© 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 3072
    9. 8-RES_FS3072-JB
    11. 93
    Prev
    Next
    INDEX
    Explanations

    phrases related to evaluations or assessments, particularly in the context of skills or characteristics

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs3072-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_3072
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    3,072
    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
    674
    +0.10
    0.5%
    538
    +0.10
    0.5%
    408
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    764
    +0.10
    0.06
    538
    +0.10
    0.04
    206
    +0.10
    0.04
    Negative Logits
    etheless
    -0.67
     Digest
    -0.65
     immensely
    -0.59
     Provides
    -0.59
     FROM
    -0.58
    irez
    -0.57
    days
    -0.55
    DI
    -0.55
    eming
    -0.54
    eness
    -0.54
    POSITIVE LOGITS
     aside
    0.89
     brakes
    0.84
     lid
    0.80
     jeopardy
    0.79
     emphasis
    0.74
     groundwork
    0.70
     certs
    0.65
     pedal
    0.65
     pressure
    0.64
     precedent
    0.64
    Activations Density 1.027%

    No Known Activations