© 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
    2. Apollo Research · Taylor · Identifying Functionally Important Features with End-to-End Sparse Dictionary Learning
    3. GPT2-Small
    4. Residual Similar CE Local
    5. 6-RES_SCL-AJT
    6. 35282
    Prev
    Next
    INDEX
    Explanations

    The neuron is looking for phrases related to biological cycles or mechanical cycles

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    neuronpedia/gpt2-small__res_scl-ajt/6-res_scl-ajt
    Prompts (Dashboard)
    12,288 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    46,080
    Data Type
    torch.float32
    Hook Point
    blocks.6.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    apollo-research/Skylion007-openwebtext-tokenizer-gpt2
    Hook Point Layer
    6
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    447
    +0.27
    1.2%
    326
    +0.16
    0.7%
    288
    +0.14
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    593
    +0.27
    0.06
    546
    +0.16
    0.07
    375
    +0.14
    0.03
    Negative Logits
    inez
    -0.99
    orage
    -0.92
    amina
    -0.90
    imus
    -0.90
    essors
    -0.89
    avez
    -0.83
    alty
    -0.81
    uckland
    -0.81
    awaru
    -0.81
    alties
    -0.81
    POSITIVE LOGITS
     cycle
    1.20
     Cycle
    0.98
     cycles
    0.98
    cycle
    0.93
     erg
    0.91
    cycles
    0.90
    cles
    0.84
     repeats
    0.83
     alternating
    0.78
     Delay
    0.78
    Activations Density 9.247%

    No Known Activations