Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    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. 34435
    Prev
    Next
    INDEX
    Explanations

    keywords related to software development and programming

    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
    +0.33
    1.6%
    139
    +0.11
    0.5%
    363
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    139
    +0.33
    0.09
    257
    +0.11
    0.08
    715
    +0.10
    0.09
    Negative Logits
    20439
    -0.78
    aeda
    -0.75
    CLASSIFIED
    -0.72
    /-
    -0.72
     Princ
    -0.71
     Territories
    -0.71
    ortment
    -0.70
    uminati
    -0.70
     Thrones
    -0.67
     Principle
    -0.64
    POSITIVE LOGITS
    _
    0.87
    fn
    0.69
     <-
    0.66
    mong
    0.63
    fac
    0.62
     :=
    0.62
    method
    0.62
     xp
    0.61
    orial
    0.61
    cmd
    0.61
    Activations Density 11.815%

    No Known Activations