "--sweep",
+    "100",
+    "--micro-batch",
+    "3",
+    "--loss-tokens-per-step",
+    "120000",
+    "--gsm8k-every",
+    "0",
+    "--save-every",
+    "100",
+    "--seed",
+    "1224",
+    "--out-dir",
+    "outputs/healed/calib/glean_keep75_s1224",
+    "--wandb",
+    "--wandb-mode",
+    "online",
+    "--wandb-project",
+    "glean-calib",
+    "--wandb-run-name",
+    "calib_keep75_100"
+  ],
+  "program": "/home/henry/Documents/PythonProjects/variable-reap/scripts/11_distill_on_policy.py",
+  "codePath": "scripts/11_distill_on_policy.py",
+  "codePathLocal": "scripts/11_distill_on_policy.py",
+  "git": {
+    "remote": "https://github.com/hbfreed/variable-reap.git",
+    "commit": "247c7f0113d7a058b1db8044deddfd4456cc998d"
+  },
+  "email": "hbfreed@protonmail.com",
+  "root": "outputs/healed/calib/glean_keep75_s1224",
+  "host": "pop-os",
+  "executable": "/home/henry/Documents/PythonProjects/variable-reap/.venv/bin/python3",
+  "cpu_count": 24,
+  "cpu_count_logical": 48,
+  "gpu": "NVIDIA GeForce RTX 3090",
+  "gpu_count": 3,
+  "disk": {
+    "/": {
+      "total": "1958315118592",
+      "used": "1624201732096"
+    }
+  },
+  "memory": {
+    "total": "134900756480"
+  },
+  "gpu_nvidia": [
+    {
+      "name": "NVIDIA GeForce RTX 3090",
+      "memoryTotal": "25769803776",
+      "cudaCores": 10496,
+      "architecture": "Ampere",
+      "uuid": "GPU-8ca70870-ddf2-d274-bcc5-182c2075bced"
+    },
+    {
+      "name": "NVIDIA GeForce RTX 3090",
+      "memoryTotal": "25769803776",
+      "cudaCores": 10496,
+      "architecture": "Ampere",
+      "uuid": "GPU-a6acf07f-31f5-618f-a5d0-c0017e7e2e27"
+    },
+    {
+      "name": "NVIDIA GeForce RTX 3090",
+      "memoryTotal": "25769803776",
+      "cudaCores": 10496,
+      "architecture": "Ampere",
+      "uuid": "GPU-864c54df-0130-7780-e271-8a5551d1733f"
+    }
+  ],
+  "cudaVersion": "13.0",
+  "writerId": "kfcng9b2jza5h90z4aoi5kb4e7j8ouke"
+}

healed/calib/glean_keep75_s1224/wandb/run-20260717_194734-j20ncveb/files/wandb-summary.json

@@ -0,0 +1 @@
+{"t_step_s":57.5,"finish_rate":0.83,"grad_norm":0.3671875,"_wandb":{"runtime":6470},"cumulative_loss_tokens":12000000,"forward_topk_kl":0.07162473855441591,"t_data_s":0,"training_mode":"off-policy","mem_gb":22.22,"_step":99,"_runtime":6470,"epoch":0,"step":100,"t_rollout_s":0,"frames":{"chat":264},"lr":3e-05,"tokens":120000,"_timestamp":1.7843492888294733e+09,"comp_len":454.5}

healed/calib/glean_keep75_s1224/wandb/run-20260717_194734-j20ncveb/logs/debug-core.log

@@ -0,0 +1,20 @@
+{"time":"2026-07-17T19:47:33.658043042-07:00","level":"INFO","msg":"main: starting server","port-filename":"/tmp/tmpi276dcan/port-388066.txt","pid":388066,"detached":false,"idle-timeout":600000000000,"log-level":0,"disable-analytics":false,"shutdown-on-parent-exit":false,"enable-dcgm-profiling":false}
+{"time":"2026-07-17T19:47:33.659896172-07:00","level":"INFO","msg":"server: will exit if parent process dies","ppid":388066}
+{"time":"2026-07-17T19:47:33.659872842-07:00","level":"INFO","msg":"server: accepting connections","addr":{"Name":"/tmp/wandb-388066-388175-2900257119/socket","Net":"unix"}}
+{"time":"2026-07-17T19:47:33.839724127-07:00","level":"INFO","msg":"connection: ManageConnectionData: new connection created","id":"1(@)"}
+{"time":"2026-07-17T19:47:34.117152253-07:00","level":"INFO","msg":"handleInformInit: received","streamId":"j20ncveb","id":"1(@)"}
+{"time":"2026-07-17T19:47:34.447830587-07:00","level":"INFO","msg":"handleInformInit: stream started","streamId":"j20ncveb","id":"1(@)"}
+{"time":"2026-07-17T19:47:39.832355986-07:00","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"220zch5j21yb"}
+{"time":"2026-07-17T21:35:25.407436018-07:00","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"220zch5j21yb"}
+{"time":"2026-07-17T21:35:26.192436203-07:00","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"220zch5j21yb"}
+{"time":"2026-07-17T21:35:26.195765546-07:00","level":"INFO","msg":"handleInformFinish: finish message received","streamId":"j20ncveb","id":"1(@)"}
+{"time":"2026-07-17T21:35:26.203672705-07:00","level":"INFO","msg":"handleInformFinish: stream closed","streamId":"j20ncveb","id":"1(@)"}
+{"time":"2026-07-17T21:35:26.262771789-07:00","level":"INFO","msg":"handleInformTeardown: server teardown initiated","id":"1(@)"}
+{"time":"2026-07-17T21:35:26.262839011-07:00","level":"INFO","msg":"handleInformTeardown: server shutdown complete","id":"1(@)"}
+{"time":"2026-07-17T21:35:26.262852762-07:00","level":"INFO","msg":"server is shutting down"}
+{"time":"2026-07-17T21:35:26.262856312-07:00","level":"INFO","msg":"connection: closing","id":"1(@)"}
+{"time":"2026-07-17T21:35:26.262903253-07:00","level":"INFO","msg":"processOutgoingData: finished","id":"1(@)"}
+{"time":"2026-07-17T21:35:26.262932804-07:00","level":"INFO","msg":"connection: closed successfully","id":"1(@)"}
+{"time":"2026-07-17T21:35:26.262947345-07:00","level":"INFO","msg":"connection: ManageConnectionData: connection closed","id":"1(@)"}
+{"time":"2026-07-17T21:35:26.263953279-07:00","level":"INFO","msg":"server: listener closed","addr":{"Name":"/tmp/wandb-388066-388175-2900257119/socket","Net":"unix"}}
+{"time":"2026-07-17T21:35:26.263993281-07:00","level":"INFO","msg":"server is closed"}

healed/calib/glean_keep75_s1224/wandb/run-20260717_194734-j20ncveb/logs/debug.log

@@ -0,0 +1,24 @@
+2026-07-17 19:47:34,115 INFO    MainThread:388066 [wandb_setup.py:_flush():81] Current SDK version is 0.28.0
+2026-07-17 19:47:34,115 INFO    MainThread:388066 [wandb_setup.py:_flush():81] Configure stats pid to 388066
+2026-07-17 19:47:34,115 INFO    MainThread:388066 [wandb_setup.py:_flush():81] Loading settings from environment variables
+2026-07-17 19:47:34,115 INFO    MainThread:388066 [wandb_init.py:setup_run_log_directory():725] Logging user logs to outputs/healed/calib/glean_keep75_s1224/wandb/run-20260717_194734-j20ncveb/logs/debug.log
+2026-07-17 19:47:34,115 INFO    MainThread:388066 [wandb_init.py:setup_run_log_directory():726] Logging internal logs to outputs/healed/calib/glean_keep75_s1224/wandb/run-20260717_194734-j20ncveb/logs/debug-internal.log
+2026-07-17 19:47:34,115 INFO    MainThread:388066 [wandb_init.py:init():768] calling init triggers
+2026-07-17 19:47:34,115 INFO    MainThread:388066 [wandb_init.py:init():773] wandb.init called with sweep_config: {}
+config: {'student': 'outputs/pruned/glean-0125inst-general-keep75', 'teacher': 'allenai/OLMoE-1B-7B-0125-Instruct', 'training_mode': 'off-policy', 'kl_direction': 'forward', 'dataset': 'allenai/RLVR-MATH', 'dataset_sources': None, 'max_difficulty': None, 'trajectories': 'outputs/teacher_trajectories/dolci_math_curated.jsonl', 'trajectory_dataset': 'allenai/Dolci-Instruct-RL', 'off_policy_frames': 'chat', 'off_policy_max_seq_len': 2048, 'topk_targets': 'outputs/teacher_trajectories/dolci_combined_top128', 'max_loss_tokens': None, 'loss_tokens_per_step': 120000, 'teacher_device': 'cuda:0', 'student_device': 'cuda:0', 'lr': 3e-05, 'optimizer': 'adamw8bit', 'weight_decay': 0.1, 'epochs': 9, 'prompts_per_step': 256, 'group_size': 1, 'rollout_batch': 64, 'micro_batch': 3, 'max_new_tokens': 256, 'max_prompt_len': 1024, 'warmup_steps': 10, 'max_grad_norm': 1.0, 'eval_every': 10, 'gsm8k_every': 0, 'gsm8k_n': 256, 'gsm8k_batch': 16, 'gsm8k_max_new_tokens': 512, 'gsm8k_frames': 'chat', 'save_every': 100, 'out_dir': 'outputs/healed/calib/glean_keep75_s1224', 'sweep': 100, 'wandb': True, 'wandb_project': 'glean-calib', 'wandb_run_name': 'calib_keep75_100', 'wandb_run_id': None, 'wandb_resume': None, 'wandb_mode': 'online', 'no_wandb_sync': False, 'debug': False, 'resume_from': None, 'start_step': 0, 'no_grad_checkpointing': False, 'seed': 1224, 'no_teacher_overlap': False, 'sync_checkpoints': False, 'rollout_engine': 'hf', 'vllm_gpu': None, 'vllm_port': 8377, 'vllm_refresh_every': 5, 'vllm_serve_bin': 'vllm-plugin/.venv/bin/python', 'vllm_gpu_mem_util': 0.85, 'vllm_refresh_mode': 'reload', 'vllm_live_dir': None, 'resolved_kl_direction': 'forward', '_wandb': {}}
+2026-07-17 19:47:34,115 INFO    MainThread:388066 [wandb_init.py:init():816] starting backend
+2026-07-17 19:47:34,115 INFO    MainThread:388066 [wandb_init.py:init():831] sending inform_init request
+2026-07-17 19:47:34,448 INFO    MainThread:388066 [wandb_init.py:init():836] backend started and connected
+2026-07-17 19:47:34,450 INFO    MainThread:388066 [wandb_init.py:init():906] updated telemetry
+2026-07-17 19:47:34,456 INFO    MainThread:388066 [wandb_init.py:init():929] communicating run to backend with 90.0 second timeout
+2026-07-17 19:47:34,695 INFO    MainThread:388066 [wandb_init.py:init():974] starting run threads in backend
+2026-07-17 19:47:34,825 INFO    MainThread:388066 [wandb_run.py:_console_start():2523] atexit reg
+2026-07-17 19:47:34,826 INFO    MainThread:388066 [wandb_run.py:_redirect():2373] redirect: wrap_raw
+2026-07-17 19:47:34,826 INFO    MainThread:388066 [wandb_run.py:_redirect():2442] Wrapping output streams.
+2026-07-17 19:47:34,826 INFO    MainThread:388066 [wandb_run.py:_redirect():2465] Redirects installed.
+2026-07-17 19:47:34,828 INFO    MainThread:388066 [wandb_init.py:init():1012] run started, returning control to user process
+2026-07-17 21:35:25,388 INFO    MainThread:388066 [wandb_run.py:_finish():2285] finishing run hbfreed/glean-calib/j20ncveb
+2026-07-17 21:35:25,406 INFO    MainThread:388066 [wandb_run.py:_atexit_cleanup():2490] got exitcode: 0
+2026-07-17 21:35:25,406 INFO    MainThread:388066 [wandb_run.py:_restore():2472] restore
+2026-07-17 21:35:25,407 INFO    MainThread:388066 [wandb_run.py:_restore():2478] restore done
+2026-07-17 21:35:26,194 INFO    MainThread:388066 [wandb_run.py:_footer_sync_info():3895] logging synced files

healed/calib/glean_keep75_s1224/wandb/run-20260717_213714-mkqm7no2/files/config.yaml

@@ -0,0 +1,246 @@
+_wandb:
+    value:
+        cli_version: 0.28.0
+        e:
+            0c0c5qhc7ujrqprydps7c2ecqc1afsdo:
+                args:
+                    - --training-mode
+                    - off-policy
+                    - --kl-direction
+                    - forward
+                    - --topk-targets
+                    - outputs/teacher_trajectories/dolci_combined_top128
+                    - --off-policy-frames
+                    - chat
+                    - --student-device
+                    - cuda:0
+                    - --student
+                    - outputs/pruned/glean-0125inst-general-keep75
+                    - --resume-from
+                    - outputs/healed/calib/glean_keep75_s1224/step0100
+                    - --lr
+                    - "3e-5"
+                    - --optimizer
+                    - adamw8bit
+                    - --epochs
+                    - "9"
+                    - --sweep
+                    - "200"
+                    - --micro-batch
+                    - "3"
+                    - --loss-tokens-per-step
+                    - "120000"
+                    - --gsm8k-every
+                    - "0"
+                    - --save-every
+                    - "200"
+                    - --seed
+                    - "1224"
+                    - --out-dir
+                    - outputs/healed/calib/glean_keep75_s1224
+                    - --wandb
+                    - --wandb-mode
+                    - online
+                    - --wandb-project
+                    - glean-calib
+                    - --wandb-run-name
+                    - calib_keep75_200
+                codePath: scripts/11_distill_on_policy.py
+                codePathLocal: scripts/11_distill_on_policy.py
+                cpu_count: 24
+                cpu_count_logical: 48
+                cudaVersion: "13.0"
+                disk:
+                    /:
+                        total: "1958315118592"
+                        used: "1668944318464"
+                email: hbfreed@protonmail.com
+                executable: /home/henry/Documents/PythonProjects/variable-reap/.venv/bin/python3
+                git:
+                    commit: 247c7f0113d7a058b1db8044deddfd4456cc998d
+                    remote: https://github.com/hbfreed/variable-reap.git
+                gpu: NVIDIA GeForce RTX 3090
+                gpu_count: 3
+                gpu_nvidia:
+                    - architecture: Ampere
+                      cudaCores: 10496
+                      memoryTotal: "25769803776"
+                      name: NVIDIA GeForce RTX 3090
+                      uuid: GPU-8ca70870-ddf2-d274-bcc5-182c2075bced
+                    - architecture: Ampere
+                      cudaCores: 10496
+                      memoryTotal: "25769803776"
+                      name: NVIDIA GeForce RTX 3090
+                      uuid: GPU-a6acf07f-31f5-618f-a5d0-c0017e7e2e27
+                    - architecture: Ampere
+                      cudaCores: 10496
+                      memoryTotal: "25769803776"
+                      name: NVIDIA GeForce RTX 3090
+                      uuid: GPU-864c54df-0130-7780-e271-8a5551d1733f
+                host: pop-os
+                memory:
+                    total: "134900756480"
+                os: Linux-7.0.11-76070011-generic-x86_64-with-glibc2.35
+                program: /home/henry/Documents/PythonProjects/variable-reap/scripts/11_distill_on_policy.py
+                python: CPython 3.12.12
+                root: outputs/healed/calib/glean_keep75_s1224
+                startedAt: "2026-07-18T04:37:14.633800Z"
+                writerId: 0c0c5qhc7ujrqprydps7c2ecqc1afsdo
+        m:
+            - "1": step
+              "6":
+                - 3
+              "7": []
+            - "2": '*'
+              "5": 1
+              "6":
+                - 1
+              "7": []
+        python_version: 3.12.12
+        t:
+            "1":
+                - 1
+                - 5
+                - 11
+                - 49
+                - 51
+                - 53
+                - 71
+            "2":
+                - 1
+                - 5
+                - 11
+                - 49
+                - 51
+                - 53
+                - 71
+            "3":
+                - 2
+                - 7
+                - 13
+                - 16
+            "4": 3.12.12
+            "5": 0.28.0
+            "6": 4.57.6
+            "12": 0.28.0
+            "13": linux-x86_64
+dataset:
+    value: allenai/RLVR-MATH
+dataset_sources:
+    value: null
+debug:
+    value: false
+epochs:
+    value: 9
+eval_every:
+    value: 10
+group_size:
+    value: 1
+gsm8k_batch:
+    value: 16
+gsm8k_every:
+    value: 0
+gsm8k_frames:
+    value: chat
+gsm8k_max_new_tokens:
+    value: 512
+gsm8k_n:
+    value: 256
+kl_direction:
+    value: forward
+loss_tokens_per_step:
+    value: 120000
+lr:
+    value: 3e-05
+max_difficulty:
+    value: null
+max_grad_norm:
+    value: 1
+max_loss_tokens:
+    value: null
+max_new_tokens:
+    value: 256
+max_prompt_len:
+    value: 1024
+micro_batch:
+    value: 3
+no_grad_checkpointing:
+    value: false
+no_teacher_overlap:
+    value: false
+no_wandb_sync:
+    value: false
+off_policy_frames:
+    value: chat
+off_policy_max_seq_len:
+    value: 2048
+optimizer:
+    value: adamw8bit
+out_dir:
+    value: outputs/healed/calib/glean_keep75_s1224
+prompts_per_step:
+    value: 256
+resolved_kl_direction:
+    value: forward
+resume_from:
+    value: outputs/healed/calib/glean_keep75_s1224/step0100
+rollout_batch:
+    value: 64
+rollout_engine:
+    value: hf
+save_every:
+    value: 200
+seed:
+    value: 1224
+start_step:
+    value: 0
+student:
+    value: outputs/pruned/glean-0125inst-general-keep75
+student_device:
+    value: cuda:0
+sweep:
+    value: 200
+sync_checkpoints:
+    value: false
+teacher:
+    value: allenai/OLMoE-1B-7B-0125-Instruct
+teacher_device:
+    value: cuda:0
+topk_targets:
+    value: outputs/teacher_trajectories/dolci_combined_top128
+training_mode:
+    value: off-policy
+trajectories:
+    value: outputs/teacher_trajectories/dolci_math_curated.jsonl
+trajectory_dataset:
+    value: allenai/Dolci-Instruct-RL
+vllm_gpu:
+    value: null
+vllm_gpu_mem_util:
+    value: 0.85
+vllm_live_dir:
+    value: null
+vllm_port:
+    value: 8377
+vllm_refresh_every:
+    value: 5
+vllm_refresh_mode:
+    value: reload
+vllm_serve_bin:
+    value: vllm-plugin/.venv/bin/python
+wandb:
+    value: true
+wandb_mode:
+    value: online
+wandb_project:
+    value: glean-calib
+wandb_resume:
+    value: null
+wandb_run_id:
+    value: null
+wandb_run_name:
+    value: calib_keep75_200
+warmup_steps:
+    value: 10
+weight_decay:
+    value: 0.1

healed/calib/glean_keep75_s1224/wandb/run-20260717_213714-mkqm7no2/files/output.log

@@ -0,0 +1,116 @@
+Loading checkpoint shards: 100%|██████████| 3/3 [00:00<00:00, 10.43it/s]
+resumed student weights from outputs/healed/calib/glean_keep75_s1224/step0100 (fresh optimizer, step counter at 0)
+58360 cached top-128 chat trajectories / 22,295,631 unique tokens | 185 steps/epoch | 200 total steps | student params 5.31B | teacher overlap=False
+restored optimizer/scheduler state from step 100; rebuilt 260 paged buffers
+{"step": 101, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.12676862803295566, "tokens": 120000, "cumulative_loss_tokens": 12120000, "grad_norm": 0.5, "lr": 3e-05, "finish_rate": 0.994, "comp_len": 343.8, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 73.8, "frames": {"chat": 349}, "mem_gb": 21.61}
+{"step": 102, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.12114182180125888, "tokens": 120000, "cumulative_loss_tokens": 12240000, "grad_norm": 0.51171875, "lr": 3e-05, "finish_rate": 0.968, "comp_len": 384.6, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 58.3, "frames": {"chat": 312}, "mem_gb": 21.82}
+{"step": 103, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.03901445078571172, "tokens": 120000, "cumulative_loss_tokens": 12360000, "grad_norm": 0.314453125, "lr": 3e-05, "finish_rate": 0.817, "comp_len": 524.0, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 48.7, "frames": {"chat": 229}, "mem_gb": 22.09}
+{"step": 104, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.1029369749694597, "tokens": 120000, "cumulative_loss_tokens": 12480000, "grad_norm": 0.4453125, "lr": 3e-05, "finish_rate": 0.938, "comp_len": 416.7, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 57.0, "frames": {"chat": 288}, "mem_gb": 22.09}
+{"step": 105, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.14577210398303966, "tokens": 120000, "cumulative_loss_tokens": 12600000, "grad_norm": 0.515625, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 341.9, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 64.0, "frames": {"chat": 351}, "mem_gb": 21.75}
+{"step": 106, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.042434302429893676, "tokens": 120000, "cumulative_loss_tokens": 12720000, "grad_norm": 0.341796875, "lr": 3e-05, "finish_rate": 0.812, "comp_len": 538.1, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 46.0, "frames": {"chat": 223}, "mem_gb": 22.12}
+{"step": 107, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.023880430516523, "tokens": 120000, "cumulative_loss_tokens": 12840000, "grad_norm": 0.30078125, "lr": 3e-05, "finish_rate": 0.793, "comp_len": 563.4, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 45.2, "frames": {"chat": 213}, "mem_gb": 22.08}
+{"step": 108, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.1253991394669128, "tokens": 120000, "cumulative_loss_tokens": 12960000, "grad_norm": 0.51171875, "lr": 3e-05, "finish_rate": 0.995, "comp_len": 315.8, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 68.9, "frames": {"chat": 380}, "mem_gb": 21.93}
+{"step": 109, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.07316874562174393, "tokens": 120000, "cumulative_loss_tokens": 13080000, "grad_norm": 0.4140625, "lr": 3e-05, "finish_rate": 0.893, "comp_len": 442.8, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 50.6, "frames": {"chat": 271}, "mem_gb": 22.11}
+{"step": 110, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.15606389061900167, "tokens": 120000, "cumulative_loss_tokens": 13200000, "grad_norm": 0.50390625, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 352.9, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 62.9, "frames": {"chat": 340}, "mem_gb": 22.01}
+The attention mask is not set and cannot be inferred from input because pad token is same as eos token. As a consequence, you may observe unexpected behavior. Please pass your input's `attention_mask` to obtain reliable results.
+[eval step 110] sample: 'To find the rank of a matrix, we need to determine the maximum number of linearly independent rows or columns in the matrix. A matrix is said to be of rank \\( r \\) if it has \\( r \\) linearly independe'
+{"step": 111, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.14502378453067505, "tokens": 120000, "cumulative_loss_tokens": 13320000, "grad_norm": 0.5234375, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 327.9, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 67.8, "frames": {"chat": 366}, "mem_gb": 21.7}
+{"step": 112, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.044772777596488594, "tokens": 120000, "cumulative_loss_tokens": 13440000, "grad_norm": 0.326171875, "lr": 3e-05, "finish_rate": 0.854, "comp_len": 515.0, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 47.5, "frames": {"chat": 233}, "mem_gb": 22.09}
+{"step": 113, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.09647136719362268, "tokens": 120000, "cumulative_loss_tokens": 13560000, "grad_norm": 0.45703125, "lr": 3e-05, "finish_rate": 0.968, "comp_len": 384.6, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 58.7, "frames": {"chat": 312}, "mem_gb": 21.96}
+{"step": 114, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.14180562431244956, "tokens": 120000, "cumulative_loss_tokens": 13680000, "grad_norm": 0.55859375, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 330.6, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 64.9, "frames": {"chat": 363}, "mem_gb": 21.67}
+{"step": 115, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.14334856677559826, "tokens": 120000, "cumulative_loss_tokens": 13800000, "grad_norm": 0.51171875, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 371.5, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 60.1, "frames": {"chat": 323}, "mem_gb": 21.94}
+{"step": 116, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.14154685637778602, "tokens": 120000, "cumulative_loss_tokens": 13920000, "grad_norm": 0.55859375, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 346.8, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 64.4, "frames": {"chat": 346}, "mem_gb": 21.69}
+{"step": 117, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.1396051885420767, "tokens": 120000, "cumulative_loss_tokens": 14040000, "grad_norm": 0.5, "lr": 3e-05, "finish_rate": 0.975, "comp_len": 377.4, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 58.0, "frames": {"chat": 318}, "mem_gb": 21.83}
+{"step": 118, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.11092782069956884, "tokens": 120000, "cumulative_loss_tokens": 14160000, "grad_norm": 0.455078125, "lr": 3e-05, "finish_rate": 0.935, "comp_len": 436.4, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 58.1, "frames": {"chat": 275}, "mem_gb": 22.02}
+{"step": 119, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.12665537766687726, "tokens": 120000, "cumulative_loss_tokens": 14280000, "grad_norm": 0.4921875, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 347.8, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 64.8, "frames": {"chat": 345}, "mem_gb": 21.76}
+{"step": 120, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.120160753988754, "tokens": 120000, "cumulative_loss_tokens": 14400000, "grad_norm": 0.53515625, "lr": 3e-05, "finish_rate": 0.992, "comp_len": 318.3, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 66.6, "frames": {"chat": 377}, "mem_gb": 21.91}
+[eval step 120] sample: 'To compute the rank of a matrix, we need to determine the maximum number of linearly independent rows or columns in the matrix. A matrix is rank-deficient if it has fewer linearly independent rows or '
+{"step": 121, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.09895123931532726, "tokens": 120000, "cumulative_loss_tokens": 14520000, "grad_norm": 0.43359375, "lr": 3e-05, "finish_rate": 0.956, "comp_len": 408.2, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 57.6, "frames": {"chat": 294}, "mem_gb": 22.14}
+{"step": 122, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.1409571954037218, "tokens": 120000, "cumulative_loss_tokens": 14640000, "grad_norm": 0.51171875, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 355.0, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 63.2, "frames": {"chat": 338}, "mem_gb": 21.83}
+{"step": 123, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.15211505075778503, "tokens": 120000, "cumulative_loss_tokens": 14760000, "grad_norm": 0.51953125, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 340.9, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 66.0, "frames": {"chat": 352}, "mem_gb": 21.73}
+{"step": 124, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.1332457024556352, "tokens": 120000, "cumulative_loss_tokens": 14880000, "grad_norm": 0.49609375, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 345.8, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 67.9, "frames": {"chat": 347}, "mem_gb": 21.78}
+{"step": 125, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.08373471758464972, "tokens": 120000, "cumulative_loss_tokens": 15000000, "grad_norm": 0.404296875, "lr": 3e-05, "finish_rate": 0.968, "comp_len": 387.1, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 57.8, "frames": {"chat": 310}, "mem_gb": 21.76}
+{"step": 126, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.15742334490390494, "tokens": 120000, "cumulative_loss_tokens": 15120000, "grad_norm": 0.578125, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 335.2, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 66.3, "frames": {"chat": 358}, "mem_gb": 21.78}
+{"step": 127, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.07713352295395452, "tokens": 120000, "cumulative_loss_tokens": 15240000, "grad_norm": 0.427734375, "lr": 3e-05, "finish_rate": 0.947, "comp_len": 376.2, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 59.3, "frames": {"chat": 319}, "mem_gb": 22.06}
+{"step": 128, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.052739765353671586, "tokens": 120000, "cumulative_loss_tokens": 15360000, "grad_norm": 0.3515625, "lr": 3e-05, "finish_rate": 0.9, "comp_len": 446.1, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 54.5, "frames": {"chat": 269}, "mem_gb": 22.09}
+{"step": 129, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.10360154020698586, "tokens": 120000, "cumulative_loss_tokens": 15480000, "grad_norm": 0.466796875, "lr": 3e-05, "finish_rate": 0.974, "comp_len": 387.1, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 63.2, "frames": {"chat": 310}, "mem_gb": 21.94}
+{"step": 130, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.06664360370513363, "tokens": 120000, "cumulative_loss_tokens": 15600000, "grad_norm": 0.41796875, "lr": 3e-05, "finish_rate": 0.834, "comp_len": 463.3, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 55.2, "frames": {"chat": 259}, "mem_gb": 22.12}
+[eval step 130] sample: "To find the rank of a matrix, we need to determine the maximum number of linearly independent rows or columns in the matrix. \n\nLet's represent the given 4x4 matrix:\n\n\\[\n\\begin{bmatrix}\n12 & -16 & 4 & "
+{"step": 131, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.06523915852874827, "tokens": 120000, "cumulative_loss_tokens": 15720000, "grad_norm": 0.41796875, "lr": 3e-05, "finish_rate": 0.868, "comp_len": 452.8, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 53.8, "frames": {"chat": 265}, "mem_gb": 22.1}
+{"step": 132, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.1503767799893394, "tokens": 120000, "cumulative_loss_tokens": 15840000, "grad_norm": 0.546875, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 356.1, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 66.1, "frames": {"chat": 337}, "mem_gb": 21.65}
+{"step": 133, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.07360061740261348, "tokens": 120000, "cumulative_loss_tokens": 15960000, "grad_norm": 0.419921875, "lr": 3e-05, "finish_rate": 0.972, "comp_len": 413.8, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 58.9, "frames": {"chat": 290}, "mem_gb": 21.91}
+{"step": 134, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.12301088459033054, "tokens": 120000, "cumulative_loss_tokens": 16080000, "grad_norm": 0.50390625, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 343.8, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 66.8, "frames": {"chat": 349}, "mem_gb": 21.84}
+{"step": 135, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.07363978966639066, "tokens": 120000, "cumulative_loss_tokens": 16200000, "grad_norm": 0.404296875, "lr": 3e-05, "finish_rate": 0.928, "comp_len": 412.4, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 57.8, "frames": {"chat": 291}, "mem_gb": 21.94}
+{"step": 136, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.12913179965493424, "tokens": 120000, "cumulative_loss_tokens": 16320000, "grad_norm": 0.515625, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 350.9, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 66.5, "frames": {"chat": 342}, "mem_gb": 21.8}
+{"step": 137, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.17013440595478752, "tokens": 120000, "cumulative_loss_tokens": 16440000, "grad_norm": 0.55078125, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 355.0, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 64.0, "frames": {"chat": 338}, "mem_gb": 21.75}
+{"step": 138, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.14697195186861792, "tokens": 120000, "cumulative_loss_tokens": 16560000, "grad_norm": 0.5234375, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 337.1, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 70.2, "frames": {"chat": 356}, "mem_gb": 21.71}
+{"step": 139, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.06590260987370275, "tokens": 120000, "cumulative_loss_tokens": 16680000, "grad_norm": 0.39453125, "lr": 3e-05, "finish_rate": 0.86, "comp_len": 452.8, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 59.6, "frames": {"chat": 265}, "mem_gb": 22.09}
+{"step": 140, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.11447063266391246, "tokens": 120000, "cumulative_loss_tokens": 16800000, "grad_norm": 0.498046875, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 339.9, "t_data_s": 0.1, "t_rollout_s": 0.0, "t_step_s": 68.9, "frames": {"chat": 353}, "mem_gb": 21.81}
+[eval step 140] sample: 'To find the rank of a matrix, we need to determine the maximum number of linearly independent rows or columns in the matrix. \n\nGiven the matrix:\n\\[\n\\begin{bmatrix}\n12 & -16 & 4 & 16 \\\\\n-9 & 11 & -1 & '
+{"step": 141, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.11048259019283578, "tokens": 120000, "cumulative_loss_tokens": 16920000, "grad_norm": 0.498046875, "lr": 3e-05, "finish_rate": 0.949, "comp_len": 379.7, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 62.9, "frames": {"chat": 316}, "mem_gb": 21.99}
+{"step": 142, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.09986399798836404, "tokens": 120000, "cumulative_loss_tokens": 17040000, "grad_norm": 0.455078125, "lr": 3e-05, "finish_rate": 0.945, "comp_len": 415.2, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 60.7, "frames": {"chat": 289}, "mem_gb": 22.14}
+{"step": 143, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.0992997664115702, "tokens": 120000, "cumulative_loss_tokens": 17160000, "grad_norm": 0.46484375, "lr": 3e-05, "finish_rate": 0.982, "comp_len": 362.5, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 65.8, "frames": {"chat": 331}, "mem_gb": 21.78}
+{"step": 144, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.09031849862008046, "tokens": 120000, "cumulative_loss_tokens": 17280000, "grad_norm": 0.42578125, "lr": 3e-05, "finish_rate": 0.972, "comp_len": 375.0, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 61.6, "frames": {"chat": 320}, "mem_gb": 22.1}
+{"step": 145, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.08688670156883696, "tokens": 120000, "cumulative_loss_tokens": 17400000, "grad_norm": 0.380859375, "lr": 3e-05, "finish_rate": 0.941, "comp_len": 441.2, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 53.3, "frames": {"chat": 272}, "mem_gb": 21.83}
+{"step": 146, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.17230489805215815, "tokens": 120000, "cumulative_loss_tokens": 17520000, "grad_norm": 0.56640625, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 326.1, "t_data_s": 0.1, "t_rollout_s": 0.0, "t_step_s": 71.0, "frames": {"chat": 368}, "mem_gb": 21.75}
+{"step": 147, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.12812672173778217, "tokens": 120000, "cumulative_loss_tokens": 17640000, "grad_norm": 0.478515625, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 365.9, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 65.5, "frames": {"chat": 328}, "mem_gb": 21.78}
+{"step": 148, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.12114051378740308, "tokens": 120000, "cumulative_loss_tokens": 17760000, "grad_norm": 0.51953125, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 368.1, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 64.5, "frames": {"chat": 326}, "mem_gb": 21.99}
+{"step": 149, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.1461482815605433, "tokens": 120000, "cumulative_loss_tokens": 17880000, "grad_norm": 0.5, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 329.7, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 72.8, "frames": {"chat": 364}, "mem_gb": 21.93}
+{"step": 150, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.09341100072071422, "tokens": 120000, "cumulative_loss_tokens": 18000000, "grad_norm": 0.412109375, "lr": 3e-05, "finish_rate": 0.97, "comp_len": 405.4, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 60.8, "frames": {"chat": 296}, "mem_gb": 21.98}
+[eval step 150] sample: 'To compute the rank of a matrix, we need to determine the maximum number of linearly independent rows or columns in the matrix. \n\nGiven the matrix:\n\\[\n\\begin{bmatrix}\n12 & -16 & 4 & 16 \\\\\n-9 & 11 & -1'
+{"step": 151, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.13002619124348275, "tokens": 120000, "cumulative_loss_tokens": 18120000, "grad_norm": 0.5078125, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 317.5, "t_data_s": 0.1, "t_rollout_s": 0.0, "t_step_s": 73.4, "frames": {"chat": 378}, "mem_gb": 21.83}
+{"step": 152, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.1715810147296327, "tokens": 120000, "cumulative_loss_tokens": 18240000, "grad_norm": 0.5703125, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 326.1, "t_data_s": 0.1, "t_rollout_s": 0.0, "t_step_s": 72.4, "frames": {"chat": 368}, "mem_gb": 21.93}
+{"step": 153, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.07034443499888682, "tokens": 120000, "cumulative_loss_tokens": 18360000, "grad_norm": 0.390625, "lr": 3e-05, "finish_rate": 0.898, "comp_len": 468.8, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 56.5, "frames": {"chat": 256}, "mem_gb": 22.03}
+{"step": 154, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.08062464235329535, "tokens": 120000, "cumulative_loss_tokens": 18480000, "grad_norm": 0.423828125, "lr": 3e-05, "finish_rate": 0.93, "comp_len": 381.0, "t_data_s": 0.1, "t_rollout_s": 0.0, "t_step_s": 62.1, "frames": {"chat": 315}, "mem_gb": 21.98}
+{"step": 155, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.07058507993586827, "tokens": 120000, "cumulative_loss_tokens": 18600000, "grad_norm": 0.384765625, "lr": 3e-05, "finish_rate": 0.974, "comp_len": 384.6, "t_data_s": 0.1, "t_rollout_s": 0.0, "t_step_s": 62.6, "frames": {"chat": 312}, "mem_gb": 21.91}
+{"step": 156, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.11778950736102803, "tokens": 120000, "cumulative_loss_tokens": 18720000, "grad_norm": 0.462890625, "lr": 3e-05, "finish_rate": 0.97, "comp_len": 363.6, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 65.6, "frames": {"chat": 330}, "mem_gb": 22.09}
+{"step": 157, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.09653343385270952, "tokens": 120000, "cumulative_loss_tokens": 18840000, "grad_norm": 0.435546875, "lr": 3e-05, "finish_rate": 0.954, "comp_len": 367.0, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 67.3, "frames": {"chat": 327}, "mem_gb": 22.09}
+{"step": 158, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.13204492282280697, "tokens": 120000, "cumulative_loss_tokens": 18960000, "grad_norm": 0.48046875, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 341.9, "t_data_s": 0.1, "t_rollout_s": 0.0, "t_step_s": 72.3, "frames": {"chat": 351}, "mem_gb": 21.75}
+{"step": 159, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.07468217376978913, "tokens": 120000, "cumulative_loss_tokens": 19080000, "grad_norm": 0.396484375, "lr": 3e-05, "finish_rate": 0.973, "comp_len": 357.1, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 66.4, "frames": {"chat": 336}, "mem_gb": 21.8}
+{"step": 160, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.13069434770564547, "tokens": 120000, "cumulative_loss_tokens": 19200000, "grad_norm": 0.51171875, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 332.4, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 74.0, "frames": {"chat": 361}, "mem_gb": 21.78}
+[eval step 160] sample: 'To compute the rank of a matrix, we need to determine the maximum number of linearly independent rows or columns in the matrix. A matrix is rank-deficient if it has fewer linearly independent rows or '
+{"step": 161, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.14887345913428193, "tokens": 120000, "cumulative_loss_tokens": 19320000, "grad_norm": 0.486328125, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 328.8, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 75.2, "frames": {"chat": 365}, "mem_gb": 21.72}
+{"step": 162, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.13350603818915163, "tokens": 120000, "cumulative_loss_tokens": 19440000, "grad_norm": 0.486328125, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 364.7, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 64.7, "frames": {"chat": 329}, "mem_gb": 21.78}
+{"step": 163, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.09072161397730621, "tokens": 120000, "cumulative_loss_tokens": 19560000, "grad_norm": 0.42578125, "lr": 3e-05, "finish_rate": 0.931, "comp_len": 416.7, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 59.5, "frames": {"chat": 288}, "mem_gb": 21.99}
+{"step": 164, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.06492729764790274, "tokens": 120000, "cumulative_loss_tokens": 19680000, "grad_norm": 0.41796875, "lr": 3e-05, "finish_rate": 0.868, "comp_len": 451.1, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 59.0, "frames": {"chat": 266}, "mem_gb": 22.14}
+{"step": 165, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.11315758886872791, "tokens": 120000, "cumulative_loss_tokens": 19800000, "grad_norm": 0.4921875, "lr": 3e-05, "finish_rate": 0.988, "comp_len": 372.7, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 58.5, "frames": {"chat": 322}, "mem_gb": 22.05}
+{"step": 166, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.1163916651834889, "tokens": 120000, "cumulative_loss_tokens": 19920000, "grad_norm": 0.47265625, "lr": 3e-05, "finish_rate": 0.989, "comp_len": 334.3, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 62.8, "frames": {"chat": 359}, "mem_gb": 21.75}
+{"step": 167, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.09991620500825811, "tokens": 120000, "cumulative_loss_tokens": 20040000, "grad_norm": 0.447265625, "lr": 3e-05, "finish_rate": 0.965, "comp_len": 385.9, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 58.9, "frames": {"chat": 311}, "mem_gb": 21.88}
+{"step": 168, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.08487513831759182, "tokens": 120000, "cumulative_loss_tokens": 20160000, "grad_norm": 0.408203125, "lr": 3e-05, "finish_rate": 0.95, "comp_len": 396.0, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 55.4, "frames": {"chat": 303}, "mem_gb": 21.93}
+{"step": 169, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.15273571721582363, "tokens": 120000, "cumulative_loss_tokens": 20280000, "grad_norm": 0.52734375, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 351.9, "t_data_s": 0.1, "t_rollout_s": 0.0, "t_step_s": 68.0, "frames": {"chat": 341}, "mem_gb": 21.94}
+{"step": 170, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.0354515667994041, "tokens": 120000, "cumulative_loss_tokens": 20400000, "grad_norm": 0.357421875, "lr": 3e-05, "finish_rate": 0.843, "comp_len": 524.0, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 47.7, "frames": {"chat": 229}, "mem_gb": 22.0}
+[eval step 170] sample: 'To compute the rank of a matrix, we need to determine the maximum number of linearly independent rows or columns in the matrix. \n\nGiven the matrix:\n\\[\n\\begin{bmatrix}\n12 & -16 & 4 & 16 \\\\\n-9 & 11 & -1'
+{"step": 171, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.0888629938938267, "tokens": 120000, "cumulative_loss_tokens": 20520000, "grad_norm": 0.423828125, "lr": 3e-05, "finish_rate": 0.964, "comp_len": 389.6, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 60.9, "frames": {"chat": 308}, "mem_gb": 22.16}
+{"step": 172, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.1270647643733304, "tokens": 120000, "cumulative_loss_tokens": 20640000, "grad_norm": 0.5078125, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 331.5, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 66.9, "frames": {"chat": 362}, "mem_gb": 21.65}
+{"step": 173, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.1342932996113319, "tokens": 120000, "cumulative_loss_tokens": 20760000, "grad_norm": 0.5234375, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 315.8, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 73.5, "frames": {"chat": 380}, "mem_gb": 21.9}
+{"step": 174, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.12381919350024934, "tokens": 120000, "cumulative_loss_tokens": 20880000, "grad_norm": 0.51953125, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 336.1, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 66.0, "frames": {"chat": 357}, "mem_gb": 21.68}
+{"step": 175, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.06805174825843424, "tokens": 120000, "cumulative_loss_tokens": 21000000, "grad_norm": 0.40625, "lr": 3e-05, "finish_rate": 0.835, "comp_len": 472.4, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 52.3, "frames": {"chat": 254}, "mem_gb": 22.16}
+{"step": 176, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.05205729084465808, "tokens": 120000, "cumulative_loss_tokens": 21120000, "grad_norm": 0.34765625, "lr": 3e-05, "finish_rate": 0.874, "comp_len": 459.8, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 52.0, "frames": {"chat": 261}, "mem_gb": 22.12}
+{"step": 177, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.1126549622541992, "tokens": 120000, "cumulative_loss_tokens": 21240000, "grad_norm": 0.4765625, "lr": 3e-05, "finish_rate": 0.982, "comp_len": 431.7, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 54.5, "frames": {"chat": 278}, "mem_gb": 21.91}
+{"step": 178, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.12549313363223336, "tokens": 120000, "cumulative_loss_tokens": 21360000, "grad_norm": 0.474609375, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 360.4, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 63.5, "frames": {"chat": 333}, "mem_gb": 22.05}
+{"step": 179, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.066560423706177, "tokens": 120000, "cumulative_loss_tokens": 21480000, "grad_norm": 0.447265625, "lr": 3e-05, "finish_rate": 0.888, "comp_len": 480.0, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 51.7, "frames": {"chat": 250}, "mem_gb": 22.02}
+{"step": 180, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.1541350697507616, "tokens": 120000, "cumulative_loss_tokens": 21600000, "grad_norm": 0.55078125, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 336.1, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 66.5, "frames": {"chat": 357}, "mem_gb": 21.79}
+[eval step 180] sample: 'To find the rank of a matrix, we need to determine the maximum number of linearly independent rows or columns in the matrix. A matrix is said to be of rank \\( r \\) if it has \\( r \\) linearly independe'
+{"step": 181, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.1540505784819601, "tokens": 120000, "cumulative_loss_tokens": 21720000, "grad_norm": 0.5390625, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 363.6, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 61.7, "frames": {"chat": 330}, "mem_gb": 21.76}
+{"step": 182, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.12838531766707698, "tokens": 120000, "cumulative_loss_tokens": 21840000, "grad_norm": 0.51171875, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 341.9, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 63.0, "frames": {"chat": 351}, "mem_gb": 21.8}
+{"step": 183, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.08405779912746511, "tokens": 120000, "cumulative_loss_tokens": 21960000, "grad_norm": 0.455078125, "lr": 3e-05, "finish_rate": 0.959, "comp_len": 413.8, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 58.7, "frames": {"chat": 290}, "mem_gb": 21.93}
+{"step": 184, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.1232834550240698, "tokens": 120000, "cumulative_loss_tokens": 22080000, "grad_norm": 0.486328125, "lr": 3e-05, "finish_rate": 0.969, "comp_len": 372.7, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 65.7, "frames": {"chat": 322}, "mem_gb": 21.87}
+{"step": 185, "epoch": 0, "training_mode": "off-policy", "forward_topk_kl": 0.056777896666216354, "tokens": 120000, "cumulative_loss_tokens": 22200000, "grad_norm": 0.37890625, "lr": 3e-05, "finish_rate": 0.827, "comp_len": 451.1, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 61.1, "frames": {"chat": 266}, "mem_gb": 22.1}
+{"step": 186, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.09375398674480773, "tokens": 120000, "cumulative_loss_tokens": 22320000, "grad_norm": 0.46875, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 362.5, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 68.1, "frames": {"chat": 331}, "mem_gb": 21.96}
+{"step": 187, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.08996357367694534, "tokens": 120000, "cumulative_loss_tokens": 22440000, "grad_norm": 0.4453125, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 350.9, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 68.9, "frames": {"chat": 342}, "mem_gb": 21.72}
+{"step": 188, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.09844830975172421, "tokens": 120000, "cumulative_loss_tokens": 22560000, "grad_norm": 0.55078125, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 335.2, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 70.0, "frames": {"chat": 358}, "mem_gb": 21.75}
+{"step": 189, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.0910161497346436, "tokens": 120000, "cumulative_loss_tokens": 22680000, "grad_norm": 0.443359375, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 348.8, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 72.4, "frames": {"chat": 344}, "mem_gb": 21.84}
+{"step": 190, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.04903289433984707, "tokens": 120000, "cumulative_loss_tokens": 22800000, "grad_norm": 0.375, "lr": 3e-05, "finish_rate": 0.861, "comp_len": 449.4, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 61.5, "frames": {"chat": 267}, "mem_gb": 22.15}
+[eval step 190] sample: 'To find the rank of a matrix, we need to determine the maximum number of linearly independent rows or columns. A matrix is said to be of rank \\( r \\) if it has \\( r \\) linearly independent rows or col'
+{"step": 191, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.02093672791371743, "tokens": 120000, "cumulative_loss_tokens": 22920000, "grad_norm": 0.216796875, "lr": 3e-05, "finish_rate": 0.835, "comp_len": 483.9, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 54.5, "frames": {"chat": 248}, "mem_gb": 21.99}
+{"step": 192, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.07855782948022631, "tokens": 120000, "cumulative_loss_tokens": 23040000, "grad_norm": 0.42578125, "lr": 3e-05, "finish_rate": 0.979, "comp_len": 364.7, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 66.9, "frames": {"chat": 329}, "mem_gb": 21.98}
+{"step": 193, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.08327089551646884, "tokens": 120000, "cumulative_loss_tokens": 23160000, "grad_norm": 0.41796875, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 354.0, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 70.5, "frames": {"chat": 339}, "mem_gb": 21.94}
+{"step": 194, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.09215007556722654, "tokens": 120000, "cumulative_loss_tokens": 23280000, "grad_norm": 0.419921875, "lr": 3e-05, "finish_rate": 0.997, "comp_len": 339.0, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 71.8, "frames": {"chat": 354}, "mem_gb": 21.94}
+{"step": 195, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.05571042603639653, "tokens": 120000, "cumulative_loss_tokens": 23400000, "grad_norm": 0.3359375, "lr": 3e-05, "finish_rate": 0.943, "comp_len": 430.1, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 58.1, "frames": {"chat": 279}, "mem_gb": 22.07}
+{"step": 196, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.06696861219073956, "tokens": 120000, "cumulative_loss_tokens": 23520000, "grad_norm": 0.439453125, "lr": 3e-05, "finish_rate": 0.959, "comp_len": 377.4, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 66.2, "frames": {"chat": 318}, "mem_gb": 22.0}
+{"step": 197, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.048480764170961144, "tokens": 120000, "cumulative_loss_tokens": 23640000, "grad_norm": 0.345703125, "lr": 3e-05, "finish_rate": 0.875, "comp_len": 452.8, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 58.6, "frames": {"chat": 265}, "mem_gb": 22.09}
+{"step": 198, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.043221544861785754, "tokens": 120000, "cumulative_loss_tokens": 23760000, "grad_norm": 0.30078125, "lr": 3e-05, "finish_rate": 0.85, "comp_len": 472.4, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 55.3, "frames": {"chat": 254}, "mem_gb": 22.03}
+{"step": 199, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.06468877917430751, "tokens": 120000, "cumulative_loss_tokens": 23880000, "grad_norm": 0.384765625, "lr": 3e-05, "finish_rate": 0.962, "comp_len": 381.0, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 66.1, "frames": {"chat": 315}, "mem_gb": 22.06}
+{"step": 200, "epoch": 1, "training_mode": "off-policy", "forward_topk_kl": 0.08913087736829184, "tokens": 120000, "cumulative_loss_tokens": 24000000, "grad_norm": 0.435546875, "lr": 3e-05, "finish_rate": 0.977, "comp_len": 400.0, "t_data_s": 0.0, "t_rollout_s": 0.0, "t_step_s": 61.7, "frames": {"chat": 300}, "mem_gb": 21.85}
+[eval step 200] sample: "To find the rank of a matrix, we need to determine the maximum number of linearly independent rows or columns. \n\nLet's start by examining the given 4x4 matrix:\n\n\\[\n\\begin{bmatrix}\n12 & -16 & 4 & 16 \\\\"
+checkpoint snapshot queued -> outputs/healed/calib/glean_keep75_s1224/step0200

healed/calib/glean_keep75_s1224/wandb/run-20260717_213714-mkqm7no2/files/requirements.txt

@@ -0,0 +1,129 @@
+certifi==2026.6.17
+fonttools==4.63.0
+urllib3==2.7.0
+requests==2.34.2
+nvidia-cublas-cu12==12.6.4.1
+packaging==26.2
+regex==2026.6.28
+portalocker==3.2.0
+safetensors==0.8.0
+datasets==5.0.0
+sqlitedict==2.1.0
+narwhals==2.23.0
+latex2sympy2_extended==1.11.0
+smmap==5.0.3
+colorama==0.4.6
+anyio==4.14.1
+cycler==0.12.1
+pytablewriter==1.2.1
+pydantic_core==2.46.4
+huggingface_hub==0.36.2
+charset-normalizer==3.4.8
+nvidia-cusolver-cu12==11.7.1.2
+joblib==1.5.3
+typing-inspection==0.4.2
+wandb==0.28.0
+bitsandbytes==0.49.2
+kiwisolver==1.5.0
+cuda-toolkit==12.6.3
+propcache==0.5.2
+Jinja2==3.1.6
+cuda-bindings==12.9.7
+pluggy==1.6.0
+pytest==9.1.1
+fsspec==2026.4.0
+tqdm==4.68.3
+psutil==7.2.2
+pydantic==2.13.4
+typepy==1.3.5
+xxhash==3.8.1
+DataProperty==1.1.1
+aiosignal==1.4.0
+annotated-types==0.7.0
+Pygments==2.20.0
+zstandard==0.25.0
+threadpoolctl==3.6.0
+nvidia-cudnn-cu12==9.10.2.21
+evaluate==0.4.6
+pillow==12.3.0
+GitPython==3.1.50
+httpx==0.28.1
+lm_eval==0.4.12
+pytz==2026.2
+mbstrdecoder==1.1.5
+immutabledict==4.3.1
+dill==0.4.1
+mpmath==1.3.0
+yarl==1.24.2
+word2number==1.1
+networkx==3.6.1
+einops==0.8.2
+tabulate==0.10.0
+pathvalidate==3.3.1
+tenacity==9.1.4
+sentry-sdk==2.64.0
+nvidia-nvshmem-cu12==3.4.5
+hf-xet==1.5.1
+six==1.17.0
+pyarrow==24.0.0
+nvidia-curand-cu12==10.3.7.77
+cuda-pathfinder==1.5.6
+sacrebleu==2.6.0
+triton==3.7.1
+frozenlist==1.8.0
+antlr4-python3-runtime==4.11.0
+tabledata==1.3.5
+h11==0.16.0
+nvidia-cusparselt-cu12==0.7.1
+nvidia-nvjitlink-cu12==12.6.85
+scipy==1.18.0
+rouge_score==0.1.2
+transformers==4.57.6
+multiprocess==0.70.19
+absl-py==2.5.0
+accelerate==1.14.0
+click==8.4.2
+lxml==6.1.1
+defusedxml==0.7.1
+setuptools==81.0.0
+nvidia-cuda-cupti-cu12==12.6.80
+httpcore==1.0.9
+langdetect==1.0.9
+nltk==3.10.0
+iniconfig==2.3.0
+nvidia-cuda-runtime-cu12==12.6.77
+nvidia-cusparse-cu12==12.5.4.2
+matplotlib==3.11.0
+protobuf==7.35.1
+nvidia-cufft-cu12==11.3.0.4
+pyparsing==3.3.2
+chardet==6.0.0.post1
+typing_extensions==4.16.0
+stanford-stk==0.7.1
+aiohappyeyeballs==2.7.1
+MarkupSafe==3.0.3
+nvidia-cufile-cu12==1.11.1.6
+tcolorpy==0.1.7
+scikit-learn==1.9.0
+tokenizers==0.22.2
+attrs==26.1.0
+gitdb==4.0.12
+aiohttp==3.14.1
+glean==0.0.1
+platformdirs==4.10.0
+megablocks==0.11.0.dev0
+nvidia-cuda-nvrtc-cu12==12.6.85
+pandas==3.0.3
+multidict==6.7.1
+torch==2.12.1+cu126
+contourpy==1.3.3
+nvidia-nvtx-cu12==12.6.77
+math-verify==0.9.0
+filelock==3.29.5
+idna==3.18
+nvidia-nccl-cu12==2.29.3
+python-dateutil==2.9.0.post0
+more-itertools==11.1.0
+numpy==2.0.2
+sympy==1.14.0
+PyYAML==6.0.3

healed/calib/glean_keep75_s1224/wandb/run-20260717_213714-mkqm7no2/files/wandb-metadata.json

@@ -0,0 +1,97 @@
+{
+  "os": "Linux-7.0.11-76070011-generic-x86_64-with-glibc2.35",
+  "python": "CPython 3.12.12",
+  "startedAt": "2026-07-18T04:37:14.633800Z",
+  "args": [
+    "--training-mode",
+    "off-policy",
+    "--kl-direction",
+    "forward",
+    "--topk-targets",
+    "outputs/teacher_trajectories/dolci_combined_top128",
+    "--off-policy-frames",
+    "chat",
+    "--student-device",
+    "cuda:0",
+    "--student",
+    "outputs/pruned/glean-0125inst-general-keep75",
+    "--resume-from",
+    "outputs/healed/calib/glean_keep75_s1224/step0100",
+    "--lr",
+    "3e-5",
+    "--optimizer",
+    "adamw8bit",
+    "--epochs",
+    "9",
+    "--sweep",
+    "200",
+    "--micro-batch",
+    "3",
+    "--loss-tokens-per-step",
+    "120000",
+    "--gsm8k-every",
+    "0",
+    "--save-every",
+    "200",
+    "--seed",
+    "1224",
+    "--out-dir",
+    "outputs/healed/calib/glean_keep75_s1224",
+    "--wandb",
+    "--wandb-mode",
+    "online",
+    "--wandb-project",
+    "glean-calib",
+    "--wandb-run-name",
+    "calib_keep75_200"
+  ],
+  "program": "/home/henry/Documents/PythonProjects/variable-reap/scripts/11_distill_on_policy.py",
+  "codePath": "scripts/11_distill_on_policy.py",
+  "codePathLocal": "scripts/11_distill_on_policy.py",
+  "git": {
+    "remote": "https://github.com/hbfreed/variable-reap.git",
+    "commit": "247c7f0113d7a058b1db8044deddfd4456cc998d"
+  },
+  "email": "hbfreed@protonmail.com",
+  "root": "outputs/healed/calib/glean_keep75_s1224",
+  "host": "pop-os",
+  "executable": "/home/henry/Documents/PythonProjects/variable-reap/.venv/bin/python3",
+  "cpu_count": 24,
+  "cpu_count_logical": 48,
+  "gpu": "NVIDIA GeForce RTX 3090",
+  "gpu_count": 3,
+  "disk": {
+    "/": {
+      "total": "1958315118592",
+      "used": "1668944318464"
+    }
+  },
+  "memory": {
+    "total": "134900756480"
+  },
+  "gpu_nvidia": [
+    {
+      "name": "NVIDIA GeForce RTX 3090",
+      "memoryTotal": "25769803776",
+      "cudaCores": 10496,
+      "architecture": "Ampere",
+      "uuid": "GPU-8ca70870-ddf2-d274-bcc5-182c2075bced"
+    },
+    {
+      "name": "NVIDIA GeForce RTX 3090",
+      "memoryTotal": "25769803776",
+      "cudaCores": 10496,
+      "architecture": "Ampere",
+      "uuid": "GPU-a6acf07f-31f5-618f-a5d0-c0017e7e2e27"
+    },
+    {
+      "name": "NVIDIA GeForce RTX 3090",
+      "memoryTotal": "25769803776",
+      "cudaCores": 10496,
+      "architecture": "Ampere",
+      "uuid": "GPU-864c54df-0130-7780-e271-8a5551d1733f"
+    }
+  ],
+  "cudaVersion": "13.0",
+  "writerId": "0c0c5qhc7ujrqprydps7c2ecqc1afsdo"
+}

healed/calib/glean_keep75_s1224/wandb/run-20260717_213714-mkqm7no2/files/wandb-summary.json

@@ -0,0 +1 @@
+{"mem_gb":21.85,"comp_len":400,"tokens":120000,"grad_norm":0.435546875,"frames":{"chat":300},"step":200,"t_data_s":0,"_wandb":{"runtime":6327},"_timestamp":1.784355730778899e+09,"cumulative_loss_tokens":24000000,"epoch":1,"training_mode":"off-policy","t_step_s":61.7,"forward_topk_kl":0.08913087736829184,"t_rollout_s":0,"_step":99,"finish_rate":0.977,"lr":3e-05,"_runtime":6327}