Manuscript revision and pending experiment work, snapshot before restructuring
Clarity pass over the main text (36-item audit), Discussion rewrite and cut, acknowledgements, Souly et al. as ref 62, lettered SI panels, model section moved under Results; plus the untracked curriculum/society/compose/smol configs, runners, figures, stats and tests that the SI already cites. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01Y64o8FKP7rCuXzC48pxpMm
This commit is contained in:
parent
e4804adabc
commit
84124de143
450 changed files with 52813 additions and 1202 deletions
5
results/llm_curriculum_v5_early_obl/README.md
Normal file
5
results/llm_curriculum_v5_early_obl/README.md
Normal file
|
|
@ -0,0 +1,5 @@
|
|||
# llm_curriculum_v5_early_obl
|
||||
|
||||
One of the four conflict-arrival curriculum bundles; the legend, numbers and reading for all four are in
|
||||
`results/llm_curriculum_v5_early/README.md` (figure `curriculum_timing.png` there). Config
|
||||
`configs/llm/curriculum_v5_early_obl.yaml`; seeds via `hpc/llm_curriculum_timing.pbs`.
|
||||
74
results/llm_curriculum_v5_early_obl/s1/manifest.json
Normal file
74
results/llm_curriculum_v5_early_obl/s1/manifest.json
Normal file
|
|
@ -0,0 +1,74 @@
|
|||
{
|
||||
"experiment": "llm_curriculum_v5_early_obl",
|
||||
"master_seed": 1,
|
||||
"git_commit": null,
|
||||
"python": "3.11.7",
|
||||
"libraries": {
|
||||
"numpy": "2.4.6",
|
||||
"scipy": "1.17.1",
|
||||
"pandas": "3.0.3",
|
||||
"pyarrow": "24.0.0",
|
||||
"torch": "2.12.1",
|
||||
"transformers": "5.16.1",
|
||||
"peft": "0.20.0"
|
||||
},
|
||||
"rows": 205,
|
||||
"results_sha256": "9e9cffa0a9c6b8ef5cb62563acce24195f062f984d1eb2fa54d91ae5e75101a3",
|
||||
"layer": "2",
|
||||
"tier": "llm",
|
||||
"base_model": "Qwen/Qwen2.5-1.5B",
|
||||
"hard": false,
|
||||
"curriculum": {
|
||||
"families": [
|
||||
"mnli",
|
||||
"arc",
|
||||
"hellaswag",
|
||||
"squad",
|
||||
"boolq",
|
||||
"winogrande"
|
||||
],
|
||||
"lineages": 3,
|
||||
"generations": 6,
|
||||
"arms": [
|
||||
"society"
|
||||
],
|
||||
"baselines": [],
|
||||
"n_new": 300,
|
||||
"n_replay": 150,
|
||||
"operator": "linear",
|
||||
"ancestor_depth": null,
|
||||
"lora": {
|
||||
"r": 16,
|
||||
"alpha": 32
|
||||
},
|
||||
"allow_veto": false,
|
||||
"merge_until": null,
|
||||
"orders": [
|
||||
[
|
||||
"boolq",
|
||||
"winogrande",
|
||||
"mnli",
|
||||
"arc",
|
||||
"hellaswag",
|
||||
"squad"
|
||||
],
|
||||
[
|
||||
"winogrande",
|
||||
"boolq",
|
||||
"arc",
|
||||
"hellaswag",
|
||||
"squad",
|
||||
"mnli"
|
||||
],
|
||||
[
|
||||
"boolq",
|
||||
"winogrande",
|
||||
"hellaswag",
|
||||
"squad",
|
||||
"mnli",
|
||||
"arc"
|
||||
]
|
||||
],
|
||||
"cull": null
|
||||
}
|
||||
}
|
||||
Binary file not shown.
64
results/llm_curriculum_v5_early_obl/s1/resolved_config.yaml
Normal file
64
results/llm_curriculum_v5_early_obl/s1/resolved_config.yaml
Normal file
|
|
@ -0,0 +1,64 @@
|
|||
experiment: llm_curriculum_v5_early_obl
|
||||
seed: 1
|
||||
n_replicates: 1
|
||||
source_config:
|
||||
experiment: llm_curriculum_v5_early_obl
|
||||
kind: llm_curriculum
|
||||
base_model: Qwen/Qwen2.5-1.5B
|
||||
seed: 1
|
||||
families:
|
||||
- mnli
|
||||
- arc
|
||||
- hellaswag
|
||||
- squad
|
||||
- boolq
|
||||
- winogrande
|
||||
lineages: 3
|
||||
generations: 6
|
||||
arms:
|
||||
- society
|
||||
baselines: []
|
||||
allow_veto: false
|
||||
n_new: 300
|
||||
n_replay: 150
|
||||
n_test: 60
|
||||
n_val: 20
|
||||
epochs: 3
|
||||
lr: 0.0001
|
||||
operator: linear
|
||||
merge_weights:
|
||||
- - 0.5
|
||||
- 0.5
|
||||
- - 0.3
|
||||
- 0.7
|
||||
- - 0.7
|
||||
- 0.3
|
||||
max_new_tokens: 48
|
||||
batch_size: 48
|
||||
train_batch_size: 4
|
||||
train_max_len: 512
|
||||
lora:
|
||||
r: 16
|
||||
alpha: 32
|
||||
output:
|
||||
dir: results/llm_curriculum_v5_early_obl/s1
|
||||
orders:
|
||||
- - boolq
|
||||
- winogrande
|
||||
- mnli
|
||||
- arc
|
||||
- hellaswag
|
||||
- squad
|
||||
- - winogrande
|
||||
- boolq
|
||||
- arc
|
||||
- hellaswag
|
||||
- squad
|
||||
- mnli
|
||||
- - boolq
|
||||
- winogrande
|
||||
- hellaswag
|
||||
- squad
|
||||
- mnli
|
||||
- arc
|
||||
n_replicates: 1
|
||||
74
results/llm_curriculum_v5_early_obl/s2/manifest.json
Normal file
74
results/llm_curriculum_v5_early_obl/s2/manifest.json
Normal file
|
|
@ -0,0 +1,74 @@
|
|||
{
|
||||
"experiment": "llm_curriculum_v5_early_obl",
|
||||
"master_seed": 2,
|
||||
"git_commit": null,
|
||||
"python": "3.11.7",
|
||||
"libraries": {
|
||||
"numpy": "2.4.6",
|
||||
"scipy": "1.17.1",
|
||||
"pandas": "3.0.3",
|
||||
"pyarrow": "24.0.0",
|
||||
"torch": "2.12.1",
|
||||
"transformers": "5.16.1",
|
||||
"peft": "0.20.0"
|
||||
},
|
||||
"rows": 205,
|
||||
"results_sha256": "7847cd841fc61baaca6f3c469be9532f5065de43f965cb5422af08bad18f7f9a",
|
||||
"layer": "2",
|
||||
"tier": "llm",
|
||||
"base_model": "Qwen/Qwen2.5-1.5B",
|
||||
"hard": false,
|
||||
"curriculum": {
|
||||
"families": [
|
||||
"mnli",
|
||||
"arc",
|
||||
"hellaswag",
|
||||
"squad",
|
||||
"boolq",
|
||||
"winogrande"
|
||||
],
|
||||
"lineages": 3,
|
||||
"generations": 6,
|
||||
"arms": [
|
||||
"society"
|
||||
],
|
||||
"baselines": [],
|
||||
"n_new": 300,
|
||||
"n_replay": 150,
|
||||
"operator": "linear",
|
||||
"ancestor_depth": null,
|
||||
"lora": {
|
||||
"r": 16,
|
||||
"alpha": 32
|
||||
},
|
||||
"allow_veto": false,
|
||||
"merge_until": null,
|
||||
"orders": [
|
||||
[
|
||||
"boolq",
|
||||
"winogrande",
|
||||
"mnli",
|
||||
"arc",
|
||||
"hellaswag",
|
||||
"squad"
|
||||
],
|
||||
[
|
||||
"winogrande",
|
||||
"boolq",
|
||||
"arc",
|
||||
"hellaswag",
|
||||
"squad",
|
||||
"mnli"
|
||||
],
|
||||
[
|
||||
"boolq",
|
||||
"winogrande",
|
||||
"hellaswag",
|
||||
"squad",
|
||||
"mnli",
|
||||
"arc"
|
||||
]
|
||||
],
|
||||
"cull": null
|
||||
}
|
||||
}
|
||||
Binary file not shown.
64
results/llm_curriculum_v5_early_obl/s2/resolved_config.yaml
Normal file
64
results/llm_curriculum_v5_early_obl/s2/resolved_config.yaml
Normal file
|
|
@ -0,0 +1,64 @@
|
|||
experiment: llm_curriculum_v5_early_obl
|
||||
seed: 2
|
||||
n_replicates: 1
|
||||
source_config:
|
||||
experiment: llm_curriculum_v5_early_obl
|
||||
kind: llm_curriculum
|
||||
base_model: Qwen/Qwen2.5-1.5B
|
||||
seed: 2
|
||||
families:
|
||||
- mnli
|
||||
- arc
|
||||
- hellaswag
|
||||
- squad
|
||||
- boolq
|
||||
- winogrande
|
||||
lineages: 3
|
||||
generations: 6
|
||||
arms:
|
||||
- society
|
||||
baselines: []
|
||||
allow_veto: false
|
||||
n_new: 300
|
||||
n_replay: 150
|
||||
n_test: 60
|
||||
n_val: 20
|
||||
epochs: 3
|
||||
lr: 0.0001
|
||||
operator: linear
|
||||
merge_weights:
|
||||
- - 0.5
|
||||
- 0.5
|
||||
- - 0.3
|
||||
- 0.7
|
||||
- - 0.7
|
||||
- 0.3
|
||||
max_new_tokens: 48
|
||||
batch_size: 48
|
||||
train_batch_size: 4
|
||||
train_max_len: 512
|
||||
lora:
|
||||
r: 16
|
||||
alpha: 32
|
||||
output:
|
||||
dir: results/llm_curriculum_v5_early_obl/s2
|
||||
orders:
|
||||
- - boolq
|
||||
- winogrande
|
||||
- mnli
|
||||
- arc
|
||||
- hellaswag
|
||||
- squad
|
||||
- - winogrande
|
||||
- boolq
|
||||
- arc
|
||||
- hellaswag
|
||||
- squad
|
||||
- mnli
|
||||
- - boolq
|
||||
- winogrande
|
||||
- hellaswag
|
||||
- squad
|
||||
- mnli
|
||||
- arc
|
||||
n_replicates: 1
|
||||
74
results/llm_curriculum_v5_early_obl/s3/manifest.json
Normal file
74
results/llm_curriculum_v5_early_obl/s3/manifest.json
Normal file
|
|
@ -0,0 +1,74 @@
|
|||
{
|
||||
"experiment": "llm_curriculum_v5_early_obl",
|
||||
"master_seed": 3,
|
||||
"git_commit": null,
|
||||
"python": "3.11.7",
|
||||
"libraries": {
|
||||
"numpy": "2.4.6",
|
||||
"scipy": "1.17.1",
|
||||
"pandas": "3.0.3",
|
||||
"pyarrow": "24.0.0",
|
||||
"torch": "2.12.1",
|
||||
"transformers": "5.16.1",
|
||||
"peft": "0.20.0"
|
||||
},
|
||||
"rows": 205,
|
||||
"results_sha256": "e43d374c1ebe852d3a814fa8abb7393afbe749af038db3bc035e5135e0ea8c36",
|
||||
"layer": "2",
|
||||
"tier": "llm",
|
||||
"base_model": "Qwen/Qwen2.5-1.5B",
|
||||
"hard": false,
|
||||
"curriculum": {
|
||||
"families": [
|
||||
"mnli",
|
||||
"arc",
|
||||
"hellaswag",
|
||||
"squad",
|
||||
"boolq",
|
||||
"winogrande"
|
||||
],
|
||||
"lineages": 3,
|
||||
"generations": 6,
|
||||
"arms": [
|
||||
"society"
|
||||
],
|
||||
"baselines": [],
|
||||
"n_new": 300,
|
||||
"n_replay": 150,
|
||||
"operator": "linear",
|
||||
"ancestor_depth": null,
|
||||
"lora": {
|
||||
"r": 16,
|
||||
"alpha": 32
|
||||
},
|
||||
"allow_veto": false,
|
||||
"merge_until": null,
|
||||
"orders": [
|
||||
[
|
||||
"boolq",
|
||||
"winogrande",
|
||||
"mnli",
|
||||
"arc",
|
||||
"hellaswag",
|
||||
"squad"
|
||||
],
|
||||
[
|
||||
"winogrande",
|
||||
"boolq",
|
||||
"arc",
|
||||
"hellaswag",
|
||||
"squad",
|
||||
"mnli"
|
||||
],
|
||||
[
|
||||
"boolq",
|
||||
"winogrande",
|
||||
"hellaswag",
|
||||
"squad",
|
||||
"mnli",
|
||||
"arc"
|
||||
]
|
||||
],
|
||||
"cull": null
|
||||
}
|
||||
}
|
||||
Binary file not shown.
64
results/llm_curriculum_v5_early_obl/s3/resolved_config.yaml
Normal file
64
results/llm_curriculum_v5_early_obl/s3/resolved_config.yaml
Normal file
|
|
@ -0,0 +1,64 @@
|
|||
experiment: llm_curriculum_v5_early_obl
|
||||
seed: 3
|
||||
n_replicates: 1
|
||||
source_config:
|
||||
experiment: llm_curriculum_v5_early_obl
|
||||
kind: llm_curriculum
|
||||
base_model: Qwen/Qwen2.5-1.5B
|
||||
seed: 3
|
||||
families:
|
||||
- mnli
|
||||
- arc
|
||||
- hellaswag
|
||||
- squad
|
||||
- boolq
|
||||
- winogrande
|
||||
lineages: 3
|
||||
generations: 6
|
||||
arms:
|
||||
- society
|
||||
baselines: []
|
||||
allow_veto: false
|
||||
n_new: 300
|
||||
n_replay: 150
|
||||
n_test: 60
|
||||
n_val: 20
|
||||
epochs: 3
|
||||
lr: 0.0001
|
||||
operator: linear
|
||||
merge_weights:
|
||||
- - 0.5
|
||||
- 0.5
|
||||
- - 0.3
|
||||
- 0.7
|
||||
- - 0.7
|
||||
- 0.3
|
||||
max_new_tokens: 48
|
||||
batch_size: 48
|
||||
train_batch_size: 4
|
||||
train_max_len: 512
|
||||
lora:
|
||||
r: 16
|
||||
alpha: 32
|
||||
output:
|
||||
dir: results/llm_curriculum_v5_early_obl/s3
|
||||
orders:
|
||||
- - boolq
|
||||
- winogrande
|
||||
- mnli
|
||||
- arc
|
||||
- hellaswag
|
||||
- squad
|
||||
- - winogrande
|
||||
- boolq
|
||||
- arc
|
||||
- hellaswag
|
||||
- squad
|
||||
- mnli
|
||||
- - boolq
|
||||
- winogrande
|
||||
- hellaswag
|
||||
- squad
|
||||
- mnli
|
||||
- arc
|
||||
n_replicates: 1
|
||||
Loading…
Add table
Add a link
Reference in a new issue