Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 49152
    5. 8-RES_FS49152-JB
    6. 41397
    Prev
    Next
    INDEX
    Explanations

    references to the concept of limitations or the idea of "beyond."

    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
    113
    +0.13
    0.6%
    99
    +0.10
    0.5%
    334
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    113
    +0.13
    0.01
    57
    +0.10
    0.01
    327
    +0.10
    0.01
    Negative Logits
    late
    -0.72
    Dub
    -0.68
    Mu
    -0.67
    nesium
    -0.66
    vati
    -0.65
    ramid
    -0.64
    ector
    -0.62
    male
    -0.61
    heastern
    -0.60
    ounces
    -0.60
    POSITIVE LOGITS
     comprehension
    0.90
    pport
    0.84
    ceivable
    0.83
     doubt
    0.79
    vable
    0.74
     Neptune
    0.71
    avorite
    0.69
     grasp
    0.69
    plain
    0.68
     unreasonable
    0.68
    Activations Density 0.024%

    No Known Activations