S-SALAAD teacher logits & student models

This repository hosts open artifacts for the SALAAD compression project.

S-SALAAD/
├── logits/       teacher-logit dataset used for offline pre-training distillation
└── models/       student checkpoints (added as they are released)

logits/ — offline pre-training distillation data

Generated by salaadpp.generate_teacher_logits. Each chunk stores the top-K teacher logprobs plus the gold token's logprob for every position of every packed sequence — enough to compute KL-based distillation losses offline without ever re-running the teacher.

Teacher	nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-NVFP4
Tokenizer	nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-NVFP4
Source corpus	HuggingFaceFW/fineweb-edu (config sample-100BT, revision 87f09149ef47)
Sequence length	512
Top-K logprobs	10
Document separator id	1
Chunks	539
Rows (sequences)	67,004,100
Total tokens	34,306,099,200

Per-chunk schema (SafeTensors)

Tensor	Dtype	Shape
input_ids	int32	[R, T]
top_idx	int32	[R, T, K]
top_logprob	float16	[R, T, K]
gold_logprob	float16	[R, T]

where R = num_rows_in_chunk, T = 512, K = 10.

Each *.safetensors file has a sibling *.meta.json with the source shard, chunk index, and the header fields above. logits/manifest.json lists every committed chunk once the run is finalized.

Reading a chunk

import json, pathlib
from huggingface_hub import snapshot_download
from safetensors import safe_open

root = pathlib.Path(snapshot_download(
    "egretwAlker/S-SALAAD", repo_type="dataset",
    allow_patterns=["logits/*"],
))
manifest = json.loads((root / "logits" / "manifest.json").read_text())
for stem in manifest["chunks"]:
    with safe_open(root / "logits" / f"{stem}.safetensors", framework="pt") as f:
        input_ids    = f.get_tensor("input_ids")     # int32 [R, T]
        top_idx      = f.get_tensor("top_idx")       # int32 [R, T, K]
        top_logprob  = f.get_tensor("top_logprob")   # fp16  [R, T, K]
        gold_logprob = f.get_tensor("gold_logprob")  # fp16  [R, T]

models/ — student checkpoints

Student models trained against these logits are added as they are released, each in its own subdirectory (models/<name>/config.json, model.safetensors, tokenizer.json, …). Load with:

from huggingface_hub import snapshot_download
from transformers import AutoModelForCausalLM, AutoTokenizer

path = snapshot_download("egretwAlker/S-SALAAD", repo_type="dataset",
                         allow_patterns=["models/<name>/*"])
tok  = AutoTokenizer.from_pretrained(f"{path}/models/<name>")
model = AutoModelForCausalLM.from_pretrained(f"{path}/models/<name>")

Licensing

The teacher logits are derivative of both the source corpus and the teacher model, and inherit the terms of both:

• Source corpus (HuggingFaceFW/fineweb-edu): ODC-BY 1.0.
• Teacher (nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-NVFP4): see the teacher's model card on the Hub.

Downstream users must comply with both. Student checkpoints under models/ carry their own per-model licensing noted in each subdirectory's README.