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 24576
    5. 8-RES_FS24576-JB
    6. 23164
    Prev
    Next
    INDEX
    Explanations

    words related to the act of capturing or preserving audio or video

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs24576-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    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
    651
    +0.14
    0.6%
    570
    +0.12
    0.6%
    609
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    609
    +0.14
    0.03
    651
    +0.12
    0.02
    540
    +0.11
    0.03
    Negative Logits
    adin
    -0.86
    ggles
    -0.76
    hiro
    -0.72
    osi
    -0.71
    enta
    -0.69
    kie
    -0.69
    pour
    -0.68
    htar
    -0.65
    adh
    -0.65
    egal
    -0.65
    POSITIVE LOGITS
    recorded
    0.94
     recordings
    0.90
     tapes
    0.85
     recording
    0.82
     Recording
    0.78
     audio
    0.77
     recorder
    0.77
     sessions
    0.74
     cassette
    0.73
     Tape
    0.73
    Activations Density 0.040%

    No Known Activations