© 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. 22097
    Prev
    Next
    INDEX
    Explanations

    references to data and statistics related to groups of individuals

    oai_token-act-pair · gpt-4o-miniTriggered 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
    440
    +0.12
    0.6%
    447
    +0.11
    0.5%
    450
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    153
    +0.12
    0.02
    341
    +0.11
    0.02
    488
    +0.11
    0.02
    Negative Logits
    --------------------------------------------------------
    -0.63
    }}}
    -0.61
    rast
    -0.60
    },"
    -0.60
     }}
    -0.60
    PDATE
    -0.59
     ramps
    -0.58
    externalActionCode
    -0.56
     actionGroup
    -0.56
    OULD
    -0.56
    POSITIVE LOGITS
     surveyed
    1.12
     interviewed
    0.96
     polled
    0.94
     screened
    0.76
     fielded
    0.72
     cited
    0.71
     participating
    0.71
     sampled
    0.69
     reviewed
    0.68
     eligible
    0.68
    Activations Density 0.095%

    No Known Activations