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. 4540
    Prev
    Next
    INDEX
    Explanations

    information related to social media platforms, especially Facebook and Twitter

    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
    292
    +0.11
    0.5%
    373
    +0.10
    0.5%
    13
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    706
    +0.11
    0.11
    167
    +0.10
    -0.02
    592
    +0.10
    0.06
    Negative Logits
    stood
    -0.69
    grass
    -0.68
    ppa
    -0.66
     Pont
    -0.66
     Rite
    -0.64
     ACS
    -0.63
    abl
    -0.62
    uits
    -0.62
    rano
    -0.61
     Rollins
    -0.60
    POSITIVE LOGITS
    ileaks
    0.84
    Facebook
    0.80
    Twitter
    0.72
     Email
    0.71
     Updates
    0.71
     Expand
    0.70
     Twitter
    0.70
     largeDownload
    0.70
     Photograph
    0.70
     Account
    0.69
    Activations Density 6.185%

    No Known Activations