© 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. 13408
    Prev
    Next
    INDEX
    Explanations

    technical terms, possibly related to computer science or programming

    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
    102
    +0.16
    0.6%
    616
    +0.14
    0.5%
    709
    +0.13
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    468
    +0.16
    0.05
    102
    +0.14
    0.05
    26
    +0.13
    0.05
    Negative Logits
     compe
    -0.94
     GOODMAN
    -0.86
     Lobby
    -0.86
    terson
    -0.86
    å§«
    -0.81
     horizont
    -0.81
    FactoryReloaded
    -0.78
     hospitality
    -0.78
     Awakens
    -0.77
     coerc
    -0.77
    POSITIVE LOGITS
    umbling
    1.56
    umbled
    1.54
    iddles
    1.53
    angers
    1.52
    acking
    1.47
    ipper
    1.45
    ascal
    1.41
    attle
    1.40
    ipp
    1.39
    iddle
    1.39
    Activations Density 0.570%

    No Known Activations