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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsBlogSlackPrivacy & 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. 45559
    Prev
    Next
    INDEX
    Explanations

    information related to Docker containers, deployment, and networking

    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
    288
    +0.24
    1.0%
    447
    +0.20
    0.8%
    216
    +0.16
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    326
    +0.24
    0.19
    740
    +0.20
    0.10
    288
    +0.16
    0.17
    Negative Logits
    teenth
    -0.89
    manship
    -0.89
    lly
    -0.88
    sembly
    -0.85
    choes
    -0.85
    tale
    -0.83
    lasses
    -0.83
    eways
    -0.80
    reditary
    -0.80
    ãĤ®
    -0.80
    POSITIVE LOGITS
    ainers
    1.24
     containers
    1.18
     container
    1.10
    icum
    1.07
     Container
    0.97
     Swarm
    0.97
     minimized
    0.95
    arium
    0.88
     reefs
    0.86
     Hub
    0.83
    Activations Density 6.427%

    No Known Activations