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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & 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. 211
    Prev
    Next
    INDEX
    Explanations

    mentions of negative events or situations

    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
    750
    +0.12
    0.5%
    161
    +0.11
    0.5%
    497
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    750
    +0.12
    0.07
    270
    +0.11
    0.04
    161
    +0.10
    0.06
    Negative Logits
     ensured
    -0.74
    eros
    -0.72
    Ĥİ
    -0.71
    ente
    -0.70
     ensures
    -0.68
    ©¶æ¥µ
    -0.67
     maintains
    -0.67
    keeping
    -0.66
     depended
    -0.66
    assisted
    -0.64
    POSITIVE LOGITS
     unfold
    1.15
     firsthand
    1.08
     afar
    0.93
     VIDEOS
    0.90
     resemblance
    0.86
     closely
    0.80
     silhou
    0.79
     replay
    0.78
     similarities
    0.78
    ideos
    0.76
    Activations Density 3.817%

    No Known Activations