© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    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. 9986
    Prev
    Next
    INDEX
    Explanations

    specific pieces of information or details in a report or document

    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
    481
    +0.18
    0.8%
    171
    +0.13
    0.6%
    71
    +0.12
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    171
    +0.18
    0.05
    362
    +0.13
    0.04
    177
    +0.12
    0.03
    Negative Logits
    pload
    -0.79
    anamo
    -0.78
    robe
    -0.77
    resses
    -0.76
    trl
    -0.74
    asking
    -0.73
    ropolis
    -0.73
    apo
    -0.72
    okes
    -0.70
    erers
    -0.69
    POSITIVE LOGITS
     observations
    1.18
     assumptions
    1.15
     criteria
    1.14
     assumption
    1.07
     evaluations
    1.04
     principles
    1.04
     feedback
    1.03
     intuition
    1.01
     observation
    1.01
     circumstances
    1.00
    Activations Density 0.356%

    No Known Activations