Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APISteerSAE EvalsBlog/PodcastNEWSlackPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlog/PodcastGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 3072
    5. 8-RES_FS3072-JB
    6. 388
    Prev
    Next
    INDEX
    Explanations

    phrases related to significant roles or impacts in various contexts

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs3072-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_3072
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    3,072
    Data Type
    torch.float32
    Hook Point
    blocks.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    447
    +0.18
    0.8%
    575
    +0.12
    0.6%
    760
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    575
    +0.18
    0.13
    531
    +0.12
    0.11
    247
    +0.10
    0.09
    Negative Logits
     peas
    -0.77
     totality
    -0.61
    oked
    -0.59
    arus
    -0.58
     recip
    -0.58
     Tsukuyomi
    -0.57
     Chains
    -0.56
     mild
    -0.56
     quint
    -0.55
     periodic
    -0.55
    POSITIVE LOGITS
     regarding
    0.82
     towards
    0.78
     toward
    0.75
     whatsoever
    0.73
     elsewhere
    0.68
     throughout
    0.67
     internally
    0.66
    ãħĭãħĭ
    0.66
     against
    0.66
    Theme
    0.64
    Activations Density 1.786%

    No Known Activations