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. 40291
    Prev
    Next
    INDEX
    Explanations

    the word "closest" along with related terms indicating proximity

    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
    565
    +0.10
    0.5%
    543
    +0.09
    0.5%
    339
    +0.08
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    565
    +0.10
    0.07
    43
    +0.09
    0.06
    471
    +0.08
    0.06
    Negative Logits
    anity
    -0.61
    inse
    -0.60
    umen
    -0.60
    dir
    -0.58
    andals
    -0.58
    olor
    -0.57
    ipel
    -0.56
    hots
    -0.56
    oat
    -0.56
    tails
    -0.55
    POSITIVE LOGITS
     approximation
    0.75
     kindred
    0.66
     imaginable
    0.64
     neighbour
    0.63
     analogue
    0.62
     confid
    0.61
     competitor
    0.61
     closest
    0.60
     possible
    0.58
     allied
    0.58
    Activations Density 7.810%

    No Known Activations