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 L0 E2E Recon
    5. 2-RES_SLEFR-AJT
    6. 5410
    Prev
    Next
    INDEX
    Explanations

    the word "nearby."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    neuronpedia/gpt2-small__res_slefr-ajt/2-res_slefr-ajt
    Prompts (Dashboard)
    12,288 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    46,080
    Data Type
    torch.float32
    Hook Point
    blocks.2.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    apollo-research/Skylion007-openwebtext-tokenizer-gpt2
    Hook Point Layer
    2
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    709
    +0.10
    0.5%
    740
    +0.09
    0.5%
    762
    +0.08
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    471
    +0.10
    0.02
    327
    +0.09
    0.01
    253
    +0.08
    0.01
    Negative Logits
    NEY
    -0.71
    RANT
    -0.63
    zai
    -0.62
    amines
    -0.60
    ãĥİ
    -0.59
    chin
    -0.57
    é¾į
    -0.57
    士
    -0.56
    cv
    -0.56
    Admin
    -0.55
    POSITIVE LOGITS
     nearby
    0.78
    resent
    0.62
     adjacent
    0.55
     neighbor
    0.54
     din
    0.53
     risen
    0.52
     glance
    0.51
    oreal
    0.50
    orate
    0.50
     neighbors
    0.49
    Activations Density 0.007%

    No Known Activations