© 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. 18926
    Prev
    Next
    INDEX
    Explanations

    phrases related to determination and achievement

    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
    505
    +0.11
    0.5%
    767
    +0.10
    0.5%
    307
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    665
    +0.11
    0.06
    608
    +0.10
    0.05
    282
    +0.10
    0.04
    Negative Logits
    soDeliveryDate
    -0.89
    anchester
    -0.86
    guards
    -0.75
    requires
    -0.70
    minus
    -0.67
     flagged
    -0.66
    houses
    -0.65
    wise
    -0.65
    vo
    -0.65
    nton
    -0.64
    POSITIVE LOGITS
     perfection
    1.16
     revenge
    1.14
     maximize
    1.11
     vengeance
    1.09
     emulate
    1.04
     fulfill
    1.03
     reproduce
    1.03
     fulfil
    0.99
     succeed
    0.97
     pursue
    0.94
    Activations Density 0.184%

    No Known Activations