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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Feature Splitting for GPT2-Small
    5. GPT2-Small
    7. Residual Feature Splitting 12288
    9. 8-RES_FS12288-JB
    11. 785
    Prev
    Next
    INDEX
    Explanations

    words related to ranks or titles

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs12288-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_12288
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    12,288
    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
    373
    +0.12
    0.6%
    436
    +0.11
    0.5%
    264
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    232
    +0.12
    0.04
    299
    +0.11
    0.04
    373
    +0.11
    -0.02
    Negative Logits
     endowed
    -0.69
     Clover
    -0.63
     Calder
    -0.60
     sterling
    -0.58
     underrated
    -0.55
     Kessler
    -0.54
     silent
    -0.54
     Reboot
    -0.54
     Bett
    -0.54
     pim
    -0.54
    POSITIVE LOGITS
    pta
    0.86
    atchewan
    0.85
    atoon
    0.84
    cific
    0.77
    oslav
    0.76
    othal
    0.73
    oku
    0.73
    adic
    0.73
    inx
    0.72
    udos
    0.71
    Activations Density 0.186%

    No Known Activations