© 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. 9490
    Prev
    Next
    INDEX
    Explanations

    phrases indicating the presence of specific content or material within a physical space

    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
    661
    +0.13
    0.6%
    756
    +0.11
    0.5%
    59
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    439
    +0.13
    0.02
    189
    +0.11
    0.01
    661
    +0.10
    0.01
    Negative Logits
    ansson
    -0.83
    roads
    -0.69
    Äĩ
    -0.67
     Shank
    -0.67
    paralle
    -0.66
    romy
    -0.64
    frog
    -0.63
    ESCO
    -0.62
    orks
    -0.62
    olyn
    -0.61
    POSITIVE LOGITS
     thereof
    1.06
     contents
    0.89
    iveness
    0.88
    ions
    0.88
    matter
    0.85
    ious
    0.81
     herein
    0.79
     contained
    0.79
     therein
    0.78
    ãĥ´ãĤ¡
    0.75
    Activations Density 0.050%

    No Known Activations