© 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. 20496
    Prev
    Next
    INDEX
    Explanations

    references to forests or forest-related activities, such as logging and conservation efforts

    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
    288
    +0.16
    0.8%
    481
    +0.12
    0.6%
    27
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.16
    0.02
    27
    +0.12
    0.02
    460
    +0.10
    0.02
    Negative Logits
    OPLE
    -0.68
    HK
    -0.67
    soever
    -0.65
    Extra
    -0.63
    pter
    -0.63
    Universal
    -0.63
    Magikarp
    -0.61
    enger
    -0.60
    realDonaldTrump
    -0.60
    Arcade
    -0.60
    POSITIVE LOGITS
    lands
    0.98
    forest
    0.97
     canopy
    0.97
    lake
    0.90
     forests
    0.88
     vegetation
    0.85
    ocene
    0.84
     forest
    0.82
    land
    0.79
     woodland
    0.79
    Activations Density 0.027%

    No Known Activations