© 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. 37297
    Prev
    Next
    INDEX
    Explanations

    instances where "mind" or thoughts are mentioned 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
    377
    +0.12
    0.6%
    414
    +0.11
    0.5%
    726
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    386
    +0.12
    0.01
    726
    +0.11
    0.01
    377
    +0.11
    0.01
    Negative Logits
    irements
    -0.69
    hered
    -0.65
    ilon
    -0.62
    imer
    -0.61
    itud
    -0.60
    elin
    -0.60
    otta
    -0.59
    mouth
    -0.59
     Cod
    -0.59
     rollout
    -0.59
    POSITIVE LOGITS
    anza
    0.66
     è£ıè
    0.65
     briefly
    0.64
    ffer
    0.63
     wondering
    0.62
    ovie
    0.59
    oldown
    0.58
    scape
    0.58
    chwitz
    0.58
     when
    0.57
    Activations Density 0.012%

    No Known Activations