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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 3072
    5. 8-RES_FS3072-JB
    6. 2566
    Prev
    Next
    INDEX
    Explanations

    information related to construction projects and criminal activities

    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
    87
    +0.39
    2.2%
    756
    +0.14
    0.8%
    464
    +0.09
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    87
    +0.39
    -0.07
    756
    +0.14
    -0.06
    459
    +0.09
    0.13
    Negative Logits
    reetings
    -0.53
    estern
    -0.53
    htaking
    -0.49
     Fasc
    -0.48
    ?",
    -0.48
    hua
    -0.47
     Anthropology
    -0.46
    abases
    -0.46
    hens
    -0.46
     Negro
    -0.44
    POSITIVE LOGITS
     nonetheless
    0.84
     accordingly
    0.82
     thereafter
    0.73
     thereto
    0.71
     afterwards
    0.69
     thereof
    0.67
     afterward
    0.63
     anyway
    0.62
     elsewhere
    0.61
     nevertheless
    0.60
    Activations Density 8.556%

    No Known Activations