© 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 E2E Recon
    5. 6-RES_SCEFR-AJT
    6. 13286
    Prev
    Next
    INDEX
    Explanations

    links and relationships between different concepts or variables

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    neuronpedia/gpt2-small__res_scefr-ajt/6-res_scefr-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
    701
    +0.15
    0.6%
    346
    +0.12
    0.5%
    195
    +0.12
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    195
    +0.15
    0.03
    346
    +0.12
    0.05
    261
    +0.12
    0.04
    Negative Logits
    sburg
    -0.89
    fl
    -0.87
    disk
    -0.86
    dp
    -0.77
    onew
    -0.75
    wagen
    -0.75
    ãģĤ
    -0.73
    adobe
    -0.73
    mac
    -0.73
    Cmd
    -0.73
    POSITIVE LOGITS
    abolic
    1.15
    SHIP
    1.04
    abol
    0.96
     linking
    0.94
     knot
    0.88
     coefficient
    0.88
     implicated
    0.86
     links
    0.86
     link
    0.85
     connecting
    0.84
    Activations Density 0.557%

    No Known Activations