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

    Neuronpedia

    Jacobian LensNEW
    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 24576
    9. 8-RES_FS24576-JB
    11. 10583
    Prev
    Next
    INDEX
    Explanations

    IP addresses and network-related terms in a technical context

    oai_token-act-pair · gpt-3.5-turboTriggered 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
    125
    +0.12
    0.5%
    654
    +0.10
    0.4%
    27
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    125
    +0.12
    0.03
    289
    +0.10
    0.01
    323
    +0.09
    0.03
    Negative Logits
    issance
    -0.70
    Increase
    -0.69
    TPPStreamerBot
    -0.68
    Grade
    -0.67
     refin
    -0.67
    sbm
    -0.66
    resso
    -0.66
    Progress
    -0.66
    ittle
    -0.65
    rodu
    -0.65
    POSITIVE LOGITS
     VPN
    1.09
     geographically
    1.08
    domain
    1.04
    outheast
    1.04
     servers
    0.98
     mainland
    0.96
     continents
    0.96
    addr
    0.96
     Address
    0.95
     domains
    0.94
    Activations Density 0.428%

    No Known Activations