Model page

MiniLLM/teacher-gpt2-1.5B warn

Weights only ship in a format that can run code when loaded; its license differs from its base model's. Plus 1 minor note.

downloads 508likes 3license apache-2.0arch gpt2updated 2024-09-26

claims base: openai-community/gpt2-xl · chat template: not found · view on Hugging Face ↗

Scan coverageStatic battery2026-08-25Weights battery2026-08-25Behavioral batterynot rundetails
BatteryLooks atStatus
Static batteryMetadata & packagingcomplete 2026-08-25
Weights batteryWeights forensics — no GPU, no downloadcomplete 2026-08-2550,257-token embedding scanned · 0 undertrained · lineage consistent · pickle audit clean
Behavioral batteryLive-inference differentialsnot run

Ingot runs three batteries against a model. What each one checks →

Findings

Scanned 2026-08-25 · published from a community scan.

medium Pickle-serialized weights, no safetensors

Weights ship only as pickle-based files (pytorch_model-00001-of-00002.bin, pytorch_model-00002-of-00002.bin). Loading pickle executes arbitrary code from the file — prefer a safetensors release or load in a sandbox.

How to fix

Convert the weights to safetensors before loading them anywhere that matters.

  1. Do not load the pickle files in-process — pickle deserialization executes arbitrary code from the file.
  2. Convert locally in a sandbox: `pip install safetensors` and use `safetensors.torch.save_file` on a state dict loaded with `torch.load(..., weights_only=True)` (refuses most code-execution payloads), or use Hugging Face's `convert.py` space/script.
  3. Pin the exact revision hash you converted from, and load only your converted safetensors artifact from then on.

low Padding token is the EOS token

The pad token and the (only) EOS token are the same. Fine-tuning frameworks mask pad positions out of the loss, so training on this checkpoint teaches the model to never emit EOS — the Phi-4 / Qwen 2.5 / DeepSeek R1 infinite-generation bug. Safe to serve, hazardous to fine-tune; repoint pad_token at a dedicated token first.

How to fixingot patch

Align the stop-token declarations — a pure metadata fix to `generation_config.json` (and `config.json`).

  1. Identify the token the chat template actually ends assistant turns with (e.g. `<|eot_id|>`, `<end_of_turn>`, `<|im_end|>`) and make sure its id is in `generation_config.json`'s `eos_token_id` list.
  2. Keep `config.json`'s `eos_token_id` consistent with (or a subset of) `generation_config.json`'s — runtimes differ in which file they read.
  3. For the pad-equals-EOS hazard: repoint `pad_token` at a dedicated padding token before fine-tuning; serving is unaffected.
  4. Until the repo is fixed, pass explicit stop tokens to your serving stack (e.g. vLLM `stop_token_ids`, llama.cpp `--override-kv tokenizer.ggml.eos_token_id`).

medium License differs from claimed parent (apache-2.0 vs mit)

This model declares apache-2.0 while its claimed base openai-community/gpt2-xl declares mit. Verify the re-license is permitted before commercial use.

How to fix

Verify the re-license is actually permitted before relying on it.

  1. Read the parent's license for derivative-work and re-licensing terms — many open-weight licenses (e.g. Llama-family) do not permit arbitrary re-licensing.
  2. If the re-license is not permitted, the parent's terms govern your use regardless of what this repo declares.

info Embedding-norm glitch scan clean

No undertrained tokens found: every non-special token's embedding norm is above 0.3× the vocabulary median (1.902). The glitch-token data-corruption class has no candidate surface in this model.

info Pickle static analysis clean

Opcode-level parse of pytorch_model-00001-of-00002.bin, pytorch_model-00002-of-00002.bin (no code executed) found only standard serialization globals (3 distinct, all torch/collections/numpy). Pickle remains an executable format — this verifies the current bytes, not future uploads; prefer a safetensors release.

info Weights consistent with claimed parent openai-community/gpt2-xl

Mean cosine similarity of 64 sampled token-embedding rows against openai-community/gpt2-xl is 0.995 — the weights plausibly descend from the declared base (relation: unspecified).

How to fix

Fix or verify the `base_model` declaration so lineage checks can run.

  1. If you own the repo: correct the `base_model` field in the model card metadata to the real, public parent.
  2. If you don't: identify the true parent (config architecture + weight shapes narrow it fast) and re-scan with that lineage in mind.

Fix it

Some findings are metadata-level and patchable — apply the fixes to your local copy (your weights never leave your machine):

npx @ingotai/scan patch MiniLLM/teacher-gpt2-1.5B

Remediation guidance addresses the documented findings only. It is evidence-driven repair, not a safety certification of the model.

Fingerprint

The durable profile of this model: measured weights-and-metadata facts, rebuilt on every scan and battery run. Updated 2026-08-25.

architecturegpt2 · 48 layers · 1600-dim
vocabulary50,257 tokens
licenseapache-2.0
serializationno safetensors pickle
chat templatenone
claimed lineageopenai-community/gpt2-xl
lineage verifiedconsistent vs openai-community/gpt2-xl — embedding-row cosine 0.995
glitch-token surfaceclean no undertrained tokens
Full measured fingerprint
architecturesGPT2LMHeadModel
pipelinetext-generation
repo files12 — pickle: pytorch_model-00001-of-00002.bin, pytorch_model-00002-of-00002.bin
revision4a8d07d52ad6
HF snapshot508 downloads · 3 likes · updated 2024-09-26 · captured 2026-08-25
pickle auditpytorch_model-00001-of-00002.bin, pytorch_model-00002-of-00002.bin3 standard global(s)
embedding tensortransformer.wte.weight · F32 · 50,257×1600
embedding normsmedian 1.9016 · mean 1.9028
lineage checkconsistent — cosine 0.9948 over 64 sampled rows vs openai-community/gpt2-xl
Battery runs (1)the run trace behind the findings — what each job measured
batterystatusqueueddurationattempts
weightscomplete2026-08-25 21:4846s1
weights run 2026-08-25 measurements
probes runglitch-norm-scan, zero-template-token-scan, pickle-static-analysis, lineage-norm-correlation
embedding tensortransformer.wte.weight · F32 · 50,257×1600
glitch surface0 undertrained, 0 plain-ASCII
lineage checkconsistent — cosine 0.9948 over 64 rows vs openai-community/gpt2-xl

Verdict badge

Ship the verdict in your README — it always shows the latest published analysis:

Ingot verdict: warn

[![Ingot scan](https://ingot.tools/api/v1/models/MiniLLM/teacher-gpt2-1.5B/badge.svg)](https://ingot.tools/models/MiniLLM/teacher-gpt2-1.5B)
Gate it in CI