© 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 EvalsExportsinterp-engineNEWAPI 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. 16651
    Prev
    Next
    INDEX
    Explanations

    verbs related to action and decision-making

    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
    337
    +0.13
    0.6%
    196
    +0.10
    0.5%
    641
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    337
    +0.13
    0.04
    665
    +0.10
    0.05
    113
    +0.10
    0.04
    Negative Logits
    misc
    -0.72
    illegal
    -0.69
    aven
    -0.69
    osate
    -0.68
    laughs
    -0.68
    dra
    -0.67
     exception
    -0.66
    cpp
    -0.66
    vol
    -0.66
    except
    -0.61
    POSITIVE LOGITS
     unfolds
    0.86
     progresses
    0.85
     prepares
    0.81
     consolid
    0.79
     progressively
    0.74
     increasingly
    0.73
     unfold
    0.72
     transitioned
    0.70
     frantically
    0.69
     dwindling
    0.69
    Activations Density 0.292%

    No Known Activations