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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & 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. 34197
    Prev
    Next
    INDEX
    Explanations

    terms related to chat rooms, messaging, and online communication

    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
    +0.15
    0.9%
    288
    +0.12
    0.7%
    361
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    675
    +0.15
    0.13
    48
    +0.12
    0.11
    314
    +0.10
    -0.02
    Negative Logits
    ©¶æ
    -0.62
    profit
    -0.56
    CVE
    -0.54
    ĻĤ
    -0.49
     graded
    -0.49
    enses
    -0.49
    è¡
    -0.48
    engeance
    -0.48
    adjusted
    -0.48
     ESV
    -0.48
    POSITIVE LOGITS
    anooga
    1.03
    rooms
    0.86
    room
    0.85
    ters
    0.80
    bots
    0.78
    tered
    0.74
     rooms
    0.70
    ting
    0.69
     chat
    0.67
    acters
    0.66
    Activations Density 6.750%

    No Known Activations