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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2026
    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. 33267
    Prev
    Next
    INDEX
    Explanations

    phone communication-related words and actions

    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
    +0.18
    0.9%
    695
    +0.11
    0.5%
    150
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    385
    +0.18
    0.16
    437
    +0.11
    0.07
    124
    +0.10
    0.13
    Negative Logits
    alde
    -0.68
    acre
    -0.68
    ottest
    -0.67
    Ranked
    -0.66
    fruit
    -0.66
    elson
    -0.66
    inctions
    -0.65
    rb
    -0.65
    cit
    -0.63
    Cola
    -0.62
    POSITIVE LOGITS
     911
    1.08
     hotline
    0.84
     voic
    0.84
    igraph
    0.80
     caller
    0.78
     timeout
    0.75
     999
    0.72
    call
    0.71
     Fargo
    0.71
     CALL
    0.70
    Activations Density 10.078%

    No Known Activations