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

    Neuronpedia

    APICircuit 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. 45395
    Prev
    Next
    INDEX
    Explanations

    academic and philosophical concepts related to social issues and political theory, especially those discussing capitalism, media, and governmental policies

    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
    99
    +0.07
    0.5%
    304
    +0.06
    0.4%
    403
    +0.06
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.07
    0.25
    326
    +0.06
    0.22
    99
    +0.06
    0.12
    Negative Logits
    iosity
    -0.54
    ctica
    -0.49
    RED
    -0.49
    idity
    -0.48
    atility
    -0.48
    anza
    -0.47
    iflower
    -0.46
    Coach
    -0.46
     Rossi
    -0.46
    rating
    -0.45
    POSITIVE LOGITS
    hower
    0.58
    ervative
    0.53
    merce
    0.53
    hops
    0.49
    auga
    0.48
    emouth
    0.47
    IME
    0.44
    paces
    0.44
    otin
    0.44
    yg
    0.43
    Activations Density 7.913%

    No Known Activations