© 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 49152
    9. 8-RES_FS49152-JB
    11. 26575
    Prev
    Next
    INDEX
    Explanations

    references to dinosaurs

    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
    351
    +0.15
    0.7%
    288
    +0.13
    0.6%
    447
    +0.12
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    351
    +0.15
    0.01
    403
    +0.13
    0.01
    549
    +0.12
    0.01
    Negative Logits
    Recommend
    -0.76
    ple
    -0.72
    76561
    -0.72
    OND
    -0.70
    OA
    -0.69
    à¨
    -0.68
    ndra
    -0.67
     Agents
    -0.65
    mberg
    -0.64
    aspx
    -0.63
    POSITIVE LOGITS
    osaurs
    1.68
    osaurus
    1.26
     fossils
    1.03
     dinosaurs
    1.01
    arium
    0.98
     skeletons
    0.88
     dinosaur
    0.86
    aurus
    0.86
     skeleton
    0.83
    inosaur
    0.82
    Activations Density 0.006%

    No Known Activations