© 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. 41681
    Prev
    Next
    INDEX
    Explanations

    references to specific Asian locations or names associated with Japan and China

    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
    631
    +0.13
    0.6%
    706
    +0.10
    0.5%
    373
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    232
    +0.13
    0.02
    373
    +0.10
    -0.01
    304
    +0.09
    0.02
    Negative Logits
    staking
    -0.80
    hered
    -0.75
    cham
    -0.72
     Bliss
    -0.70
    forth
    -0.69
     Malf
    -0.68
    ibly
    -0.65
    nown
    -0.65
    BLIC
    -0.65
    erent
    -0.64
    POSITIVE LOGITS
    代
    0.67
    insk
    0.62
    MG
    0.60
    士
    0.58
    hare
    0.58
     Picture
    0.58
    810
    0.56
    Conf
    0.55
    Picture
    0.55
    atom
    0.55
    Activations Density 0.044%

    No Known Activations