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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & 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 E2E Recon
    5. 6-RES_SCEFR-AJT
    6. 4424
    Prev
    Next
    INDEX
    Explanations

    references to the names "Anna" and "Elsa", as well as words related to baking or desserts

    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
    14
    +0.15
    0.6%
    78
    +0.13
    0.5%
    99
    +0.13
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.15
    0.14
    326
    +0.13
    0.11
    99
    +0.13
    0.09
    Negative Logits
    ours
    -1.02
    aments
    -1.01
    ors
    -0.94
    igious
    -0.91
    ugu
    -0.89
    ician
    -0.86
    racted
    -0.83
    otyp
    -0.82
    ually
    -0.81
    ributed
    -0.81
    POSITIVE LOGITS
    flake
    1.15
    blind
    0.83
     Darling
    0.81
     Leopard
    0.81
    âĺħâĺħ
    0.79
     desserts
    0.79
    âĹ¼
    0.77
     Wonderland
    0.77
    ãĤ¼ãĤ¦ãĤ¹
    0.77
    ħĭ
    0.76
    Activations Density 1.780%

    No Known Activations