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 Local
    5. 6-RES_SCL-AJT
    6. 45117
    Prev
    Next
    INDEX
    Explanations

    phrases related to video games, Nintendo, and technology

    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
    606
    +0.09
    0.5%
    78
    +0.08
    0.4%
    277
    +0.07
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    388
    +0.09
    0.09
    326
    +0.08
    0.19
    620
    +0.07
    0.02
    Negative Logits
    hospital
    -0.58
    tackle
    -0.58
     chlorine
    -0.56
     surgeon
    -0.56
     gren
    -0.55
     wrench
    -0.54
    inished
    -0.54
    sett
    -0.53
    akespe
    -0.53
     tremend
    -0.52
    POSITIVE LOGITS
    XL
    0.69
    ettings
    0.68
    +,
    0.68
    +.
    0.67
     Anniversary
    0.67
     XL
    0.66
    XT
    0.65
     Wireless
    0.62
    display
    0.62
    âĦ¢:
    0.61
    Activations Density 6.851%

    No Known Activations