{ "wikitext2_token_ppl": 29.277406271326807, "wikitext2_byte_ppl": 2.2718694747091326, "bytes_per_token": 4.115042105665474, "checkpoint_path": "/root/slayer149-balanced/runs/arcft-balanced/epoch-02.pt", "checkpoint_sha256": "79e4d06b1b95408c4b47e6d9fab34ed2a26609997b80d45698e6ec23db8724e8", "step": 1118, "training_tokens": null, "parameters": 149333081, "torch": "2.14.0+cu130", "gpu": "NVIDIA GeForce RTX 3090", "dataset_revisions": { "Salesforce/wikitext": "b08601e04326c79dfdd32d625aee71d232d685c3", "nyu-mll/blimp": "877fba0801ffb7cbd8c39c1ff314a46f053f6036", "allenai/ai2_arc": "210d026faf9955653af8916fad021475a3f00453" }, "recipe_sha256": "4a4ab81f9a749b04cf476672c0967512ec951a77947dea02fec3d77bf13fa091", "elapsed_seconds": 15.584210969973356, "protocol": "GLINT-derived pinned harness; CUDA BF16 forward, FP32 logprobs, context 256, no BOS, raw answer LL", "dataset_files": { "wikitext-2-raw-v1": [ "wikitext-2-raw-v1/test-00000-of-00001.parquet" ] } }