© 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 24576
    9. 8-RES_FS24576-JB
    11. 1773
    Prev
    Next
    INDEX
    Explanations

    instances where there are multiple overlapping locations mentioned

    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
    314
    +0.12
    0.5%
    254
    +0.10
    0.5%
    140
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    426
    +0.12
    0.02
    254
    +0.10
    0.02
    687
    +0.10
    0.02
    Negative Logits
    76561
    -0.65
    susp
    -0.60
     psychiatry
    -0.59
     sshd
    -0.59
    phony
    -0.58
     Susp
    -0.57
    HAM
    -0.57
    bsp
    -0.56
     resusc
    -0.55
     Detection
    -0.55
    POSITIVE LOGITS
    nesday
    0.78
    女
    0.76
    çīĪ
    0.69
    NOW
    0.69
     Chile
    0.69
     Nieto
    0.67
    abad
    0.66
    anchester
    0.66
    iah
    0.65
    orld
    0.64
    Activations Density 0.091%

    No Known Activations