© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Feature Splitting for GPT2-Small
    5. GPT2-Small
    7. Residual Feature Splitting 12288
    9. 8-RES_FS12288-JB
    11. 758
    Prev
    Next
    INDEX
    Explanations

    legal and financial terms related to relationships

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs12288-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_12288
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    12,288
    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
    203
    +0.10
    0.5%
    373
    +0.10
    0.5%
    332
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    232
    +0.10
    0.03
    406
    +0.10
    0.03
    373
    +0.10
    -0.01
    Negative Logits
    ©¶æ
    -0.79
    £ı
    -0.67
    nih
    -0.65
    ovych
    -0.65
     proceed
    -0.62
     behold
    -0.59
    hower
    -0.58
     "$:/
    -0.58
    clerosis
    -0.58
    âĶĢâĶĢâĶĢâĶĢ
    -0.58
    POSITIVE LOGITS
    essim
    0.92
    unky
    0.83
    hett
    0.82
    idon
    0.74
    hawk
    0.71
    lator
    0.71
    aneously
    0.69
    aker
    0.69
    aghetti
    0.69
    ivot
    0.68
    Activations Density 0.137%

    No Known Activations