RESERVED Session 1 | 20,000 examples | crash-safe reservation

README.md

@@ -0,0 +1,102 @@
+---
+license: mit
+library_name: transformers
+tags:
+- causal-lm
+- decoder-only
+- reasoning
+- deep-reasoning
+- recurrent-transformer
+- adaptive-computation
+- chain-of-thought
+- adaptive-reasoning
+---
+
+# Nano.Deep.Reasoner.11m-HyperMini
+
+An approximately 11,094,003-parameter decoder-only
+adaptive recurrent reasoning language model.
+
+## Architecture
+
+- Parameters: 11,094,003
+- Context: 1096
+- Vocabulary: 16,000
+- Hidden size: 240
+- Base Transformer blocks: 6
+- Attention heads: 8
+- Head dimension: 30
+- Intermediate size: 1072
+- Shared recurrent reasoning block
+- Learned latent reasoning memory: 8 tokens
+- Adaptive reasoning depth: 2-16
+- Gated recurrent memory updates
+- Adaptive halting controller
+- Verification head
+- Revision head
+- RoPE
+- Tied input/output embeddings
+- Padding-aware causal attention
+- Explicit padded-state suppression
+
+## Reasoning tokens
+
+<think>
+</think>
+
+<thought>
+</thought>
+
+<reasoning>
+</reasoning>
+
+<answer>
+</answer>
+
+## Training
+
+Dataset:
+
+Plans11/Organized_PreTrain_1k_Context
+
+Each session contains up to 20,000 NEW examples.
+
+Examples are protected by SHA-256 hashes.
+
+Session reservations are committed before training so a hard
+Kaggle interruption cannot cause the same reserved examples
+to be selected again.
+
+## Resume safety
+
+The checkpoint contains:
+
+- model.safetensors
+- optimizer.pt
+- rng_state.pt
+- training_state.json
+- example_ledger.json
+- tokenizer.json
+- tokenizer_config.json
+- token_id_manifest.json
+- config.json
+
+Dataset fingerprint and tokenizer artifact hashes are verified
+before resume.
+
+## Current state
+
+Completed sessions:
+0
+
+Unique examples reserved/trained:
+20,000
+
+Unique completed examples:
+0
+
+Global optimizer steps:
+0
+
+Last session loss:
+None

config.json

@@ -0,0 +1,48 @@
+{
+  "adaptive_halting": true,
+  "architectures": [
+    "HyperMiniReasoner"
+  ],
+  "base_layers": 6,
+  "context_length": 1096,
+  "dataset": "Plans11/Organized_PreTrain_1k_Context",
+  "head_dimension": 30,
+  "hidden_size": 240,
+  "intermediate_size": 1072,
+  "max_position_embeddings": 1096,
+  "max_reasoning_steps": 16,
+  "memory_tokens": 8,
+  "min_reasoning_steps": 2,
+  "model_name": "Nano.Deep.Reasoner.11m-HyperMini",
+  "model_type": "nano_deep_reasoner_hypermini",
+  "num_attention_heads": 8,
+  "padding_aware_attention": true,
+  "padding_query_zeroing": true,
+  "parameter_count": 11094003,
+  "reasoning_tokens": [
+    "<pad>",
+    "<bos>",
+    "<eos>",
+    "<unk>",
+    "<think>",
+    "</think>",
+    "<thought>",
+    "</thought>",
+    "<reasoning>",
+    "</reasoning>",
+    "<answer>",
+    "</answer>",
+    "<question>",
+    "</question>",
+    "<analysis>",
+    "</analysis>",
+    "<tool>",
+    "</tool>"
+  ],
+  "revision_head": true,
+  "rope": true,
+  "shared_recurrent_reasoner": true,
+  "tie_word_embeddings": true,
+  "verification_head": true,
+  "vocab_size": 16000
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7395c7c3b61893cadfa369d36f78a12f61c3c844856821b5cd88ccf24b9b0971
+size 59742900

optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d493a143ff47266f40b916d678f4bce6de34ba3c9592c59d1f92078ab1b7c9de
+size 1721

rng_state.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:55c3041a7be4d734bf7df211c94ad7620ecd54feff1b53e73c11c884179f2879
+size 10931

token_id_manifest.json

@@ -0,0 +1,31 @@
+{
+  "artifact_sha256": {
+    "tokenizer.json": "9619297bcfd3ea8f40d56f991e214d166b7c3bf45aee8cabb908881adeaf6ff9",
+    "tokenizer_config.json": "2e82dbc3e20a656b022331e8b616deb37897c4544b5c4489a598793cae589c91"
+  },
+  "immutable": true,
+  "reasoning_token_ids": {
+    "</analysis>": 15,
+    "</answer>": 11,
+    "</question>": 13,
+    "</reasoning>": 9,
+    "</think>": 5,
+    "</thought>": 7,
+    "</tool>": 17,
+    "<analysis>": 14,
+    "<answer>": 10,
+    "<question>": 12,
+    "<reasoning>": 8,
+    "<think>": 4,
+    "<thought>": 6,
+    "<tool>": 16
+  },
+  "special_token_ids": {
+    "<bos>": 1,
+    "<eos>": 2,
+    "<pad>": 0,
+    "<unk>": 3
+  },
+  "tokenizer_version": 3,
+  "vocab_size": 16000
+}

tokenizer_config.json

@@ -0,0 +1,9 @@
+{
+  "backend": "tokenizers",
+  "bos_token": "<bos>",
+  "eos_token": "<eos>",
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<pad>",
+  "tokenizer_class": "TokenizersBackend",
+  "unk_token": "<unk>"
+}