© 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. 20043
    Prev
    Next
    INDEX
    Explanations

    words related to strength and resilience

    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
    251
    +0.13
    0.6%
    503
    +0.11
    0.5%
    98
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    251
    +0.13
    0.03
    503
    +0.11
    0.02
    98
    +0.10
    0.02
    Negative Logits
     honour
    -0.66
     Halls
    -0.64
     Franz
    -0.62
     Masters
    -0.62
     approve
    -0.61
     Apprentice
    -0.61
     STATES
    -0.61
    İĭ
    -0.60
     exclusive
    -0.59
     Incarnation
    -0.59
    POSITIVE LOGITS
    ooky
    1.23
    anish
    1.15
    rit
    1.11
    rawl
    1.11
    atter
    1.09
    itt
    1.07
    acer
    1.07
    ooked
    1.07
    irtual
    1.07
    leen
    1.06
    Activations Density 0.011%

    No Known Activations