© 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 49152
    9. 8-RES_FS49152-JB
    11. 41533
    Prev
    Next
    INDEX
    Explanations

    phrases emphasizing transitions or actions directed towards a target

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs49152-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    49,152
    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
    608
    +0.13
    0.6%
    320
    +0.10
    0.5%
    385
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    608
    +0.13
    0.02
    385
    +0.10
    0.02
    305
    +0.10
    0.01
    Negative Logits
     largeDownload
    -0.83
    anted
    -0.76
    ahime
    -0.74
    soDeliveryDate
    -0.74
    ersed
    -0.70
    rared
    -0.70
    URI
    -0.69
    handled
    -0.69
    published
    -0.67
    ersive
    -0.67
    POSITIVE LOGITS
     accommodate
    0.75
     appease
    0.75
     temptation
    0.75
     avoid
    0.73
     rival
    0.72
    ilet
    0.72
     injure
    0.71
     Thatcher
    0.71
     Tie
    0.70
     defeat
    0.70
    Activations Density 0.040%

    No Known Activations