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. 1633
    Prev
    Next
    INDEX
    Explanations

    biographical information about notable individuals

    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
    105
    +0.10
    0.4%
    642
    +0.10
    0.4%
    732
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    603
    +0.10
    0.07
    545
    +0.10
    0.06
    469
    +0.10
    0.05
    Negative Logits
     takedown
    -0.88
    forcement
    -0.80
    aminer
    -0.79
    parency
    -0.77
    cific
    -0.77
     effectiveness
    -0.76
    aptic
    -0.76
    ħĭ
    -0.75
    nox
    -0.72
    awatts
    -0.71
    POSITIVE LOGITS
     upbringing
    1.13
     childhood
    1.05
     Married
    0.98
     married
    0.98
     eldest
    0.97
     lifelong
    0.93
     father
    0.93
     parents
    0.92
     immigrant
    0.91
     grandparents
    0.91
    Activations Density 1.818%

    No Known Activations