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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Feature Splitting for GPT2-Small
    5. GPT2-Small
    7. Residual Feature Splitting 49152
    9. 8-RES_FS49152-JB
    11. 26555
    Prev
    Next
    INDEX
    Explanations

    instances of the word "ret" in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs49152-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    49,152
    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
    651
    +0.10
    0.5%
    425
    +0.10
    0.4%
    114
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    70
    +0.10
    0.01
    651
    +0.10
    0.01
    114
    +0.09
    0.01
    Negative Logits
     Thames
    -0.64
    BuyableInstoreAndOnline
    -0.64
     ILCS
    -0.60
    OPER
    -0.60
    catentry
    -0.60
    ADRA
    -0.59
    CHR
    -0.59
    omnia
    -0.58
    latest
    -0.58
    isconsin
    -0.57
    POSITIVE LOGITS
    ting
    0.96
    ooth
    0.94
    zel
    0.93
    ribution
    0.91
    rans
    0.88
    ilde
    0.83
     unfocusedRange
    0.82
    ract
    0.82
    sov
    0.82
    zman
    0.82
    Activations Density 0.004%

    No Known Activations