Feat/intragen experiments (#414)

* added curriculum

* readapted readme

* corrected small errors

* Delete eval/eval/r1/algorithmic/word_sorting.json

* removed redundant argument

* added spell

* removed duplicated fit

* changed config

* added composite changes

* added composite changes

* updated yaml

* added spell backward

* updated read me

* added qwen2.5

* added

* Add files via upload

* updated missing trainer func

* updated curr

* updated spell back

* updated correctness score func

* updated configs

* added local evals

* added updates

* updated datasets

* added fsdp to hf utility

* added algorithmic qwen 3b yaml

* updated read me

* updated configs

* added preappend token

* updated with thinking token

* updated test score board

* resolved comments

* added evaluation scripts

* removed results from pr

* added config

* added partial reward scoring

* added evaluation composites

* added training configs

* added games eval

* added rubriks cube

* resolved merge cinflicts

* added games config

* added latest eval configs

* updated strucutre

* Delete training/evaluations/eval_graphs_composite.yaml

---------

Co-authored-by: joesharratt1229 <joesharrat1229@gmail.com>

training/evaluations/eval_cognition_composite.yaml

@@ -0,0 +1,36 @@
+# Model configuration
+model_path: ../utils/qwen3b_cognition
+max_tokens: 1024
+temperature: 0.6  # Lower temperature for more focused responses
+top_p: 0.9  # From rollout top_p
+developer_prompt: DeepSeekZero
+developer_role: system  # Standard role for system prompts
+
+# Output configuration
+output_dir: results
+save_metadata: true
+save_full_results: true
+eval_repeats: 3
+
+# Categories and datasets to evaluate
+categories:
+  - category: reasoning
+    datasets:
+    - dataset: number_sequence
+      size: 100
+      seed: 42
+      params:
+        min_terms: 4  # Minimum visible terms
+        max_terms:  8  # Maximum visible terms
+        min_value: -100  # Minimum allowed number
+        max_value: 100  # Maximum allowed number
+        max_complexity: 3  # Maximum number of operations to combine
+    - dataset: modulo_grid
+      size: 100
+      seed: 42
+      params:
+        size_x: 20
+        size_y: 20
+        max_divisor: 20
+        max_target: 20
+        max_holes: 1