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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Feature Splitting for GPT2-Small
    5. GPT2-Small
    7. Residual Feature Splitting 49152
    9. 8-RES_FS49152-JB
    11. 45641
    Prev
    Next
    INDEX
    Explanations

    phrases that express contradictions or contrasting ideas

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs49152-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    49,152
    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
    480
    +0.44
    2.2%
    158
    +0.11
    0.5%
    351
    +0.09
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    351
    +0.44
    0.03
    480
    +0.11
    0.00
    384
    +0.09
    0.03
    Negative Logits
     Cosponsors
    -0.76
    chool
    -0.74
    ogun
    -0.72
    itsch
    -0.70
    lass
    -0.69
    ©¶æ
    -0.63
    psc
    -0.60
    abase
    -0.60
    unes
    -0.59
    cients
    -0.58
    POSITIVE LOGITS
     satisfaction
    0.92
     disappointment
    0.89
     excitement
    0.89
     sadness
    0.88
     certainty
    0.88
     acknowledgement
    0.87
     acknowledgment
    0.85
     happiness
    0.83
     surprises
    0.83
     degradation
    0.82
    Activations Density 0.311%

    No Known Activations