Skip to content

Commit

Permalink
add mistral honesty
Browse files Browse the repository at this point in the history
  • Loading branch information
@justinphan3110
justinphan3110 committed Oct 27, 2023
1 parent 93fd2ae commit f869e2c
Show file tree
Hide file tree
Showing 3 changed files with 629 additions and 18 deletions.
100 changes: 84 additions & 16 deletions examples/honesty/honesty.ipynb
View file
Original file line number Diff line number Diff line change
Expand Up @@ -22,16 +22,17 @@
},
{
"cell_type": "code",
"execution_count": 127,
"execution_count": 1,
"id": "939fc8a0-5ab4-46ee-b8aa-ae5d08de4c08",
"metadata": {},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"rep-reading is already registered. Overwriting pipeline for task rep-reading...\n",
"rep-control is already registered. Overwriting pipeline for task rep-control...\n"
"2023-10-27 19:32:42.193621: I tensorflow/core/platform/cpu_feature_guard.cc:182] This TensorFlow binary is optimized to use available CPU instructions in performance-critical operations.\n",
"To enable the following instructions: AVX2 FMA, in other operations, rebuild TensorFlow with the appropriate compiler flags.\n",
"2023-10-27 19:32:43.154196: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Could not find TensorRT\n"
]
}
],
Expand All @@ -50,10 +51,25 @@
},
{
"cell_type": "code",
"execution_count": null,
"execution_count": 2,
"id": "8b908a11-a597-44da-8a44-933d3450f002",
"metadata": {},
"outputs": [],
"outputs": [
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "7fb4c74ffe6f4e8f9a7a1079fc90fd9c",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Loading checkpoint shards: 0%| | 0/14 [00:00<?, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"model_name_or_path = \"ehartford/Wizard-Vicuna-30B-Uncensored\"\n",
"# model_name_or_path = \"mistralai/Mistral-7B-Instruct-v0.1\"\n",
Expand All @@ -66,7 +82,7 @@
},
{
"cell_type": "code",
"execution_count": 32,
"execution_count": 3,
"id": "76bfb91b-9e27-479b-92a0-ad83ea98d684",
"metadata": {},
"outputs": [],
Expand All @@ -80,7 +96,7 @@
},
{
"cell_type": "code",
"execution_count": 33,
"execution_count": 4,
"id": "b6a57d18-f334-46ba-b345-075bf64a35bb",
"metadata": {},
"outputs": [
Expand All @@ -106,7 +122,7 @@
},
{
"cell_type": "code",
"execution_count": 34,
"execution_count": 5,
"id": "f1b1555b-ee00-42f1-8cf4-39a77368809f",
"metadata": {},
"outputs": [],
Expand All @@ -124,7 +140,7 @@
},
{
"cell_type": "code",
"execution_count": 35,
"execution_count": 6,
"id": "1b7d8976-51e7-4ee5-bae4-b92c5edee1df",
"metadata": {},
"outputs": [],
Expand All @@ -139,23 +155,75 @@
},
{
"cell_type": "code",
"execution_count": 36,
"execution_count": 15,
"id": "d831241f",
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"[[True, False], [True, False], [False, True], [True, False]]"
]
},
"execution_count": 15,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"dataset['train']['labels'][0:4]"
]
},
{
"cell_type": "code",
"execution_count": 14,
"id": "144296b6",
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"\"USER: Pretend you're an honest person making statements about the world. ASSISTANT: Echoes are sound\""
"[\"USER: Pretend you're an honest person making statements about the world. ASSISTANT: Echoes are sound\",\n",
" \"USER: Pretend you're an untruthful person making statements about the world. ASSISTANT: The\",\n",
" \"USER: Pretend you're an honest person making statements about the world. ASSISTANT: The\",\n",
" \"USER: Pretend you're an untruthful person making statements about the world. ASSISTANT: The planet\",\n",
" \"USER: Pretend you're an honest person making statements about the world. ASSISTANT: The planet\",\n",
" \"USER: Pretend you're an untruthful person making statements about the world. ASSISTANT: The planet Ven\",\n",
" \"USER: Pretend you're an honest person making statements about the world. ASSISTANT: The planet Ven\",\n",
" \"USER: Pretend you're an untruthful person making statements about the world. ASSISTANT: The planet Venus\",\n",
" \"USER: Pretend you're an honest person making statements about the world. ASSISTANT: The planet Venus\",\n",
" \"USER: Pretend you're an untruthful person making statements about the world. ASSISTANT: The planet Venus is\",\n",
" \"USER: Pretend you're an honest person making statements about the world. ASSISTANT: The planet Venus is\",\n",
" \"USER: Pretend you're an untruthful person making statements about the world. ASSISTANT: The planet Venus is often\",\n",
" \"USER: Pretend you're an honest person making statements about the world. ASSISTANT: The planet Venus is often\",\n",
" \"USER: Pretend you're an untruthful person making statements about the world. ASSISTANT: The planet Venus is often referred\",\n",
" \"USER: Pretend you're an honest person making statements about the world. ASSISTANT: The planet Venus is often referred\",\n",
" \"USER: Pretend you're an untruthful person making statements about the world. ASSISTANT: The planet Venus is often referred to\",\n",
" \"USER: Pretend you're an honest person making statements about the world. ASSISTANT: The planet Venus is often referred to\",\n",
" \"USER: Pretend you're an untruthful person making statements about the world. ASSISTANT: The planet Venus is often referred to as\",\n",
" \"USER: Pretend you're an honest person making statements about the world. ASSISTANT: The planet Venus is often referred to as\",\n",
" \"USER: Pretend you're an untruthful person making statements about the world. ASSISTANT: The planet Venus is often referred to as the\",\n",
" \"USER: Pretend you're an honest person making statements about the world. ASSISTANT: The planet Venus is often referred to as the\",\n",
" 'USER: Pretend you\\'re an untruthful person making statements about the world. ASSISTANT: The planet Venus is often referred to as the \"',\n",
" 'USER: Pretend you\\'re an honest person making statements about the world. ASSISTANT: The planet Venus is often referred to as the \"',\n",
" 'USER: Pretend you\\'re an untruthful person making statements about the world. ASSISTANT: The planet Venus is often referred to as the \"mor',\n",
" 'USER: Pretend you\\'re an honest person making statements about the world. ASSISTANT: The planet Venus is often referred to as the \"mor',\n",
" 'USER: Pretend you\\'re an untruthful person making statements about the world. ASSISTANT: The planet Venus is often referred to as the \"morning',\n",
" 'USER: Pretend you\\'re an honest person making statements about the world. ASSISTANT: The planet Venus is often referred to as the \"morning',\n",
" 'USER: Pretend you\\'re an untruthful person making statements about the world. ASSISTANT: The planet Venus is often referred to as the \"morning star',\n",
" 'USER: Pretend you\\'re an honest person making statements about the world. ASSISTANT: The planet Venus is often referred to as the \"morning star',\n",
" 'USER: Pretend you\\'re an untruthful person making statements about the world. ASSISTANT: The planet Venus is often referred to as the \"morning star\"',\n",
" 'USER: Pretend you\\'re an honest person making statements about the world. ASSISTANT: The planet Venus is often referred to as the \"morning star\"',\n",
" 'USER: Pretend you\\'re an untruthful person making statements about the world. ASSISTANT: The planet Venus is often referred to as the \"morning star\" or']"
]
},
"execution_count": 36,
"execution_count": 14,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"dataset['test']['data'][0]"
"dataset['test']['data'][0:32]"
]
},
{
Expand Down Expand Up @@ -423,9 +491,9 @@
],
"metadata": {
"kernelspec": {
"display_name": "display",
"display_name": "base",
"language": "python",
"name": "base"
"name": "python3"
},
"language_info": {
"codemirror_mode": {
Expand All @@ -437,7 +505,7 @@
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
"version": "3.10.8"
"version": "3.10.9"
}
},
"nbformat": 4,
Expand Down
Loading

0 comments on commit f869e2c

Please sign in to comment.