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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsBlogSlackPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 1536
    5. 8-RES_FS1536-JB
    6. 878
    Prev
    Next
    INDEX
    Explanations

    information related to a specific person and their life events

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs1536-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_1536
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    1,536
    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
    314
    +0.13
    0.6%
    133
    +0.09
    0.4%
    378
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    106
    +0.13
    0.10
    253
    +0.09
    0.10
    133
    +0.09
    0.09
    Negative Logits
     Paste
    -0.54
     Amendments
    -0.52
     Beg
    -0.50
     conclusions
    -0.49
    idious
    -0.49
     advances
    -0.49
    phasis
    -0.48
     CoC
    -0.48
    MSN
    -0.47
    oidal
    -0.47
    POSITIVE LOGITS
    etheless
    0.90
    ħĭ
    0.90
    morning
    0.88
     morning
    0.88
    til
    0.87
    ©¶æ
    0.84
     way
    0.84
     week
    0.81
     mornings
    0.79
    Ĥİ
    0.78
    Activations Density 3.486%

    No Known Activations