© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI 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. 4479
    Prev
    Next
    INDEX
    Explanations

    personal pronouns referring to oneself or others in reflective or self-analytical contexts

    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
    715
    +0.11
    0.5%
    503
    +0.10
    0.5%
    96
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    715
    +0.11
    0.04
    571
    +0.10
    0.04
    656
    +0.09
    0.04
    Negative Logits
    interstitial
    -0.74
    Reviewer
    -0.65
    quartered
    -0.64
    ©¶æ¥µ
    -0.60
    Sharp
    -0.57
     ILCS
    -0.57
     Levine
    -0.56
    âĸijâĸij
    -0.56
    Ts
    -0.56
    Extreme
    -0.56
    POSITIVE LOGITS
     succeed
    1.00
     weren
    0.98
     hadn
    0.92
     succeeds
    0.88
     ever
    0.85
     someday
    0.84
    're
    0.83
     wanna
    0.82
     were
    0.80
     fail
    0.80
    Activations Density 0.132%

    No Known Activations