© 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
    3. Joseph Bloom · Feature Splitting for GPT2-Small
    5. GPT2-Small
    7. Residual Feature Splitting 3072
    9. 8-RES_FS3072-JB
    11. 237
    Prev
    Next
    INDEX
    Explanations

    strings formatted as numerical stats accompanied by specific keywords

    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
    238
    +0.12
    0.5%
    316
    +0.11
    0.5%
    373
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    316
    +0.12
    0.08
    176
    +0.11
    0.06
    326
    +0.11
    0.07
    Negative Logits
    DragonMagazine
    -0.66
    lehem
    -0.65
    catentry
    -0.63
    aido
    -0.60
    dayName
    -0.60
    orough
    -0.58
    IDs
    -0.58
    reb
    -0.57
     rehe
    -0.57
    ¨
    -0.57
    POSITIVE LOGITS
    99
    0.69
    98
    0.67
    96
    0.66
    686
    0.65
    antage
    0.65
    75
    0.64
    66
    0.64
    997
    0.63
    93
    0.63
    34
    0.63
    Activations Density 0.634%

    No Known Activations