id stringlengths 12 32 | archetype stringclasses 30
values | split stringclasses 1
value | prompt stringlengths 4.71k 5.31k | gold stringlengths 68 186 | meta stringclasses 30
values |
|---|---|---|---|---|---|
train-lr_at_step-017 | lr_at_step | train | You are debugging a failed or misbehaving LLM training run.
Pick exactly ONE diagnosis code and give the minimal fix: only the config fields that must change, with corrected values.
Diagnosis codes:
- OOM_BATCH_TOO_LARGE
- ACTIVATION_MEMORY_NO_CHECKPOINTING
- EFFECTIVE_BATCH_MISMATCH
- LR_TOO_HIGH_NO_WARMUP
- PRECISIO... | {"diagnosis": "LR_VALUE_AT_STEP", "fix": {"current_learning_rate": 0.00011}, "work": {"step_fraction": 0.5}} | {"archetype": "lr_at_step", "computations": ["R15"], "depth": 2, "interaction": null, "tier": "computed"} |
train-lr_at_step-062 | lr_at_step | train | You are debugging a failed or misbehaving LLM training run.
Pick exactly ONE diagnosis code and give the minimal fix: only the config fields that must change, with corrected values.
Diagnosis codes:
- OOM_BATCH_TOO_LARGE
- ACTIVATION_MEMORY_NO_CHECKPOINTING
- EFFECTIVE_BATCH_MISMATCH
- LR_TOO_HIGH_NO_WARMUP
- PRECISIO... | {"diagnosis": "LR_VALUE_AT_STEP", "fix": {"current_learning_rate": 0.00011}, "work": {"step_fraction": 0.5}} | {"archetype": "lr_at_step", "computations": ["R15"], "depth": 2, "interaction": null, "tier": "computed"} |
train-two_oom_clipping-037 | two_oom_clipping | train | You are debugging a failed or misbehaving LLM training run.
Pick exactly ONE diagnosis code and give the minimal fix: only the config fields that must change, with corrected values.
Diagnosis codes:
- OOM_BATCH_TOO_LARGE
- ACTIVATION_MEMORY_NO_CHECKPOINTING
- EFFECTIVE_BATCH_MISMATCH
- LR_TOO_HIGH_NO_WARMUP
- PRECISIO... | {"diagnosis": ["OOM_BATCH_TOO_LARGE", "NO_GRADIENT_CLIPPING"], "fix": {"max_grad_norm": 1.0, "per_device_batch_size": 33}} | {"archetype": "two_oom_clipping", "computations": ["R1", "R12"], "depth": 3, "interaction": "oom+clipping", "tier": "two_fault"} |
train-shard_size-000 | shard_size | train | You are debugging a failed or misbehaving LLM training run.
Pick exactly ONE diagnosis code and give the minimal fix: only the config fields that must change, with corrected values.
Diagnosis codes:
- OOM_BATCH_TOO_LARGE
- ACTIVATION_MEMORY_NO_CHECKPOINTING
- EFFECTIVE_BATCH_MISMATCH
- LR_TOO_HIGH_NO_WARMUP
- PRECISIO... | {"diagnosis": "SHARD_SIZE_PER_RANK", "fix": {"examples_per_rank": 25000, "shard_size_gb": 60.0}} | {"archetype": "shard_size", "computations": ["R18"], "depth": 2, "interaction": null, "tier": "computed"} |
train-ckpt_batch_accum-069 | ckpt_batch_accum | train | You are debugging a failed or misbehaving LLM training run.
Pick exactly ONE diagnosis code and give the minimal fix: only the config fields that must change, with corrected values.
Diagnosis codes:
- OOM_BATCH_TOO_LARGE
- ACTIVATION_MEMORY_NO_CHECKPOINTING
- EFFECTIVE_BATCH_MISMATCH
- LR_TOO_HIGH_NO_WARMUP
- PRECISIO... | {"diagnosis": "CHECKPOINTED_BATCH_AND_ACCUM", "fix": {"grad_accum_steps": 2, "per_device_batch_size": 66}, "work": {"activation_per_sample_gb": 1.0, "max_per_device_batch": 66}} | {"archetype": "ckpt_batch_accum", "computations": ["R20", "R1", "R3"], "depth": 4, "interaction": "checkpointing x accum", "tier": "interacting"} |
train-parallel_batch-005 | parallel_batch | train | You are debugging a failed or misbehaving LLM training run.
Pick exactly ONE diagnosis code and give the minimal fix: only the config fields that must change, with corrected values.
Diagnosis codes:
- OOM_BATCH_TOO_LARGE
- ACTIVATION_MEMORY_NO_CHECKPOINTING
- EFFECTIVE_BATCH_MISMATCH
- LR_TOO_HIGH_NO_WARMUP
- PRECISIO... | {"diagnosis": "PARALLEL_SHARDED_BATCH", "fix": {"per_device_batch_size": 32}, "work": {"fixed_memory_gb": 128.0, "params_per_rank_billions": 8.0}} | {"archetype": "parallel_batch", "computations": ["R22", "R13", "R1"], "depth": 4, "interaction": "parallelism x memory", "tier": "interacting"} |
train-lr_at_step-019 | lr_at_step | train | You are debugging a failed or misbehaving LLM training run.
Pick exactly ONE diagnosis code and give the minimal fix: only the config fields that must change, with corrected values.
Diagnosis codes:
- OOM_BATCH_TOO_LARGE
- ACTIVATION_MEMORY_NO_CHECKPOINTING
- EFFECTIVE_BATCH_MISMATCH
- LR_TOO_HIGH_NO_WARMUP
- PRECISIO... | {"diagnosis": "LR_VALUE_AT_STEP", "fix": {"current_learning_rate": 0.0002325}, "work": {"step_fraction": 0.25}} | {"archetype": "lr_at_step", "computations": ["R15"], "depth": 2, "interaction": null, "tier": "computed"} |
train-packing_time-053 | packing_time | train | You are debugging a failed or misbehaving LLM training run.
Pick exactly ONE diagnosis code and give the minimal fix: only the config fields that must change, with corrected values.
Diagnosis codes:
- OOM_BATCH_TOO_LARGE
- ACTIVATION_MEMORY_NO_CHECKPOINTING
- EFFECTIVE_BATCH_MISMATCH
- LR_TOO_HIGH_NO_WARMUP
- PRECISIO... | {"diagnosis": "PACKED_TRAINING_TIME", "fix": {"planned_time_hours": 1.0}, "work": {"total_padded_tokens": 245760000, "total_real_tokens": 216000000}} | {"archetype": "packing_time", "computations": ["R23", "R16"], "depth": 4, "interaction": "packing x time", "tier": "interacting"} |
train-batch_from_components-041 | batch_from_components | train | You are debugging a failed or misbehaving LLM training run.
Pick exactly ONE diagnosis code and give the minimal fix: only the config fields that must change, with corrected values.
Diagnosis codes:
- OOM_BATCH_TOO_LARGE
- ACTIVATION_MEMORY_NO_CHECKPOINTING
- EFFECTIVE_BATCH_MISMATCH
- LR_TOO_HIGH_NO_WARMUP
- PRECISIO... | {"diagnosis": "BATCH_FROM_MEMORY_COMPONENTS", "fix": {"per_device_batch_size": 24}, "work": {"fixed_memory_gb": 32.0, "remaining_memory_gb": 48.0}} | {"archetype": "batch_from_components", "computations": ["R13", "R1"], "depth": 3, "interaction": null, "tier": "computed"} |
train-ckpt_batch_accum-039 | ckpt_batch_accum | train | You are debugging a failed or misbehaving LLM training run.
Pick exactly ONE diagnosis code and give the minimal fix: only the config fields that must change, with corrected values.
Diagnosis codes:
- OOM_BATCH_TOO_LARGE
- ACTIVATION_MEMORY_NO_CHECKPOINTING
- EFFECTIVE_BATCH_MISMATCH
- LR_TOO_HIGH_NO_WARMUP
- PRECISIO... | {"diagnosis": "CHECKPOINTED_BATCH_AND_ACCUM", "fix": {"grad_accum_steps": 2, "per_device_batch_size": 13}, "work": {"activation_per_sample_gb": 2.0, "max_per_device_batch": 13}} | {"archetype": "ckpt_batch_accum", "computations": ["R20", "R1", "R3"], "depth": 4, "interaction": "checkpointing x accum", "tier": "interacting"} |
train-zero_batch-001 | zero_batch | train | You are debugging a failed or misbehaving LLM training run.
Pick exactly ONE diagnosis code and give the minimal fix: only the config fields that must change, with corrected values.
Diagnosis codes:
- OOM_BATCH_TOO_LARGE
- ACTIVATION_MEMORY_NO_CHECKPOINTING
- EFFECTIVE_BATCH_MISMATCH
- LR_TOO_HIGH_NO_WARMUP
- PRECISIO... | {"diagnosis": "ZERO_SHARDED_BATCH_SIZE", "fix": {"per_device_batch_size": 40}, "work": {"bytes_per_param_per_rank": 10.0, "fixed_memory_gb": 40.0, "remaining_memory_gb": 40.0}} | {"archetype": "zero_batch", "computations": ["R19", "R13", "R1"], "depth": 4, "interaction": "zero x memory", "tier": "interacting"} |
train-ckpt_batch_accum-014 | ckpt_batch_accum | train | You are debugging a failed or misbehaving LLM training run.
Pick exactly ONE diagnosis code and give the minimal fix: only the config fields that must change, with corrected values.
Diagnosis codes:
- OOM_BATCH_TOO_LARGE
- ACTIVATION_MEMORY_NO_CHECKPOINTING
- EFFECTIVE_BATCH_MISMATCH
- LR_TOO_HIGH_NO_WARMUP
- PRECISIO... | {"diagnosis": "CHECKPOINTED_BATCH_AND_ACCUM", "fix": {"grad_accum_steps": 8, "per_device_batch_size": 66}, "work": {"activation_per_sample_gb": 1.0, "max_per_device_batch": 66}} | {"archetype": "ckpt_batch_accum", "computations": ["R20", "R1", "R3"], "depth": 4, "interaction": "checkpointing x accum", "tier": "interacting"} |
train-ckpt_batch_accum-042 | ckpt_batch_accum | train | You are debugging a failed or misbehaving LLM training run.
Pick exactly ONE diagnosis code and give the minimal fix: only the config fields that must change, with corrected values.
Diagnosis codes:
- OOM_BATCH_TOO_LARGE
- ACTIVATION_MEMORY_NO_CHECKPOINTING
- EFFECTIVE_BATCH_MISMATCH
- LR_TOO_HIGH_NO_WARMUP
- PRECISIO... | {"diagnosis": "CHECKPOINTED_BATCH_AND_ACCUM", "fix": {"grad_accum_steps": 2, "per_device_batch_size": 22}, "work": {"activation_per_sample_gb": 1.5, "max_per_device_batch": 22}} | {"archetype": "ckpt_batch_accum", "computations": ["R20", "R1", "R3"], "depth": 4, "interaction": "checkpointing x accum", "tier": "interacting"} |
train-batch_from_components-056 | batch_from_components | train | You are debugging a failed or misbehaving LLM training run.
Pick exactly ONE diagnosis code and give the minimal fix: only the config fields that must change, with corrected values.
Diagnosis codes:
- OOM_BATCH_TOO_LARGE
- ACTIVATION_MEMORY_NO_CHECKPOINTING
- EFFECTIVE_BATCH_MISMATCH
- LR_TOO_HIGH_NO_WARMUP
- PRECISIO... | {"diagnosis": "BATCH_FROM_MEMORY_COMPONENTS", "fix": {"per_device_batch_size": 8}, "work": {"fixed_memory_gb": 64.0, "remaining_memory_gb": 16.0}} | {"archetype": "batch_from_components", "computations": ["R13", "R1"], "depth": 3, "interaction": null, "tier": "computed"} |
train-packing_time-061 | packing_time | train | You are debugging a failed or misbehaving LLM training run.
Pick exactly ONE diagnosis code and give the minimal fix: only the config fields that must change, with corrected values.
Diagnosis codes:
- OOM_BATCH_TOO_LARGE
- ACTIVATION_MEMORY_NO_CHECKPOINTING
- EFFECTIVE_BATCH_MISMATCH
- LR_TOO_HIGH_NO_WARMUP
- PRECISIO... | {"diagnosis": "PACKED_TRAINING_TIME", "fix": {"planned_time_hours": 0.5}, "work": {"total_padded_tokens": 491520000, "total_real_tokens": 216000000}} | {"archetype": "packing_time", "computations": ["R23", "R16"], "depth": 4, "interaction": "packing x time", "tier": "interacting"} |
train-batch_from_components-044 | batch_from_components | train | You are debugging a failed or misbehaving LLM training run.
Pick exactly ONE diagnosis code and give the minimal fix: only the config fields that must change, with corrected values.
Diagnosis codes:
- OOM_BATCH_TOO_LARGE
- ACTIVATION_MEMORY_NO_CHECKPOINTING
- EFFECTIVE_BATCH_MISMATCH
- LR_TOO_HIGH_NO_WARMUP
- PRECISIO... | {"diagnosis": "BATCH_FROM_MEMORY_COMPONENTS", "fix": {"per_device_batch_size": 16}, "work": {"fixed_memory_gb": 64.0, "remaining_memory_gb": 16.0}} | {"archetype": "batch_from_components", "computations": ["R13", "R1"], "depth": 3, "interaction": null, "tier": "computed"} |
train-checkpoint_retention-054 | checkpoint_retention | train | You are debugging a failed or misbehaving LLM training run.
Pick exactly ONE diagnosis code and give the minimal fix: only the config fields that must change, with corrected values.
Diagnosis codes:
- OOM_BATCH_TOO_LARGE
- ACTIVATION_MEMORY_NO_CHECKPOINTING
- EFFECTIVE_BATCH_MISMATCH
- LR_TOO_HIGH_NO_WARMUP
- PRECISIO... | {"diagnosis": "CHECKPOINT_RETENTION_LIMIT", "fix": {"max_checkpoints_to_keep": 8}, "work": {"checkpoint_size_gb": 24.0}} | {"archetype": "checkpoint_retention", "computations": ["R17"], "depth": 2, "interaction": null, "tier": "computed"} |
train-zero_batch-052 | zero_batch | train | You are debugging a failed or misbehaving LLM training run.
Pick exactly ONE diagnosis code and give the minimal fix: only the config fields that must change, with corrected values.
Diagnosis codes:
- OOM_BATCH_TOO_LARGE
- ACTIVATION_MEMORY_NO_CHECKPOINTING
- EFFECTIVE_BATCH_MISMATCH
- LR_TOO_HIGH_NO_WARMUP
- PRECISIO... | {"diagnosis": "ZERO_SHARDED_BATCH_SIZE", "fix": {"per_device_batch_size": 56}, "work": {"bytes_per_param_per_rank": 2.0, "fixed_memory_gb": 16.0, "remaining_memory_gb": 112.0}} | {"archetype": "zero_batch", "computations": ["R19", "R13", "R1"], "depth": 4, "interaction": "zero x memory", "tier": "interacting"} |
train-optimizer_batch-034 | optimizer_batch | train | You are debugging a failed or misbehaving LLM training run.
Pick exactly ONE diagnosis code and give the minimal fix: only the config fields that must change, with corrected values.
Diagnosis codes:
- OOM_BATCH_TOO_LARGE
- ACTIVATION_MEMORY_NO_CHECKPOINTING
- EFFECTIVE_BATCH_MISMATCH
- LR_TOO_HIGH_NO_WARMUP
- PRECISIO... | {"diagnosis": "OPTIMIZER_MEMORY_BATCH", "fix": {"per_device_batch_size": 104}, "work": {"bytes_per_param": 12, "fixed_memory_gb": 24.0}} | {"archetype": "optimizer_batch", "computations": ["R21", "R1"], "depth": 3, "interaction": "optimizer x memory", "tier": "interacting"} |
train-training_time-011 | training_time | train | You are debugging a failed or misbehaving LLM training run.
Pick exactly ONE diagnosis code and give the minimal fix: only the config fields that must change, with corrected values.
Diagnosis codes:
- OOM_BATCH_TOO_LARGE
- ACTIVATION_MEMORY_NO_CHECKPOINTING
- EFFECTIVE_BATCH_MISMATCH
- LR_TOO_HIGH_NO_WARMUP
- PRECISIO... | {"diagnosis": "TRAINING_TIME_ESTIMATE", "fix": {"planned_time_hours": 1.1377777777777778}, "work": {"total_tokens": 245760000}} | {"archetype": "training_time", "computations": ["R16"], "depth": 3, "interaction": null, "tier": "computed"} |
train-packing_time-033 | packing_time | train | You are debugging a failed or misbehaving LLM training run.
Pick exactly ONE diagnosis code and give the minimal fix: only the config fields that must change, with corrected values.
Diagnosis codes:
- OOM_BATCH_TOO_LARGE
- ACTIVATION_MEMORY_NO_CHECKPOINTING
- EFFECTIVE_BATCH_MISMATCH
- LR_TOO_HIGH_NO_WARMUP
- PRECISIO... | {"diagnosis": "PACKED_TRAINING_TIME", "fix": {"planned_time_hours": 0.5}, "work": {"total_padded_tokens": 163840000, "total_real_tokens": 144000000}} | {"archetype": "packing_time", "computations": ["R23", "R16"], "depth": 4, "interaction": "packing x time", "tier": "interacting"} |
train-two_lora_vocab-012 | two_lora_vocab | train | You are debugging a failed or misbehaving LLM training run.
Pick exactly ONE diagnosis code and give the minimal fix: only the config fields that must change, with corrected values.
Diagnosis codes:
- OOM_BATCH_TOO_LARGE
- ACTIVATION_MEMORY_NO_CHECKPOINTING
- EFFECTIVE_BATCH_MISMATCH
- LR_TOO_HIGH_NO_WARMUP
- PRECISIO... | {"diagnosis": ["ADAPTERS_FROZEN_ZERO_TRAINABLE", "EMBEDDING_VOCAB_MISMATCH"], "fix": {"adapter_requires_grad": true, "embedding_vocab_size": 50304}} | {"archetype": "two_lora_vocab", "computations": ["R8", "R11"], "depth": 2, "interaction": "mixed", "tier": "two_fault"} |
train-zero_batch-033 | zero_batch | train | You are debugging a failed or misbehaving LLM training run.
Pick exactly ONE diagnosis code and give the minimal fix: only the config fields that must change, with corrected values.
Diagnosis codes:
- OOM_BATCH_TOO_LARGE
- ACTIVATION_MEMORY_NO_CHECKPOINTING
- EFFECTIVE_BATCH_MISMATCH
- LR_TOO_HIGH_NO_WARMUP
- PRECISIO... | {"diagnosis": "ZERO_SHARDED_BATCH_SIZE", "fix": {"per_device_batch_size": 8}, "work": {"bytes_per_param_per_rank": 10.0, "fixed_memory_gb": 80.0, "remaining_memory_gb": 16.0}} | {"archetype": "zero_batch", "computations": ["R19", "R13", "R1"], "depth": 4, "interaction": "zero x memory", "tier": "interacting"} |
train-shard_size-031 | shard_size | train | You are debugging a failed or misbehaving LLM training run.
Pick exactly ONE diagnosis code and give the minimal fix: only the config fields that must change, with corrected values.
Diagnosis codes:
- OOM_BATCH_TOO_LARGE
- ACTIVATION_MEMORY_NO_CHECKPOINTING
- EFFECTIVE_BATCH_MISMATCH
- LR_TOO_HIGH_NO_WARMUP
- PRECISIO... | {"diagnosis": "SHARD_SIZE_PER_RANK", "fix": {"examples_per_rank": 25000, "shard_size_gb": 12.5}} | {"archetype": "shard_size", "computations": ["R18"], "depth": 2, "interaction": null, "tier": "computed"} |
train-lr_at_step-046 | lr_at_step | train | You are debugging a failed or misbehaving LLM training run.
Pick exactly ONE diagnosis code and give the minimal fix: only the config fields that must change, with corrected values.
Diagnosis codes:
- OOM_BATCH_TOO_LARGE
- ACTIVATION_MEMORY_NO_CHECKPOINTING
- EFFECTIVE_BATCH_MISMATCH
- LR_TOO_HIGH_NO_WARMUP
- PRECISIO... | {"diagnosis": "LR_VALUE_AT_STEP", "fix": {"current_learning_rate": 0.000465}, "work": {"step_fraction": 0.25}} | {"archetype": "lr_at_step", "computations": ["R15"], "depth": 2, "interaction": null, "tier": "computed"} |
train-optimizer_batch-045 | optimizer_batch | train | You are debugging a failed or misbehaving LLM training run.
Pick exactly ONE diagnosis code and give the minimal fix: only the config fields that must change, with corrected values.
Diagnosis codes:
- OOM_BATCH_TOO_LARGE
- ACTIVATION_MEMORY_NO_CHECKPOINTING
- EFFECTIVE_BATCH_MISMATCH
- LR_TOO_HIGH_NO_WARMUP
- PRECISIO... | {"diagnosis": "OPTIMIZER_MEMORY_BATCH", "fix": {"per_device_batch_size": 17}, "work": {"bytes_per_param": 10, "fixed_memory_gb": 70.0}} | {"archetype": "optimizer_batch", "computations": ["R21", "R1"], "depth": 3, "interaction": "optimizer x memory", "tier": "interacting"} |
train-lora_frozen-002 | lora_frozen | train | You are debugging a failed or misbehaving LLM training run.
Pick exactly ONE diagnosis code and give the minimal fix: only the config fields that must change, with corrected values.
Diagnosis codes:
- OOM_BATCH_TOO_LARGE
- ACTIVATION_MEMORY_NO_CHECKPOINTING
- EFFECTIVE_BATCH_MISMATCH
- LR_TOO_HIGH_NO_WARMUP
- PRECISIO... | {"diagnosis": "ADAPTERS_FROZEN_ZERO_TRAINABLE", "fix": {"adapter_requires_grad": true}} | {"archetype": "lora_frozen", "computations": ["R8"], "depth": 1, "interaction": null, "tier": "diagnosis"} |
train-two_oom_clipping-006 | two_oom_clipping | train | You are debugging a failed or misbehaving LLM training run.
Pick exactly ONE diagnosis code and give the minimal fix: only the config fields that must change, with corrected values.
Diagnosis codes:
- OOM_BATCH_TOO_LARGE
- ACTIVATION_MEMORY_NO_CHECKPOINTING
- EFFECTIVE_BATCH_MISMATCH
- LR_TOO_HIGH_NO_WARMUP
- PRECISIO... | {"diagnosis": ["OOM_BATCH_TOO_LARGE", "NO_GRADIENT_CLIPPING"], "fix": {"max_grad_norm": 1.0, "per_device_batch_size": 17}} | {"archetype": "two_oom_clipping", "computations": ["R1", "R12"], "depth": 3, "interaction": "oom+clipping", "tier": "two_fault"} |
train-checkpoint_retention-004 | checkpoint_retention | train | You are debugging a failed or misbehaving LLM training run.
Pick exactly ONE diagnosis code and give the minimal fix: only the config fields that must change, with corrected values.
Diagnosis codes:
- OOM_BATCH_TOO_LARGE
- ACTIVATION_MEMORY_NO_CHECKPOINTING
- EFFECTIVE_BATCH_MISMATCH
- LR_TOO_HIGH_NO_WARMUP
- PRECISIO... | {"diagnosis": "CHECKPOINT_RETENTION_LIMIT", "fix": {"max_checkpoints_to_keep": 6}, "work": {"checkpoint_size_gb": 112.0}} | {"archetype": "checkpoint_retention", "computations": ["R17"], "depth": 2, "interaction": null, "tier": "computed"} |
train-ckpt_batch_accum-035 | ckpt_batch_accum | train | You are debugging a failed or misbehaving LLM training run.
Pick exactly ONE diagnosis code and give the minimal fix: only the config fields that must change, with corrected values.
Diagnosis codes:
- OOM_BATCH_TOO_LARGE
- ACTIVATION_MEMORY_NO_CHECKPOINTING
- EFFECTIVE_BATCH_MISMATCH
- LR_TOO_HIGH_NO_WARMUP
- PRECISIO... | {"diagnosis": "CHECKPOINTED_BATCH_AND_ACCUM", "fix": {"grad_accum_steps": 2, "per_device_batch_size": 26}, "work": {"activation_per_sample_gb": 1.0, "max_per_device_batch": 26}} | {"archetype": "ckpt_batch_accum", "computations": ["R20", "R1", "R3"], "depth": 4, "interaction": "checkpointing x accum", "tier": "interacting"} |
train-oom_batch-002 | oom_batch | train | You are debugging a failed or misbehaving LLM training run.
Pick exactly ONE diagnosis code and give the minimal fix: only the config fields that must change, with corrected values.
Diagnosis codes:
- OOM_BATCH_TOO_LARGE
- ACTIVATION_MEMORY_NO_CHECKPOINTING
- EFFECTIVE_BATCH_MISMATCH
- LR_TOO_HIGH_NO_WARMUP
- PRECISIO... | {"diagnosis": "OOM_BATCH_TOO_LARGE", "fix": {"per_device_batch_size": 24}} | {"archetype": "oom_batch", "computations": ["R1"], "depth": 2, "interaction": null, "tier": "diagnosis"} |
train-fp16_scaler-001 | fp16_scaler | train | You are debugging a failed or misbehaving LLM training run.
Pick exactly ONE diagnosis code and give the minimal fix: only the config fields that must change, with corrected values.
Diagnosis codes:
- OOM_BATCH_TOO_LARGE
- ACTIVATION_MEMORY_NO_CHECKPOINTING
- EFFECTIVE_BATCH_MISMATCH
- LR_TOO_HIGH_NO_WARMUP
- PRECISIO... | {"diagnosis": "FP16_NO_GRAD_SCALER", "fix": {"use_grad_scaler": true}} | {"archetype": "fp16_scaler", "computations": ["R6"], "depth": 1, "interaction": null, "tier": "diagnosis"} |
train-accum_and_steps-013 | accum_and_steps | train | You are debugging a failed or misbehaving LLM training run.
Pick exactly ONE diagnosis code and give the minimal fix: only the config fields that must change, with corrected values.
Diagnosis codes:
- OOM_BATCH_TOO_LARGE
- ACTIVATION_MEMORY_NO_CHECKPOINTING
- EFFECTIVE_BATCH_MISMATCH
- LR_TOO_HIGH_NO_WARMUP
- PRECISIO... | {"diagnosis": "ACCUM_AND_TOTAL_STEPS", "fix": {"grad_accum_steps": 2, "planned_total_steps": 800}, "work": {"steps_per_epoch": 200}} | {"archetype": "accum_and_steps", "computations": ["R3", "R14"], "depth": 3, "interaction": null, "tier": "computed"} |
train-accum_and_steps-023 | accum_and_steps | train | You are debugging a failed or misbehaving LLM training run.
Pick exactly ONE diagnosis code and give the minimal fix: only the config fields that must change, with corrected values.
Diagnosis codes:
- OOM_BATCH_TOO_LARGE
- ACTIVATION_MEMORY_NO_CHECKPOINTING
- EFFECTIVE_BATCH_MISMATCH
- LR_TOO_HIGH_NO_WARMUP
- PRECISIO... | {"diagnosis": "ACCUM_AND_TOTAL_STEPS", "fix": {"grad_accum_steps": 4, "planned_total_steps": 1500}, "work": {"steps_per_epoch": 500}} | {"archetype": "accum_and_steps", "computations": ["R3", "R14"], "depth": 3, "interaction": null, "tier": "computed"} |
train-checkpoint_retention-053 | checkpoint_retention | train | You are debugging a failed or misbehaving LLM training run.
Pick exactly ONE diagnosis code and give the minimal fix: only the config fields that must change, with corrected values.
Diagnosis codes:
- OOM_BATCH_TOO_LARGE
- ACTIVATION_MEMORY_NO_CHECKPOINTING
- EFFECTIVE_BATCH_MISMATCH
- LR_TOO_HIGH_NO_WARMUP
- PRECISIO... | {"diagnosis": "CHECKPOINT_RETENTION_LIMIT", "fix": {"max_checkpoints_to_keep": 6}, "work": {"checkpoint_size_gb": 112.0}} | {"archetype": "checkpoint_retention", "computations": ["R17"], "depth": 2, "interaction": null, "tier": "computed"} |
train-zero_batch-061 | zero_batch | train | You are debugging a failed or misbehaving LLM training run.
Pick exactly ONE diagnosis code and give the minimal fix: only the config fields that must change, with corrected values.
Diagnosis codes:
- OOM_BATCH_TOO_LARGE
- ACTIVATION_MEMORY_NO_CHECKPOINTING
- EFFECTIVE_BATCH_MISMATCH
- LR_TOO_HIGH_NO_WARMUP
- PRECISIO... | {"diagnosis": "ZERO_SHARDED_BATCH_SIZE", "fix": {"per_device_batch_size": 152}, "work": {"bytes_per_param_per_rank": 10.0, "fixed_memory_gb": 40.0, "remaining_memory_gb": 152.0}} | {"archetype": "zero_batch", "computations": ["R19", "R13", "R1"], "depth": 4, "interaction": "zero x memory", "tier": "interacting"} |
train-batch_from_components-031 | batch_from_components | train | You are debugging a failed or misbehaving LLM training run.
Pick exactly ONE diagnosis code and give the minimal fix: only the config fields that must change, with corrected values.
Diagnosis codes:
- OOM_BATCH_TOO_LARGE
- ACTIVATION_MEMORY_NO_CHECKPOINTING
- EFFECTIVE_BATCH_MISMATCH
- LR_TOO_HIGH_NO_WARMUP
- PRECISIO... | {"diagnosis": "BATCH_FROM_MEMORY_COMPONENTS", "fix": {"per_device_batch_size": 22}, "work": {"fixed_memory_gb": 24.0, "remaining_memory_gb": 56.0}} | {"archetype": "batch_from_components", "computations": ["R13", "R1"], "depth": 3, "interaction": null, "tier": "computed"} |
train-training_time-016 | training_time | train | You are debugging a failed or misbehaving LLM training run.
Pick exactly ONE diagnosis code and give the minimal fix: only the config fields that must change, with corrected values.
Diagnosis codes:
- OOM_BATCH_TOO_LARGE
- ACTIVATION_MEMORY_NO_CHECKPOINTING
- EFFECTIVE_BATCH_MISMATCH
- LR_TOO_HIGH_NO_WARMUP
- PRECISIO... | {"diagnosis": "TRAINING_TIME_ESTIMATE", "fix": {"planned_time_hours": 0.37925925925925924}, "work": {"total_tokens": 81920000}} | {"archetype": "training_time", "computations": ["R16"], "depth": 3, "interaction": null, "tier": "computed"} |
train-checkpoint_retention-050 | checkpoint_retention | train | You are debugging a failed or misbehaving LLM training run.
Pick exactly ONE diagnosis code and give the minimal fix: only the config fields that must change, with corrected values.
Diagnosis codes:
- OOM_BATCH_TOO_LARGE
- ACTIVATION_MEMORY_NO_CHECKPOINTING
- EFFECTIVE_BATCH_MISMATCH
- LR_TOO_HIGH_NO_WARMUP
- PRECISIO... | {"diagnosis": "CHECKPOINT_RETENTION_LIMIT", "fix": {"max_checkpoints_to_keep": 10}, "work": {"checkpoint_size_gb": 48.0}} | {"archetype": "checkpoint_retention", "computations": ["R17"], "depth": 2, "interaction": null, "tier": "computed"} |
train-shard_size-043 | shard_size | train | You are debugging a failed or misbehaving LLM training run.
Pick exactly ONE diagnosis code and give the minimal fix: only the config fields that must change, with corrected values.
Diagnosis codes:
- OOM_BATCH_TOO_LARGE
- ACTIVATION_MEMORY_NO_CHECKPOINTING
- EFFECTIVE_BATCH_MISMATCH
- LR_TOO_HIGH_NO_WARMUP
- PRECISIO... | {"diagnosis": "SHARD_SIZE_PER_RANK", "fix": {"examples_per_rank": 5000, "shard_size_gb": 40.0}} | {"archetype": "shard_size", "computations": ["R18"], "depth": 2, "interaction": null, "tier": "computed"} |
train-accum_steps_lr-004 | accum_steps_lr | train | You are debugging a failed or misbehaving LLM training run.
Pick exactly ONE diagnosis code and give the minimal fix: only the config fields that must change, with corrected values.
Diagnosis codes:
- OOM_BATCH_TOO_LARGE
- ACTIVATION_MEMORY_NO_CHECKPOINTING
- EFFECTIVE_BATCH_MISMATCH
- LR_TOO_HIGH_NO_WARMUP
- PRECISIO... | {"diagnosis": "ACCUM_STEPS_AND_LR", "fix": {"current_learning_rate": 0.000465, "grad_accum_steps": 1, "planned_total_steps": 500}, "work": {"audit_step": 125, "steps_per_epoch": 250}} | {"archetype": "accum_steps_lr", "computations": ["R3", "R14", "R15"], "depth": 5, "interaction": "accum x schedule", "tier": "interacting"} |
train-dataloader-004 | dataloader | train | You are debugging a failed or misbehaving LLM training run.
Pick exactly ONE diagnosis code and give the minimal fix: only the config fields that must change, with corrected values.
Diagnosis codes:
- OOM_BATCH_TOO_LARGE
- ACTIVATION_MEMORY_NO_CHECKPOINTING
- EFFECTIVE_BATCH_MISMATCH
- LR_TOO_HIGH_NO_WARMUP
- PRECISIO... | {"diagnosis": "DATALOADER_FORK_WITH_CUDA", "fix": {"dataloader_multiprocessing_context": "spawn"}} | {"archetype": "dataloader", "computations": ["R9"], "depth": 1, "interaction": null, "tier": "diagnosis"} |
train-two_v100_dataloader-054 | two_v100_dataloader | train | You are debugging a failed or misbehaving LLM training run.
Pick exactly ONE diagnosis code and give the minimal fix: only the config fields that must change, with corrected values.
Diagnosis codes:
- OOM_BATCH_TOO_LARGE
- ACTIVATION_MEMORY_NO_CHECKPOINTING
- EFFECTIVE_BATCH_MISMATCH
- LR_TOO_HIGH_NO_WARMUP
- PRECISIO... | {"diagnosis": ["PRECISION_UNSUPPORTED_ON_GPU", "DATALOADER_FORK_WITH_CUDA"], "fix": {"dataloader_multiprocessing_context": "spawn", "dtype": "fp16", "use_grad_scaler": true}} | {"archetype": "two_v100_dataloader", "computations": ["R5", "R9"], "depth": 2, "interaction": "mixed", "tier": "two_fault"} |
train-two_sampler_accum-045 | two_sampler_accum | train | You are debugging a failed or misbehaving LLM training run.
Pick exactly ONE diagnosis code and give the minimal fix: only the config fields that must change, with corrected values.
Diagnosis codes:
- OOM_BATCH_TOO_LARGE
- ACTIVATION_MEMORY_NO_CHECKPOINTING
- EFFECTIVE_BATCH_MISMATCH
- LR_TOO_HIGH_NO_WARMUP
- PRECISIO... | {"diagnosis": ["DUPLICATE_DATA_NO_DISTRIBUTED_SAMPLER", "EFFECTIVE_BATCH_MISMATCH"], "fix": {"grad_accum_steps": 2, "sampler": "DistributedSampler"}} | {"archetype": "two_sampler_accum", "computations": ["R7", "R3"], "depth": 3, "interaction": "sampler+accum", "tier": "two_fault"} |
train-zero_batch-050 | zero_batch | train | You are debugging a failed or misbehaving LLM training run.
Pick exactly ONE diagnosis code and give the minimal fix: only the config fields that must change, with corrected values.
Diagnosis codes:
- OOM_BATCH_TOO_LARGE
- ACTIVATION_MEMORY_NO_CHECKPOINTING
- EFFECTIVE_BATCH_MISMATCH
- LR_TOO_HIGH_NO_WARMUP
- PRECISIO... | {"diagnosis": "ZERO_SHARDED_BATCH_SIZE", "fix": {"per_device_batch_size": 16}, "work": {"bytes_per_param_per_rank": 11.0, "fixed_memory_gb": 88.0, "remaining_memory_gb": 40.0}} | {"archetype": "zero_batch", "computations": ["R19", "R13", "R1"], "depth": 4, "interaction": "zero x memory", "tier": "interacting"} |
train-shard_size-057 | shard_size | train | You are debugging a failed or misbehaving LLM training run.
Pick exactly ONE diagnosis code and give the minimal fix: only the config fields that must change, with corrected values.
Diagnosis codes:
- OOM_BATCH_TOO_LARGE
- ACTIVATION_MEMORY_NO_CHECKPOINTING
- EFFECTIVE_BATCH_MISMATCH
- LR_TOO_HIGH_NO_WARMUP
- PRECISIO... | {"diagnosis": "SHARD_SIZE_PER_RANK", "fix": {"examples_per_rank": 5000, "shard_size_gb": 60.0}} | {"archetype": "shard_size", "computations": ["R18"], "depth": 2, "interaction": null, "tier": "computed"} |
train-shard_size-040 | shard_size | train | You are debugging a failed or misbehaving LLM training run.
Pick exactly ONE diagnosis code and give the minimal fix: only the config fields that must change, with corrected values.
Diagnosis codes:
- OOM_BATCH_TOO_LARGE
- ACTIVATION_MEMORY_NO_CHECKPOINTING
- EFFECTIVE_BATCH_MISMATCH
- LR_TOO_HIGH_NO_WARMUP
- PRECISIO... | {"diagnosis": "SHARD_SIZE_PER_RANK", "fix": {"examples_per_rank": 12500, "shard_size_gb": 25.0}} | {"archetype": "shard_size", "computations": ["R18"], "depth": 2, "interaction": null, "tier": "computed"} |
train-two_oom_clipping-010 | two_oom_clipping | train | You are debugging a failed or misbehaving LLM training run.
Pick exactly ONE diagnosis code and give the minimal fix: only the config fields that must change, with corrected values.
Diagnosis codes:
- OOM_BATCH_TOO_LARGE
- ACTIVATION_MEMORY_NO_CHECKPOINTING
- EFFECTIVE_BATCH_MISMATCH
- LR_TOO_HIGH_NO_WARMUP
- PRECISIO... | {"diagnosis": ["OOM_BATCH_TOO_LARGE", "NO_GRADIENT_CLIPPING"], "fix": {"max_grad_norm": 1.0, "per_device_batch_size": 9}} | {"archetype": "two_oom_clipping", "computations": ["R1", "R12"], "depth": 3, "interaction": "oom+clipping", "tier": "two_fault"} |
train-checkpoint_retention-064 | checkpoint_retention | train | You are debugging a failed or misbehaving LLM training run.
Pick exactly ONE diagnosis code and give the minimal fix: only the config fields that must change, with corrected values.
Diagnosis codes:
- OOM_BATCH_TOO_LARGE
- ACTIVATION_MEMORY_NO_CHECKPOINTING
- EFFECTIVE_BATCH_MISMATCH
- LR_TOO_HIGH_NO_WARMUP
- PRECISIO... | {"diagnosis": "CHECKPOINT_RETENTION_LIMIT", "fix": {"max_checkpoints_to_keep": 8}, "work": {"checkpoint_size_gb": 112.0}} | {"archetype": "checkpoint_retention", "computations": ["R17"], "depth": 2, "interaction": null, "tier": "computed"} |
train-two_v100_dataloader-047 | two_v100_dataloader | train | You are debugging a failed or misbehaving LLM training run.
Pick exactly ONE diagnosis code and give the minimal fix: only the config fields that must change, with corrected values.
Diagnosis codes:
- OOM_BATCH_TOO_LARGE
- ACTIVATION_MEMORY_NO_CHECKPOINTING
- EFFECTIVE_BATCH_MISMATCH
- LR_TOO_HIGH_NO_WARMUP
- PRECISIO... | {"diagnosis": ["PRECISION_UNSUPPORTED_ON_GPU", "DATALOADER_FORK_WITH_CUDA"], "fix": {"dataloader_multiprocessing_context": "spawn", "dtype": "fp16", "use_grad_scaler": true}} | {"archetype": "two_v100_dataloader", "computations": ["R5", "R9"], "depth": 2, "interaction": "mixed", "tier": "two_fault"} |
train-accum_and_steps-005 | accum_and_steps | train | You are debugging a failed or misbehaving LLM training run.
Pick exactly ONE diagnosis code and give the minimal fix: only the config fields that must change, with corrected values.
Diagnosis codes:
- OOM_BATCH_TOO_LARGE
- ACTIVATION_MEMORY_NO_CHECKPOINTING
- EFFECTIVE_BATCH_MISMATCH
- LR_TOO_HIGH_NO_WARMUP
- PRECISIO... | {"diagnosis": "ACCUM_AND_TOTAL_STEPS", "fix": {"grad_accum_steps": 1, "planned_total_steps": 600}, "work": {"steps_per_epoch": 200}} | {"archetype": "accum_and_steps", "computations": ["R3", "R14"], "depth": 3, "interaction": null, "tier": "computed"} |
train-optimizer_batch-033 | optimizer_batch | train | You are debugging a failed or misbehaving LLM training run.
Pick exactly ONE diagnosis code and give the minimal fix: only the config fields that must change, with corrected values.
Diagnosis codes:
- OOM_BATCH_TOO_LARGE
- ACTIVATION_MEMORY_NO_CHECKPOINTING
- EFFECTIVE_BATCH_MISMATCH
- LR_TOO_HIGH_NO_WARMUP
- PRECISIO... | {"diagnosis": "OPTIMIZER_MEMORY_BATCH", "fix": {"per_device_batch_size": 12}, "work": {"bytes_per_param": 12, "fixed_memory_gb": 84.0}} | {"archetype": "optimizer_batch", "computations": ["R21", "R1"], "depth": 3, "interaction": "optimizer x memory", "tier": "interacting"} |
train-zero_batch-056 | zero_batch | train | You are debugging a failed or misbehaving LLM training run.
Pick exactly ONE diagnosis code and give the minimal fix: only the config fields that must change, with corrected values.
Diagnosis codes:
- OOM_BATCH_TOO_LARGE
- ACTIVATION_MEMORY_NO_CHECKPOINTING
- EFFECTIVE_BATCH_MISMATCH
- LR_TOO_HIGH_NO_WARMUP
- PRECISIO... | {"diagnosis": "ZERO_SHARDED_BATCH_SIZE", "fix": {"per_device_batch_size": 24}, "work": {"bytes_per_param_per_rank": 8.5, "fixed_memory_gb": 34.0, "remaining_memory_gb": 62.0}} | {"archetype": "zero_batch", "computations": ["R19", "R13", "R1"], "depth": 4, "interaction": "zero x memory", "tier": "interacting"} |
train-trap_nan_clipping-010 | trap_nan_clipping | train | You are debugging a failed or misbehaving LLM training run.
Pick exactly ONE diagnosis code and give the minimal fix: only the config fields that must change, with corrected values.
Diagnosis codes:
- OOM_BATCH_TOO_LARGE
- ACTIVATION_MEMORY_NO_CHECKPOINTING
- EFFECTIVE_BATCH_MISMATCH
- LR_TOO_HIGH_NO_WARMUP
- PRECISIO... | {"diagnosis": "NO_GRADIENT_CLIPPING", "fix": {"max_grad_norm": 1.0}} | {"archetype": "trap_nan_clipping", "computations": ["R4", "R12"], "depth": 2, "interaction": "lr", "tier": "trap"} |
train-two_v100_dataloader-033 | two_v100_dataloader | train | You are debugging a failed or misbehaving LLM training run.
Pick exactly ONE diagnosis code and give the minimal fix: only the config fields that must change, with corrected values.
Diagnosis codes:
- OOM_BATCH_TOO_LARGE
- ACTIVATION_MEMORY_NO_CHECKPOINTING
- EFFECTIVE_BATCH_MISMATCH
- LR_TOO_HIGH_NO_WARMUP
- PRECISIO... | {"diagnosis": ["PRECISION_UNSUPPORTED_ON_GPU", "DATALOADER_FORK_WITH_CUDA"], "fix": {"dataloader_multiprocessing_context": "spawn", "dtype": "fp16", "use_grad_scaler": true}} | {"archetype": "two_v100_dataloader", "computations": ["R5", "R9"], "depth": 2, "interaction": "mixed", "tier": "two_fault"} |
train-ckpt_batch_accum-048 | ckpt_batch_accum | train | You are debugging a failed or misbehaving LLM training run.
Pick exactly ONE diagnosis code and give the minimal fix: only the config fields that must change, with corrected values.
Diagnosis codes:
- OOM_BATCH_TOO_LARGE
- ACTIVATION_MEMORY_NO_CHECKPOINTING
- EFFECTIVE_BATCH_MISMATCH
- LR_TOO_HIGH_NO_WARMUP
- PRECISIO... | {"diagnosis": "CHECKPOINTED_BATCH_AND_ACCUM", "fix": {"grad_accum_steps": 4, "per_device_batch_size": 15}, "work": {"activation_per_sample_gb": 2.0, "max_per_device_batch": 15}} | {"archetype": "ckpt_batch_accum", "computations": ["R20", "R1", "R3"], "depth": 4, "interaction": "checkpointing x accum", "tier": "interacting"} |
train-lr_at_step-007 | lr_at_step | train | You are debugging a failed or misbehaving LLM training run.
Pick exactly ONE diagnosis code and give the minimal fix: only the config fields that must change, with corrected values.
Diagnosis codes:
- OOM_BATCH_TOO_LARGE
- ACTIVATION_MEMORY_NO_CHECKPOINTING
- EFFECTIVE_BATCH_MISMATCH
- LR_TOO_HIGH_NO_WARMUP
- PRECISIO... | {"diagnosis": "LR_VALUE_AT_STEP", "fix": {"current_learning_rate": 0.0002325}, "work": {"step_fraction": 0.25}} | {"archetype": "lr_at_step", "computations": ["R15"], "depth": 2, "interaction": null, "tier": "computed"} |
train-device_mismatch-006 | device_mismatch | train | You are debugging a failed or misbehaving LLM training run.
Pick exactly ONE diagnosis code and give the minimal fix: only the config fields that must change, with corrected values.
Diagnosis codes:
- OOM_BATCH_TOO_LARGE
- ACTIVATION_MEMORY_NO_CHECKPOINTING
- EFFECTIVE_BATCH_MISMATCH
- LR_TOO_HIGH_NO_WARMUP
- PRECISIO... | {"diagnosis": "DEVICE_MISMATCH_LOSS", "fix": {"loss_device": "cuda:0"}} | {"archetype": "device_mismatch", "computations": ["R10"], "depth": 1, "interaction": null, "tier": "diagnosis"} |
train-optimizer_batch-040 | optimizer_batch | train | You are debugging a failed or misbehaving LLM training run.
Pick exactly ONE diagnosis code and give the minimal fix: only the config fields that must change, with corrected values.
Diagnosis codes:
- OOM_BATCH_TOO_LARGE
- ACTIVATION_MEMORY_NO_CHECKPOINTING
- EFFECTIVE_BATCH_MISMATCH
- LR_TOO_HIGH_NO_WARMUP
- PRECISIO... | {"diagnosis": "OPTIMIZER_MEMORY_BATCH", "fix": {"per_device_batch_size": 17}, "work": {"bytes_per_param": 10, "fixed_memory_gb": 70.0}} | {"archetype": "optimizer_batch", "computations": ["R21", "R1"], "depth": 3, "interaction": "optimizer x memory", "tier": "interacting"} |
train-oom_batch-001 | oom_batch | train | You are debugging a failed or misbehaving LLM training run.
Pick exactly ONE diagnosis code and give the minimal fix: only the config fields that must change, with corrected values.
Diagnosis codes:
- OOM_BATCH_TOO_LARGE
- ACTIVATION_MEMORY_NO_CHECKPOINTING
- EFFECTIVE_BATCH_MISMATCH
- LR_TOO_HIGH_NO_WARMUP
- PRECISIO... | {"diagnosis": "OOM_BATCH_TOO_LARGE", "fix": {"per_device_batch_size": 68}} | {"archetype": "oom_batch", "computations": ["R1"], "depth": 2, "interaction": null, "tier": "diagnosis"} |
train-two_lora_vocab-052 | two_lora_vocab | train | You are debugging a failed or misbehaving LLM training run.
Pick exactly ONE diagnosis code and give the minimal fix: only the config fields that must change, with corrected values.
Diagnosis codes:
- OOM_BATCH_TOO_LARGE
- ACTIVATION_MEMORY_NO_CHECKPOINTING
- EFFECTIVE_BATCH_MISMATCH
- LR_TOO_HIGH_NO_WARMUP
- PRECISIO... | {"diagnosis": ["ADAPTERS_FROZEN_ZERO_TRAINABLE", "EMBEDDING_VOCAB_MISMATCH"], "fix": {"adapter_requires_grad": true, "embedding_vocab_size": 32064}} | {"archetype": "two_lora_vocab", "computations": ["R8", "R11"], "depth": 2, "interaction": "mixed", "tier": "two_fault"} |
train-batch_from_components-057 | batch_from_components | train | You are debugging a failed or misbehaving LLM training run.
Pick exactly ONE diagnosis code and give the minimal fix: only the config fields that must change, with corrected values.
Diagnosis codes:
- OOM_BATCH_TOO_LARGE
- ACTIVATION_MEMORY_NO_CHECKPOINTING
- EFFECTIVE_BATCH_MISMATCH
- LR_TOO_HIGH_NO_WARMUP
- PRECISIO... | {"diagnosis": "BATCH_FROM_MEMORY_COMPONENTS", "fix": {"per_device_batch_size": 22}, "work": {"fixed_memory_gb": 24.0, "remaining_memory_gb": 56.0}} | {"archetype": "batch_from_components", "computations": ["R13", "R1"], "depth": 3, "interaction": null, "tier": "computed"} |
train-shard_size-024 | shard_size | train | You are debugging a failed or misbehaving LLM training run.
Pick exactly ONE diagnosis code and give the minimal fix: only the config fields that must change, with corrected values.
Diagnosis codes:
- OOM_BATCH_TOO_LARGE
- ACTIVATION_MEMORY_NO_CHECKPOINTING
- EFFECTIVE_BATCH_MISMATCH
- LR_TOO_HIGH_NO_WARMUP
- PRECISIO... | {"diagnosis": "SHARD_SIZE_PER_RANK", "fix": {"examples_per_rank": 5000, "shard_size_gb": 60.0}} | {"archetype": "shard_size", "computations": ["R18"], "depth": 2, "interaction": null, "tier": "computed"} |
train-optimizer_batch-029 | optimizer_batch | train | You are debugging a failed or misbehaving LLM training run.
Pick exactly ONE diagnosis code and give the minimal fix: only the config fields that must change, with corrected values.
Diagnosis codes:
- OOM_BATCH_TOO_LARGE
- ACTIVATION_MEMORY_NO_CHECKPOINTING
- EFFECTIVE_BATCH_MISMATCH
- LR_TOO_HIGH_NO_WARMUP
- PRECISIO... | {"diagnosis": "OPTIMIZER_MEMORY_BATCH", "fix": {"per_device_batch_size": 48}, "work": {"bytes_per_param": 8, "fixed_memory_gb": 24.0}} | {"archetype": "optimizer_batch", "computations": ["R21", "R1"], "depth": 3, "interaction": "optimizer x memory", "tier": "interacting"} |
train-batch_from_components-045 | batch_from_components | train | You are debugging a failed or misbehaving LLM training run.
Pick exactly ONE diagnosis code and give the minimal fix: only the config fields that must change, with corrected values.
Diagnosis codes:
- OOM_BATCH_TOO_LARGE
- ACTIVATION_MEMORY_NO_CHECKPOINTING
- EFFECTIVE_BATCH_MISMATCH
- LR_TOO_HIGH_NO_WARMUP
- PRECISIO... | {"diagnosis": "BATCH_FROM_MEMORY_COMPONENTS", "fix": {"per_device_batch_size": 10}, "work": {"fixed_memory_gb": 64.0, "remaining_memory_gb": 16.0}} | {"archetype": "batch_from_components", "computations": ["R13", "R1"], "depth": 3, "interaction": null, "tier": "computed"} |
train-device_mismatch-007 | device_mismatch | train | You are debugging a failed or misbehaving LLM training run.
Pick exactly ONE diagnosis code and give the minimal fix: only the config fields that must change, with corrected values.
Diagnosis codes:
- OOM_BATCH_TOO_LARGE
- ACTIVATION_MEMORY_NO_CHECKPOINTING
- EFFECTIVE_BATCH_MISMATCH
- LR_TOO_HIGH_NO_WARMUP
- PRECISIO... | {"diagnosis": "DEVICE_MISMATCH_LOSS", "fix": {"loss_device": "cuda:0"}} | {"archetype": "device_mismatch", "computations": ["R10"], "depth": 1, "interaction": null, "tier": "diagnosis"} |
train-two_oom_clipping-030 | two_oom_clipping | train | You are debugging a failed or misbehaving LLM training run.
Pick exactly ONE diagnosis code and give the minimal fix: only the config fields that must change, with corrected values.
Diagnosis codes:
- OOM_BATCH_TOO_LARGE
- ACTIVATION_MEMORY_NO_CHECKPOINTING
- EFFECTIVE_BATCH_MISMATCH
- LR_TOO_HIGH_NO_WARMUP
- PRECISIO... | {"diagnosis": ["OOM_BATCH_TOO_LARGE", "NO_GRADIENT_CLIPPING"], "fix": {"max_grad_norm": 1.0, "per_device_batch_size": 26}} | {"archetype": "two_oom_clipping", "computations": ["R1", "R12"], "depth": 3, "interaction": "oom+clipping", "tier": "two_fault"} |
train-parallel_batch-047 | parallel_batch | train | You are debugging a failed or misbehaving LLM training run.
Pick exactly ONE diagnosis code and give the minimal fix: only the config fields that must change, with corrected values.
Diagnosis codes:
- OOM_BATCH_TOO_LARGE
- ACTIVATION_MEMORY_NO_CHECKPOINTING
- EFFECTIVE_BATCH_MISMATCH
- LR_TOO_HIGH_NO_WARMUP
- PRECISIO... | {"diagnosis": "PARALLEL_SHARDED_BATCH", "fix": {"per_device_batch_size": 16}, "work": {"fixed_memory_gb": 64.0, "params_per_rank_billions": 4.0}} | {"archetype": "parallel_batch", "computations": ["R22", "R13", "R1"], "depth": 4, "interaction": "parallelism x memory", "tier": "interacting"} |
train-lr_at_step-006 | lr_at_step | train | You are debugging a failed or misbehaving LLM training run.
Pick exactly ONE diagnosis code and give the minimal fix: only the config fields that must change, with corrected values.
Diagnosis codes:
- OOM_BATCH_TOO_LARGE
- ACTIVATION_MEMORY_NO_CHECKPOINTING
- EFFECTIVE_BATCH_MISMATCH
- LR_TOO_HIGH_NO_WARMUP
- PRECISIO... | {"diagnosis": "LR_VALUE_AT_STEP", "fix": {"current_learning_rate": 9.75e-05}, "work": {"step_fraction": 0.75}} | {"archetype": "lr_at_step", "computations": ["R15"], "depth": 2, "interaction": null, "tier": "computed"} |
train-shard_size-032 | shard_size | train | You are debugging a failed or misbehaving LLM training run.
Pick exactly ONE diagnosis code and give the minimal fix: only the config fields that must change, with corrected values.
Diagnosis codes:
- OOM_BATCH_TOO_LARGE
- ACTIVATION_MEMORY_NO_CHECKPOINTING
- EFFECTIVE_BATCH_MISMATCH
- LR_TOO_HIGH_NO_WARMUP
- PRECISIO... | {"diagnosis": "SHARD_SIZE_PER_RANK", "fix": {"examples_per_rank": 5000, "shard_size_gb": 60.0}} | {"archetype": "shard_size", "computations": ["R18"], "depth": 2, "interaction": null, "tier": "computed"} |
train-shard_size-020 | shard_size | train | You are debugging a failed or misbehaving LLM training run.
Pick exactly ONE diagnosis code and give the minimal fix: only the config fields that must change, with corrected values.
Diagnosis codes:
- OOM_BATCH_TOO_LARGE
- ACTIVATION_MEMORY_NO_CHECKPOINTING
- EFFECTIVE_BATCH_MISMATCH
- LR_TOO_HIGH_NO_WARMUP
- PRECISIO... | {"diagnosis": "SHARD_SIZE_PER_RANK", "fix": {"examples_per_rank": 5000, "shard_size_gb": 25.0}} | {"archetype": "shard_size", "computations": ["R18"], "depth": 2, "interaction": null, "tier": "computed"} |
train-batch_from_components-051 | batch_from_components | train | You are debugging a failed or misbehaving LLM training run.
Pick exactly ONE diagnosis code and give the minimal fix: only the config fields that must change, with corrected values.
Diagnosis codes:
- OOM_BATCH_TOO_LARGE
- ACTIVATION_MEMORY_NO_CHECKPOINTING
- EFFECTIVE_BATCH_MISMATCH
- LR_TOO_HIGH_NO_WARMUP
- PRECISIO... | {"diagnosis": "BATCH_FROM_MEMORY_COMPONENTS", "fix": {"per_device_batch_size": 21}, "work": {"fixed_memory_gb": 48.0, "remaining_memory_gb": 32.0}} | {"archetype": "batch_from_components", "computations": ["R13", "R1"], "depth": 3, "interaction": null, "tier": "computed"} |
train-parallel_batch-039 | parallel_batch | train | You are debugging a failed or misbehaving LLM training run.
Pick exactly ONE diagnosis code and give the minimal fix: only the config fields that must change, with corrected values.
Diagnosis codes:
- OOM_BATCH_TOO_LARGE
- ACTIVATION_MEMORY_NO_CHECKPOINTING
- EFFECTIVE_BATCH_MISMATCH
- LR_TOO_HIGH_NO_WARMUP
- PRECISIO... | {"diagnosis": "PARALLEL_SHARDED_BATCH", "fix": {"per_device_batch_size": 64}, "work": {"fixed_memory_gb": 128.0, "params_per_rank_billions": 8.0}} | {"archetype": "parallel_batch", "computations": ["R22", "R13", "R1"], "depth": 4, "interaction": "parallelism x memory", "tier": "interacting"} |
train-zero_batch-064 | zero_batch | train | You are debugging a failed or misbehaving LLM training run.
Pick exactly ONE diagnosis code and give the minimal fix: only the config fields that must change, with corrected values.
Diagnosis codes:
- OOM_BATCH_TOO_LARGE
- ACTIVATION_MEMORY_NO_CHECKPOINTING
- EFFECTIVE_BATCH_MISMATCH
- LR_TOO_HIGH_NO_WARMUP
- PRECISIO... | {"diagnosis": "ZERO_SHARDED_BATCH_SIZE", "fix": {"per_device_batch_size": 41}, "work": {"bytes_per_param_per_rank": 8.5, "fixed_memory_gb": 34.0, "remaining_memory_gb": 62.0}} | {"archetype": "zero_batch", "computations": ["R19", "R13", "R1"], "depth": 4, "interaction": "zero x memory", "tier": "interacting"} |
train-two_v100_dataloader-013 | two_v100_dataloader | train | You are debugging a failed or misbehaving LLM training run.
Pick exactly ONE diagnosis code and give the minimal fix: only the config fields that must change, with corrected values.
Diagnosis codes:
- OOM_BATCH_TOO_LARGE
- ACTIVATION_MEMORY_NO_CHECKPOINTING
- EFFECTIVE_BATCH_MISMATCH
- LR_TOO_HIGH_NO_WARMUP
- PRECISIO... | {"diagnosis": ["PRECISION_UNSUPPORTED_ON_GPU", "DATALOADER_FORK_WITH_CUDA"], "fix": {"dataloader_multiprocessing_context": "spawn", "dtype": "fp16", "use_grad_scaler": true}} | {"archetype": "two_v100_dataloader", "computations": ["R5", "R9"], "depth": 2, "interaction": "mixed", "tier": "two_fault"} |
train-batch_from_components-035 | batch_from_components | train | You are debugging a failed or misbehaving LLM training run.
Pick exactly ONE diagnosis code and give the minimal fix: only the config fields that must change, with corrected values.
Diagnosis codes:
- OOM_BATCH_TOO_LARGE
- ACTIVATION_MEMORY_NO_CHECKPOINTING
- EFFECTIVE_BATCH_MISMATCH
- LR_TOO_HIGH_NO_WARMUP
- PRECISIO... | {"diagnosis": "BATCH_FROM_MEMORY_COMPONENTS", "fix": {"per_device_batch_size": 16}, "work": {"fixed_memory_gb": 64.0, "remaining_memory_gb": 16.0}} | {"archetype": "batch_from_components", "computations": ["R13", "R1"], "depth": 3, "interaction": null, "tier": "computed"} |
train-batch_from_components-017 | batch_from_components | train | You are debugging a failed or misbehaving LLM training run.
Pick exactly ONE diagnosis code and give the minimal fix: only the config fields that must change, with corrected values.
Diagnosis codes:
- OOM_BATCH_TOO_LARGE
- ACTIVATION_MEMORY_NO_CHECKPOINTING
- EFFECTIVE_BATCH_MISMATCH
- LR_TOO_HIGH_NO_WARMUP
- PRECISIO... | {"diagnosis": "BATCH_FROM_MEMORY_COMPONENTS", "fix": {"per_device_batch_size": 37}, "work": {"fixed_memory_gb": 24.0, "remaining_memory_gb": 56.0}} | {"archetype": "batch_from_components", "computations": ["R13", "R1"], "depth": 3, "interaction": null, "tier": "computed"} |
train-two_v100_dataloader-050 | two_v100_dataloader | train | You are debugging a failed or misbehaving LLM training run.
Pick exactly ONE diagnosis code and give the minimal fix: only the config fields that must change, with corrected values.
Diagnosis codes:
- OOM_BATCH_TOO_LARGE
- ACTIVATION_MEMORY_NO_CHECKPOINTING
- EFFECTIVE_BATCH_MISMATCH
- LR_TOO_HIGH_NO_WARMUP
- PRECISIO... | {"diagnosis": ["PRECISION_UNSUPPORTED_ON_GPU", "DATALOADER_FORK_WITH_CUDA"], "fix": {"dataloader_multiprocessing_context": "spawn", "dtype": "fp16", "use_grad_scaler": true}} | {"archetype": "two_v100_dataloader", "computations": ["R5", "R9"], "depth": 2, "interaction": "mixed", "tier": "two_fault"} |
train-checkpoint_retention-029 | checkpoint_retention | train | You are debugging a failed or misbehaving LLM training run.
Pick exactly ONE diagnosis code and give the minimal fix: only the config fields that must change, with corrected values.
Diagnosis codes:
- OOM_BATCH_TOO_LARGE
- ACTIVATION_MEMORY_NO_CHECKPOINTING
- EFFECTIVE_BATCH_MISMATCH
- LR_TOO_HIGH_NO_WARMUP
- PRECISIO... | {"diagnosis": "CHECKPOINT_RETENTION_LIMIT", "fix": {"max_checkpoints_to_keep": 1}, "work": {"checkpoint_size_gb": 112.0}} | {"archetype": "checkpoint_retention", "computations": ["R17"], "depth": 2, "interaction": null, "tier": "computed"} |
train-two_v100_dataloader-051 | two_v100_dataloader | train | You are debugging a failed or misbehaving LLM training run.
Pick exactly ONE diagnosis code and give the minimal fix: only the config fields that must change, with corrected values.
Diagnosis codes:
- OOM_BATCH_TOO_LARGE
- ACTIVATION_MEMORY_NO_CHECKPOINTING
- EFFECTIVE_BATCH_MISMATCH
- LR_TOO_HIGH_NO_WARMUP
- PRECISIO... | {"diagnosis": ["PRECISION_UNSUPPORTED_ON_GPU", "DATALOADER_FORK_WITH_CUDA"], "fix": {"dataloader_multiprocessing_context": "spawn", "dtype": "fp16", "use_grad_scaler": true}} | {"archetype": "two_v100_dataloader", "computations": ["R5", "R9"], "depth": 2, "interaction": "mixed", "tier": "two_fault"} |
train-shard_size-050 | shard_size | train | You are debugging a failed or misbehaving LLM training run.
Pick exactly ONE diagnosis code and give the minimal fix: only the config fields that must change, with corrected values.
Diagnosis codes:
- OOM_BATCH_TOO_LARGE
- ACTIVATION_MEMORY_NO_CHECKPOINTING
- EFFECTIVE_BATCH_MISMATCH
- LR_TOO_HIGH_NO_WARMUP
- PRECISIO... | {"diagnosis": "SHARD_SIZE_PER_RANK", "fix": {"examples_per_rank": 25000, "shard_size_gb": 60.0}} | {"archetype": "shard_size", "computations": ["R18"], "depth": 2, "interaction": null, "tier": "computed"} |
train-two_v100_dataloader-021 | two_v100_dataloader | train | You are debugging a failed or misbehaving LLM training run.
Pick exactly ONE diagnosis code and give the minimal fix: only the config fields that must change, with corrected values.
Diagnosis codes:
- OOM_BATCH_TOO_LARGE
- ACTIVATION_MEMORY_NO_CHECKPOINTING
- EFFECTIVE_BATCH_MISMATCH
- LR_TOO_HIGH_NO_WARMUP
- PRECISIO... | {"diagnosis": ["PRECISION_UNSUPPORTED_ON_GPU", "DATALOADER_FORK_WITH_CUDA"], "fix": {"dataloader_multiprocessing_context": "spawn", "dtype": "fp16", "use_grad_scaler": true}} | {"archetype": "two_v100_dataloader", "computations": ["R5", "R9"], "depth": 2, "interaction": "mixed", "tier": "two_fault"} |
train-two_sampler_accum-020 | two_sampler_accum | train | You are debugging a failed or misbehaving LLM training run.
Pick exactly ONE diagnosis code and give the minimal fix: only the config fields that must change, with corrected values.
Diagnosis codes:
- OOM_BATCH_TOO_LARGE
- ACTIVATION_MEMORY_NO_CHECKPOINTING
- EFFECTIVE_BATCH_MISMATCH
- LR_TOO_HIGH_NO_WARMUP
- PRECISIO... | {"diagnosis": ["DUPLICATE_DATA_NO_DISTRIBUTED_SAMPLER", "EFFECTIVE_BATCH_MISMATCH"], "fix": {"grad_accum_steps": 8, "sampler": "DistributedSampler"}} | {"archetype": "two_sampler_accum", "computations": ["R7", "R3"], "depth": 3, "interaction": "sampler+accum", "tier": "two_fault"} |
train-two_lora_vocab-039 | two_lora_vocab | train | You are debugging a failed or misbehaving LLM training run.
Pick exactly ONE diagnosis code and give the minimal fix: only the config fields that must change, with corrected values.
Diagnosis codes:
- OOM_BATCH_TOO_LARGE
- ACTIVATION_MEMORY_NO_CHECKPOINTING
- EFFECTIVE_BATCH_MISMATCH
- LR_TOO_HIGH_NO_WARMUP
- PRECISIO... | {"diagnosis": ["ADAPTERS_FROZEN_ZERO_TRAINABLE", "EMBEDDING_VOCAB_MISMATCH"], "fix": {"adapter_requires_grad": true, "embedding_vocab_size": 32064}} | {"archetype": "two_lora_vocab", "computations": ["R8", "R11"], "depth": 2, "interaction": "mixed", "tier": "two_fault"} |
train-optimizer_batch-063 | optimizer_batch | train | You are debugging a failed or misbehaving LLM training run.
Pick exactly ONE diagnosis code and give the minimal fix: only the config fields that must change, with corrected values.
Diagnosis codes:
- OOM_BATCH_TOO_LARGE
- ACTIVATION_MEMORY_NO_CHECKPOINTING
- EFFECTIVE_BATCH_MISMATCH
- LR_TOO_HIGH_NO_WARMUP
- PRECISIO... | {"diagnosis": "OPTIMIZER_MEMORY_BATCH", "fix": {"per_device_batch_size": 80}, "work": {"bytes_per_param": 12, "fixed_memory_gb": 48.0}} | {"archetype": "optimizer_batch", "computations": ["R21", "R1"], "depth": 3, "interaction": "optimizer x memory", "tier": "interacting"} |
train-oom_batch-007 | oom_batch | train | You are debugging a failed or misbehaving LLM training run.
Pick exactly ONE diagnosis code and give the minimal fix: only the config fields that must change, with corrected values.
Diagnosis codes:
- OOM_BATCH_TOO_LARGE
- ACTIVATION_MEMORY_NO_CHECKPOINTING
- EFFECTIVE_BATCH_MISMATCH
- LR_TOO_HIGH_NO_WARMUP
- PRECISIO... | {"diagnosis": "OOM_BATCH_TOO_LARGE", "fix": {"per_device_batch_size": 17}} | {"archetype": "oom_batch", "computations": ["R1"], "depth": 2, "interaction": null, "tier": "diagnosis"} |
train-batch_from_components-055 | batch_from_components | train | You are debugging a failed or misbehaving LLM training run.
Pick exactly ONE diagnosis code and give the minimal fix: only the config fields that must change, with corrected values.
Diagnosis codes:
- OOM_BATCH_TOO_LARGE
- ACTIVATION_MEMORY_NO_CHECKPOINTING
- EFFECTIVE_BATCH_MISMATCH
- LR_TOO_HIGH_NO_WARMUP
- PRECISIO... | {"diagnosis": "BATCH_FROM_MEMORY_COMPONENTS", "fix": {"per_device_batch_size": 5}, "work": {"fixed_memory_gb": 32.0, "remaining_memory_gb": 8.0}} | {"archetype": "batch_from_components", "computations": ["R13", "R1"], "depth": 3, "interaction": null, "tier": "computed"} |
train-shard_size-068 | shard_size | train | You are debugging a failed or misbehaving LLM training run.
Pick exactly ONE diagnosis code and give the minimal fix: only the config fields that must change, with corrected values.
Diagnosis codes:
- OOM_BATCH_TOO_LARGE
- ACTIVATION_MEMORY_NO_CHECKPOINTING
- EFFECTIVE_BATCH_MISMATCH
- LR_TOO_HIGH_NO_WARMUP
- PRECISIO... | {"diagnosis": "SHARD_SIZE_PER_RANK", "fix": {"examples_per_rank": 25000, "shard_size_gb": 40.0}} | {"archetype": "shard_size", "computations": ["R18"], "depth": 2, "interaction": null, "tier": "computed"} |
train-optimizer_batch-066 | optimizer_batch | train | You are debugging a failed or misbehaving LLM training run.
Pick exactly ONE diagnosis code and give the minimal fix: only the config fields that must change, with corrected values.
Diagnosis codes:
- OOM_BATCH_TOO_LARGE
- ACTIVATION_MEMORY_NO_CHECKPOINTING
- EFFECTIVE_BATCH_MISMATCH
- LR_TOO_HIGH_NO_WARMUP
- PRECISIO... | {"diagnosis": "OPTIMIZER_MEMORY_BATCH", "fix": {"per_device_batch_size": 26}, "work": {"bytes_per_param": 8, "fixed_memory_gb": 56.0}} | {"archetype": "optimizer_batch", "computations": ["R21", "R1"], "depth": 3, "interaction": "optimizer x memory", "tier": "interacting"} |
train-two_v100_dataloader-016 | two_v100_dataloader | train | You are debugging a failed or misbehaving LLM training run.
Pick exactly ONE diagnosis code and give the minimal fix: only the config fields that must change, with corrected values.
Diagnosis codes:
- OOM_BATCH_TOO_LARGE
- ACTIVATION_MEMORY_NO_CHECKPOINTING
- EFFECTIVE_BATCH_MISMATCH
- LR_TOO_HIGH_NO_WARMUP
- PRECISIO... | {"diagnosis": ["PRECISION_UNSUPPORTED_ON_GPU", "DATALOADER_FORK_WITH_CUDA"], "fix": {"dataloader_multiprocessing_context": "spawn", "dtype": "fp16", "use_grad_scaler": true}} | {"archetype": "two_v100_dataloader", "computations": ["R5", "R9"], "depth": 2, "interaction": "mixed", "tier": "two_fault"} |
train-trap_nan_clipping-008 | trap_nan_clipping | train | You are debugging a failed or misbehaving LLM training run.
Pick exactly ONE diagnosis code and give the minimal fix: only the config fields that must change, with corrected values.
Diagnosis codes:
- OOM_BATCH_TOO_LARGE
- ACTIVATION_MEMORY_NO_CHECKPOINTING
- EFFECTIVE_BATCH_MISMATCH
- LR_TOO_HIGH_NO_WARMUP
- PRECISIO... | {"diagnosis": "NO_GRADIENT_CLIPPING", "fix": {"max_grad_norm": 1.0}} | {"archetype": "trap_nan_clipping", "computations": ["R4", "R12"], "depth": 2, "interaction": "lr", "tier": "trap"} |
train-checkpoint_retention-012 | checkpoint_retention | train | You are debugging a failed or misbehaving LLM training run.
Pick exactly ONE diagnosis code and give the minimal fix: only the config fields that must change, with corrected values.
Diagnosis codes:
- OOM_BATCH_TOO_LARGE
- ACTIVATION_MEMORY_NO_CHECKPOINTING
- EFFECTIVE_BATCH_MISMATCH
- LR_TOO_HIGH_NO_WARMUP
- PRECISIO... | {"diagnosis": "CHECKPOINT_RETENTION_LIMIT", "fix": {"max_checkpoints_to_keep": 20}, "work": {"checkpoint_size_gb": 24.0}} | {"archetype": "checkpoint_retention", "computations": ["R17"], "depth": 2, "interaction": null, "tier": "computed"} |
train-batch_from_components-034 | batch_from_components | train | You are debugging a failed or misbehaving LLM training run.
Pick exactly ONE diagnosis code and give the minimal fix: only the config fields that must change, with corrected values.
Diagnosis codes:
- OOM_BATCH_TOO_LARGE
- ACTIVATION_MEMORY_NO_CHECKPOINTING
- EFFECTIVE_BATCH_MISMATCH
- LR_TOO_HIGH_NO_WARMUP
- PRECISIO... | {"diagnosis": "BATCH_FROM_MEMORY_COMPONENTS", "fix": {"per_device_batch_size": 16}, "work": {"fixed_memory_gb": 64.0, "remaining_memory_gb": 16.0}} | {"archetype": "batch_from_components", "computations": ["R13", "R1"], "depth": 3, "interaction": null, "tier": "computed"} |
train-lr_at_step-060 | lr_at_step | train | You are debugging a failed or misbehaving LLM training run.
Pick exactly ONE diagnosis code and give the minimal fix: only the config fields that must change, with corrected values.
Diagnosis codes:
- OOM_BATCH_TOO_LARGE
- ACTIVATION_MEMORY_NO_CHECKPOINTING
- EFFECTIVE_BATCH_MISMATCH
- LR_TOO_HIGH_NO_WARMUP
- PRECISIO... | {"diagnosis": "LR_VALUE_AT_STEP", "fix": {"current_learning_rate": 0.000195}, "work": {"step_fraction": 0.75}} | {"archetype": "lr_at_step", "computations": ["R15"], "depth": 2, "interaction": null, "tier": "computed"} |
train-accum_steps_lr-008 | accum_steps_lr | train | You are debugging a failed or misbehaving LLM training run.
Pick exactly ONE diagnosis code and give the minimal fix: only the config fields that must change, with corrected values.
Diagnosis codes:
- OOM_BATCH_TOO_LARGE
- ACTIVATION_MEMORY_NO_CHECKPOINTING
- EFFECTIVE_BATCH_MISMATCH
- LR_TOO_HIGH_NO_WARMUP
- PRECISIO... | {"diagnosis": "ACCUM_STEPS_AND_LR", "fix": {"current_learning_rate": 9.75e-05, "grad_accum_steps": 2, "planned_total_steps": 400}, "work": {"audit_step": 300, "steps_per_epoch": 200}} | {"archetype": "accum_steps_lr", "computations": ["R3", "R14", "R15"], "depth": 5, "interaction": "accum x schedule", "tier": "interacting"} |
train-training_time-038 | training_time | train | You are debugging a failed or misbehaving LLM training run.
Pick exactly ONE diagnosis code and give the minimal fix: only the config fields that must change, with corrected values.
Diagnosis codes:
- OOM_BATCH_TOO_LARGE
- ACTIVATION_MEMORY_NO_CHECKPOINTING
- EFFECTIVE_BATCH_MISMATCH
- LR_TOO_HIGH_NO_WARMUP
- PRECISIO... | {"diagnosis": "TRAINING_TIME_ESTIMATE", "fix": {"planned_time_hours": 1.7066666666666668}, "work": {"total_tokens": 245760000}} | {"archetype": "training_time", "computations": ["R16"], "depth": 3, "interaction": null, "tier": "computed"} |
train-zero_batch-041 | zero_batch | train | You are debugging a failed or misbehaving LLM training run.
Pick exactly ONE diagnosis code and give the minimal fix: only the config fields that must change, with corrected values.
Diagnosis codes:
- OOM_BATCH_TOO_LARGE
- ACTIVATION_MEMORY_NO_CHECKPOINTING
- EFFECTIVE_BATCH_MISMATCH
- LR_TOO_HIGH_NO_WARMUP
- PRECISIO... | {"diagnosis": "ZERO_SHARDED_BATCH_SIZE", "fix": {"per_device_batch_size": 48}, "work": {"bytes_per_param_per_rank": 2.0, "fixed_memory_gb": 32.0, "remaining_memory_gb": 96.0}} | {"archetype": "zero_batch", "computations": ["R19", "R13", "R1"], "depth": 4, "interaction": "zero x memory", "tier": "interacting"} |
train-parallel_batch-020 | parallel_batch | train | You are debugging a failed or misbehaving LLM training run.
Pick exactly ONE diagnosis code and give the minimal fix: only the config fields that must change, with corrected values.
Diagnosis codes:
- OOM_BATCH_TOO_LARGE
- ACTIVATION_MEMORY_NO_CHECKPOINTING
- EFFECTIVE_BATCH_MISMATCH
- LR_TOO_HIGH_NO_WARMUP
- PRECISIO... | {"diagnosis": "PARALLEL_SHARDED_BATCH", "fix": {"per_device_batch_size": 25}, "work": {"fixed_memory_gb": 64.0, "params_per_rank_billions": 4.0}} | {"archetype": "parallel_batch", "computations": ["R22", "R13", "R1"], "depth": 4, "interaction": "parallelism x memory", "tier": "interacting"} |
train-checkpoint_retention-036 | checkpoint_retention | train | You are debugging a failed or misbehaving LLM training run.
Pick exactly ONE diagnosis code and give the minimal fix: only the config fields that must change, with corrected values.
Diagnosis codes:
- OOM_BATCH_TOO_LARGE
- ACTIVATION_MEMORY_NO_CHECKPOINTING
- EFFECTIVE_BATCH_MISMATCH
- LR_TOO_HIGH_NO_WARMUP
- PRECISIO... | {"diagnosis": "CHECKPOINT_RETENTION_LIMIT", "fix": {"max_checkpoints_to_keep": 31}, "work": {"checkpoint_size_gb": 24.0}} | {"archetype": "checkpoint_retention", "computations": ["R17"], "depth": 2, "interaction": null, "tier": "computed"} |
train-two_oom_clipping-048 | two_oom_clipping | train | You are debugging a failed or misbehaving LLM training run.
Pick exactly ONE diagnosis code and give the minimal fix: only the config fields that must change, with corrected values.
Diagnosis codes:
- OOM_BATCH_TOO_LARGE
- ACTIVATION_MEMORY_NO_CHECKPOINTING
- EFFECTIVE_BATCH_MISMATCH
- LR_TOO_HIGH_NO_WARMUP
- PRECISIO... | {"diagnosis": ["OOM_BATCH_TOO_LARGE", "NO_GRADIENT_CLIPPING"], "fix": {"max_grad_norm": 1.0, "per_device_batch_size": 12}} | {"archetype": "two_oom_clipping", "computations": ["R1", "R12"], "depth": 3, "interaction": "oom+clipping", "tier": "two_fault"} |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.