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. 34175
    Prev
    Next
    INDEX
    Explanations

    informational articles or blog posts with a focus on technical topics and coding concepts

    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
    +0.27
    1.8%
    393
    +0.10
    0.7%
    260
    +0.08
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    270
    +0.27
    0.06
    393
    +0.10
    0.22
    97
    +0.08
    0.06
    Negative Logits
    chen
    -0.55
    quez
    -0.52
    rang
    -0.50
    chens
    -0.50
    ches
    -0.49
    qua
    -0.47
    held
    -0.47
    rift
    -0.46
    hesis
    -0.45
    ulla
    -0.45
    POSITIVE LOGITS
    =#
    0.60
     how
    0.51
     topics
    0.49
     halfway
    0.45
     why
    0.42
     trivial
    0.41
     overcoming
    0.41
     bund
    0.40
     calib
    0.40
     convol
    0.40
    Activations Density 9.335%

    No Known Activations