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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsBlogSlackPrivacy & 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. 13559
    Prev
    Next
    INDEX
    Explanations

    items that can be purchased or considered for purchase

    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
    +0.23
    1.1%
    567
    +0.10
    0.4%
    165
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    652
    +0.23
    0.09
    270
    +0.10
    0.04
    567
    +0.09
    0.07
    Negative Logits
    xual
    -0.95
    ĸļ
    -0.84
    ulty
    -0.76
    amination
    -0.75
    ilial
    -0.71
    icable
    -0.69
    ogenesis
    -0.69
    ional
    -0.69
    ngth
    -0.69
    ukemia
    -0.68
    POSITIVE LOGITS
     groceries
    1.16
     cheapest
    0.97
     cheap
    0.96
     cheaply
    0.90
     lottery
    0.87
     Crate
    0.86
     discounted
    0.84
     tickets
    0.83
     licences
    0.78
     discount
    0.77
    Activations Density 15.280%

    No Known Activations