© 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. 33783
    Prev
    Next
    INDEX
    Explanations

    instances where the phrase "go ahead" is used

    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
    665
    +0.14
    0.6%
    447
    +0.13
    0.6%
    679
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    665
    +0.14
    0.13
    354
    +0.13
    0.07
    551
    +0.11
    0.06
    Negative Logits
    mini
    -0.70
    Pokémon
    -0.67
    Net
    -0.67
    brid
    -0.66
    Effective
    -0.66
    cci
    -0.66
    Ranked
    -0.66
    Phoenix
    -0.65
    CVE
    -0.65
    vol
    -0.63
    POSITIVE LOGITS
     reproduce
    0.72
     undet
    0.72
    erk
    0.69
     anyway
    0.67
     unnoticed
    0.67
    acht
    0.65
     Skip
    0.65
     blindly
    0.64
     rusher
    0.63
     Continue
    0.63
    Activations Density 6.606%

    No Known Activations