© 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. 9154
    Prev
    Next
    INDEX
    Explanations

    mentions of personal memories and storytelling, particularly focusing on specific recollections or reminiscences

    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
    718
    +0.10
    0.4%
    239
    +0.09
    0.4%
    684
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    755
    +0.10
    0.02
    750
    +0.09
    0.02
    739
    +0.09
    0.01
    Negative Logits
    tan
    -0.96
    acies
    -0.90
    oit
    -0.89
    redit
    -0.82
    ourage
    -0.82
    ahime
    -0.79
    ocrats
    -0.76
    Div
    -0.75
    ÃŁ
    -0.75
    blogspot
    -0.75
    POSITIVE LOGITS
     vividly
    1.38
     waking
    1.13
     seeing
    1.11
     encountering
    1.10
     hearing
    1.09
     remembering
    1.01
     distinctly
    0.99
     recalling
    0.94
     fond
    0.94
     having
    0.92
    Activations Density 0.031%

    No Known Activations