© 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. 41794
    Prev
    Next
    INDEX
    Explanations

    references to lever actions and mechanisms

    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
    288
    +0.13
    0.6%
    62
    +0.12
    0.6%
    624
    +0.12
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    624
    +0.13
    0.01
    62
    +0.12
    0.01
    288
    +0.12
    0.01
    Negative Logits
    ħĭ
    -0.80
    soever
    -0.72
    ij士
    -0.72
     millenn
    -0.68
    sterdam
    -0.67
    ĻĤ
    -0.66
     Parables
    -0.66
    Interstitial
    -0.65
    bleacher
    -0.65
    gdala
    -0.64
    POSITIVE LOGITS
    aged
    1.46
    aging
    1.12
    ages
    1.09
    agement
    0.98
     levers
    0.92
    ulic
    0.91
    agements
    0.91
    age
    0.90
     lever
    0.84
    ror
    0.84
    Activations Density 0.005%

    No Known Activations