Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APISteerSAE EvalsBlog/PodcastNEWSlackPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlog/PodcastGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 3072
    5. 8-RES_FS3072-JB
    6. 392
    Prev
    Next
    INDEX
    Explanations

    terms related to detachment, dissociation, and extraction in the context of psychological coping mechanisms

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs3072-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_3072
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    3,072
    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
    639
    +0.11
    0.5%
    472
    +0.11
    0.5%
    382
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.11
    0.11
    665
    +0.11
    0.08
    335
    +0.10
    0.06
    Negative Logits
     heartbeat
    -0.62
    ounce
    -0.60
    olver
    -0.59
    dream
    -0.59
    umbers
    -0.59
    ãĥ¥
    -0.58
    enegger
    -0.58
    iop
    -0.57
    emate
    -0.56
    cond
    -0.55
    POSITIVE LOGITS
     from
    1.18
    from
    1.15
    From
    1.10
     FROM
    1.09
     From
    1.04
    owship
    0.81
    away
    0.76
    ring
    0.75
    comings
    0.68
    egu
    0.67
    Activations Density 1.212%

    No Known Activations