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. 34131
    Prev
    Next
    INDEX
    Explanations

    phrases indicating no more of something

    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
    393
    +0.15
    0.6%
    754
    +0.12
    0.5%
    737
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    726
    +0.15
    0.10
    647
    +0.12
    0.07
    614
    +0.11
    0.04
    Negative Logits
    irez
    -0.83
    isite
    -0.73
    iage
    -0.71
    erest
    -0.71
    omnia
    -0.70
    itime
    -0.68
    cius
    -0.67
    OS
    -0.66
    OCK
    -0.66
    ress
    -0.66
    POSITIVE LOGITS
     than
    0.99
     whatsoever
    0.82
     nor
    0.77
     ado
    0.73
    than
    0.72
    Fake
    0.71
     excuses
    0.71
    cial
    0.69
    xious
    0.65
     info
    0.64
    Activations Density 4.154%

    No Known Activations