# Cron eval result #3 for Nanthasit/sakthai-context-1.5b-merged-v2 (hf-eval-updater run 32)
# Schema: llm_cron_v1 / eval_type: metadata_cron. Metadata-based snapshot, no inference run.
# Data: HF API + config.json + existing .eval_results/, 2026-08-01.

target_model:
  id: Nanthasit/sakthai-context-1.5b-merged-v2
  pipeline_tag: text-generation
  library_name: transformers
  base_model: Qwen/Qwen2.5-1.5B-Instruct
  downloads: 337
  likes: 0
  private: false
  gated: false
  created: "2026-07-30T10:46:30+00:00"
  last_modified: "2026-07-31T19:49:44+00:00"
  model_age_days: 2.003
  model_type: llm
  has_weights: true

architecture:
  model_type: qwen2
  architectures:
    - Qwen2ForCausalLM
  hidden_size: 1536
  num_hidden_layers: 28
  num_attention_heads: 12
  num_key_value_heads: 2
  intermediate_size: 8960
  vocab_size: 151936
  max_position_embeddings: 32768
  rope_theta: 1000000.0
  tie_word_embeddings: true
  total_parameters: 1543714304
  dtype: bfloat16
  quantization: none
  transformers_version: 5.14.1
  arch_source: "config.json fetched live 2026-08-01"

repo_summary:
  siblings_count: 20
  total_repo_bytes: 3098928651
  total_gb: 2.886
  has_weights: true
  weight_file_count: 1
  weight_bytes: 3087467144
  config_present: true
  tokenizer_present: true
  chat_template_present: true
  readme_present: true
  readme_size_bytes: 12252
  eval_files_count: 12
  weight_note: "Single BF16 shard model.safetensors (2.875 GiB)."

benchmarks:
  model_index_count: 1
  metrics_count: 3
  all_verified: false
  pending_metrics: 0
  entries:
    - dataset: Nanthasit/sakthai-bench-v2
      task: text-generation
      metrics:
        - name: Selection Accuracy
          value: 34.9
          verified: false
        - name: Arguments Accuracy
          value: 44.2
          verified: false
        - name: Strict Accuracy
          value: 34.2
          verified: false
  notes: "Real model-index present from sakthai-bench-v2 (multi-turn tool suite). Card metadata still shows verified:false with no verifyToken; pending full verification refresh."

card_quality:
  license: apache-2.0
  base_model_documented: true
  base_model: Qwen/Qwen2.5-1.5B-Instruct
  datasets_count: 3
  datasets:
    - Nanthasit/sakthai-combined-v6
    - Nanthasit/sakthai-combined-v7
    - Nanthasit/sakthai-irrelevance-supplement
  tags_count: 25
  model_index_present: true
  readme_size_bytes: 12252
  score: 90

health_score:
  overall: 62.4
  components:
    popularity: 12.3
    momentum: 43.6
    benchmarks: 50.0
    card_quality: 90.0
    repo_hygiene: 100.0
  weights:
    popularity: 0.20
    momentum: 0.20
    benchmarks: 0.25
    card_quality: 0.20
    repo_hygiene: 0.15
  formula_note: "0.20*12.3 + 0.20*43.6 + 0.25*50.0 + 0.20*90.0 + 0.15*100.0 = 62.42"

sibling_comparison:
  rank_by_downloads: 8
  total_author_models: 20
  models_with_positive_downloads: 19
  max_sibling_downloads: 1855
  velocity_note: "Daily velocity computed from download delta since first tracked 2026-07-30."

eval_metadata:
  model: Nanthasit/sakthai-context-1.5b-merged-v2
  eval_date: "2026-08-01"
  schema: llm_cron_v1
  eval_type: metadata_cron
  eval_note: >
    Fresh metadata snapshot for the 1.5B merged-v2 checkpoint. No inference
    rerun in cron env. Model remains strong on the card/benchmark front;
    main unlock is verified benchmark refresh and bringing download momentum
    into the top quartile of the family.