© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    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. 13780
    Prev
    Next
    INDEX
    Explanations

    phrases related to email subscriptions or newsletters

    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
    339
    +0.09
    0.6%
    69
    +0.07
    0.5%
    447
    +0.07
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    339
    +0.09
    0.10
    176
    +0.07
    0.07
    640
    +0.07
    0.22
    Negative Logits
    minist
    -0.65
    ISC
    -0.54
    handedly
    -0.48
     detachment
    -0.47
    gd
    -0.45
    dk
    -0.45
    uana
    -0.44
    pled
    -0.43
     unpop
    -0.43
    mith
    -0.43
    POSITIVE LOGITS
     Email
    0.71
     Updates
    0.68
     Your
    0.65
    zens
    0.65
     Slate
    0.64
    Alert
    0.58
     Tickets
    0.56
     Emails
    0.56
    cloneembedreportprint
    0.56
     Recap
    0.55
    Activations Density 11.100%

    No Known Activations