© 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. 17521
    Prev
    Next
    INDEX
    Explanations

    information related to neuroscience, biology, and research studies

    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
    160
    +0.13
    0.6%
    314
    +0.11
    0.5%
    102
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    160
    +0.13
    0.04
    112
    +0.11
    0.03
    17
    +0.10
    0.03
    Negative Logits
     deprivation
    -0.71
     redu
    -0.70
     superf
    -0.68
     bands
    -0.66
     diminishing
    -0.66
     lesser
    -0.65
     strap
    -0.65
     regression
    -0.64
     upl
    -0.64
     hourly
    -0.63
    POSITIVE LOGITS
    sic
    1.40
    ?]
    1.37
    !]
    1.28
     edit
    1.23
    emphasis
    1.22
    ].
    1.21
     ].
    1.20
    Footnote
    1.18
    :]
    1.16
    ](
    1.13
    Activations Density 0.030%

    No Known Activations