© 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. 7619
    Prev
    Next
    INDEX
    Explanations

    phrases related to speculation or evaluation of actions

    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
    551
    +0.13
    0.6%
    278
    +0.11
    0.5%
    687
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    551
    +0.13
    0.10
    278
    +0.11
    0.05
    49
    +0.10
    0.06
    Negative Logits
     occurs
    -0.71
     Indo
    -0.67
     occurring
    -0.67
     occurrence
    -0.63
     Celsius
    -0.63
     occurred
    -0.61
     occurrences
    -0.61
    Eva
    -0.60
     Sci
    -0.59
    irection
    -0.59
    POSITIVE LOGITS
     surely
    0.89
    bes
    0.88
     doubtless
    0.86
     ideally
    0.86
    ÄŁ
    0.86
     bask
    0.85
     understandably
    0.84
     reconsider
    0.83
    udos
    0.82
    ering
    0.81
    Activations Density 0.256%

    No Known Activations