Uploading saes for ['model.layers.2', 'model.layers.3', 'model.layers.4', 'model.layers.5', 'model.layers.6', 'model.layers.7', 'model.layers.8', 'model.layers.9', 'model.layers.10', 'model.layers.11', 'model.layers.12', 'model.layers.13', 'model.layers.14']

config.json

@@ -0,0 +1 @@
+{"sae": {"expansion_factor": 4, "normalize_decoder": true, "num_latents": 0, "k": 128, "multi_topk": false, "skip_connection": false}, "batch_size": 8, "grad_acc_steps": 1, "micro_acc_steps": 1, "lr": 0.0005, "lr_warmup_steps": 1000, "auxk_alpha": 0.0, "dead_feature_threshold": 10000000, "hookpoints": ["model.layers.2", "model.layers.3", "model.layers.4", "model.layers.5", "model.layers.6", "model.layers.7", "model.layers.8", "model.layers.9", "model.layers.10", "model.layers.11", "model.layers.12", "model.layers.13", "model.layers.14"], "init_seeds": [0], "layers": [], "layer_stride": 1, "transcode": false, "distribute_modules": false, "save_every": 100000, "log_to_wandb": true, "run_name": "saes_meta-llama/Llama-3.2-1B-Instruct/k=128", "wandb_log_frequency": 100}

lr_scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9c5ccd3bb9f2aeda2b31fc7539bb013bd8efaa29e9c5d1a89e7b2f4b97f5a6ff
+size 1268

model.layers.10/cfg.json

@@ -0,0 +1 @@
+{"expansion_factor": 4, "normalize_decoder": true, "num_latents": 0, "k": 128, "multi_topk": false, "skip_connection": false, "d_in": 2048}

model.layers.10/metrics.json

@@ -0,0 +1 @@
+null

model.layers.10/sae.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fac286cc73bda7a92318dbb6aa3f161ab5bc804d282fdc1056db333f5241ce54
+size 134259016

model.layers.11/cfg.json

@@ -0,0 +1 @@
+{"expansion_factor": 4, "normalize_decoder": true, "num_latents": 0, "k": 128, "multi_topk": false, "skip_connection": false, "d_in": 2048}

model.layers.11/metrics.json

@@ -0,0 +1 @@
+null

model.layers.11/sae.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5e3682ab25bdbe7c967f3af961ca7710f203acedb44ee0502a3c4da34a36f7b6
+size 134259016

model.layers.12/cfg.json

@@ -0,0 +1 @@
+{"expansion_factor": 4, "normalize_decoder": true, "num_latents": 0, "k": 128, "multi_topk": false, "skip_connection": false, "d_in": 2048}

model.layers.12/metrics.json

@@ -0,0 +1 @@
+null

model.layers.12/sae.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f42088c4ee71cc09bfbe00a31b18b518cedfa096e61c7237408de46c40e3b5cc
+size 134259016

model.layers.13/cfg.json

@@ -0,0 +1 @@
+{"expansion_factor": 4, "normalize_decoder": true, "num_latents": 0, "k": 128, "multi_topk": false, "skip_connection": false, "d_in": 2048}

model.layers.13/metrics.json

@@ -0,0 +1 @@
+null

model.layers.13/sae.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2eb0112892e140bd9302541f485ab90ea0f3f9439548fdbef3c3e5efa23e6e69
+size 134259016

model.layers.14/cfg.json

@@ -0,0 +1 @@
+{"expansion_factor": 4, "normalize_decoder": true, "num_latents": 0, "k": 128, "multi_topk": false, "skip_connection": false, "d_in": 2048}

model.layers.14/metrics.json

@@ -0,0 +1 @@
+null

model.layers.14/sae.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b55b1d3859923776f97194f64ba7d8162f0010c4a007a29fa1163281e1b79954
+size 134259016

model.layers.2/cfg.json

@@ -0,0 +1 @@
+{"expansion_factor": 4, "normalize_decoder": true, "num_latents": 0, "k": 128, "multi_topk": false, "skip_connection": false, "d_in": 2048}

model.layers.2/metrics.json

@@ -0,0 +1 @@
+null

model.layers.2/sae.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:53f7764b3cb5c55b1921a3079bde899d30056a3427a60c78e1d177b5e4459918
+size 134259016

model.layers.3/cfg.json

@@ -0,0 +1 @@
+{"expansion_factor": 4, "normalize_decoder": true, "num_latents": 0, "k": 128, "multi_topk": false, "skip_connection": false, "d_in": 2048}

model.layers.3/metrics.json

@@ -0,0 +1 @@
+null

model.layers.3/sae.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a0ecff217049a44d6e92acac77ed5fda1eedf8f7ff87712e63dff1aab5894c4a
+size 134259016

model.layers.4/cfg.json

@@ -0,0 +1 @@
+{"expansion_factor": 4, "normalize_decoder": true, "num_latents": 0, "k": 128, "multi_topk": false, "skip_connection": false, "d_in": 2048}

model.layers.4/metrics.json

@@ -0,0 +1 @@
+null

model.layers.4/sae.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:51047b1518f13acc7e536c927868fc95a9b09079269377097296ce0efd09ed15
+size 134259016

model.layers.5/cfg.json

@@ -0,0 +1 @@
+{"expansion_factor": 4, "normalize_decoder": true, "num_latents": 0, "k": 128, "multi_topk": false, "skip_connection": false, "d_in": 2048}

model.layers.5/metrics.json

@@ -0,0 +1 @@
+null

model.layers.5/sae.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6ee7fc310eb5c6c8ffd014fb65e047b16b7e87c865c32afec142dc5c09a4279d
+size 134259016

model.layers.6/cfg.json

@@ -0,0 +1 @@
+{"expansion_factor": 4, "normalize_decoder": true, "num_latents": 0, "k": 128, "multi_topk": false, "skip_connection": false, "d_in": 2048}

model.layers.6/metrics.json

@@ -0,0 +1 @@
+null

model.layers.6/sae.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2d7e9ffa702cc16e79d24a1b516ffa610fbd7a3b8eedeedfb248ce0cdbfb4296
+size 134259016

model.layers.7/cfg.json

@@ -0,0 +1 @@
+{"expansion_factor": 4, "normalize_decoder": true, "num_latents": 0, "k": 128, "multi_topk": false, "skip_connection": false, "d_in": 2048}

model.layers.7/metrics.json

@@ -0,0 +1 @@
+null

model.layers.7/sae.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:602581fc9a2bade3be49401c90ab33f2899bcfa0f979b717452b983db154d16f
+size 134259016

model.layers.8/cfg.json

@@ -0,0 +1 @@
+{"expansion_factor": 4, "normalize_decoder": true, "num_latents": 0, "k": 128, "multi_topk": false, "skip_connection": false, "d_in": 2048}

model.layers.8/metrics.json

@@ -0,0 +1 @@
+null

model.layers.8/sae.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e1d0439a7ad3ef0de2121c68cfd636ca64c61da84e04fe666aa9682ab8de8c4c
+size 134259016

model.layers.9/cfg.json

@@ -0,0 +1 @@
+{"expansion_factor": 4, "normalize_decoder": true, "num_latents": 0, "k": 128, "multi_topk": false, "skip_connection": false, "d_in": 2048}

model.layers.9/metrics.json

@@ -0,0 +1 @@
+null

model.layers.9/sae.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:51a4122b3039cf9de61efcdf525b2773713bbd6178b6e9b9428e3efabdda725d
+size 134259016

model_config.json

@@ -0,0 +1 @@
+{"model_name": "meta-llama/Llama-3.2-1B-Instruct", "dataset_name": "lukemarks/vader-post-training"}

optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a49ae529432a6b168cfa5fcde6fb865575942fd966d033caa9de606ee3b0bc98
+size 886533370

state.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1fc26200b4842d45e17a6b29b104c541b9f77163fd39ffb9e15b3ba2fd2759f6
+size 855629