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. 34409
    Prev
    Next
    INDEX
    Explanations

    references to external sources and articles, possibly related to research, publications, and websites

    oai_token-act-pair · gpt-3.5-turbo

    URL addresses and associated components such as domain extensions (.org, .com), specific paths (/news, /page), and document formats (.pdf, .html)

    oai_token-act-pair · gpt-3.5-turboTriggered by @danbraun
    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
    594
    +0.14
    0.5%
    331
    +0.13
    0.5%
    314
    +0.13
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    314
    +0.14
    -0.02
    331
    +0.13
    0.11
    145
    +0.13
    0.07
    Negative Logits
     Sinai
    -0.77
     Yel
    -0.74
     «
    -0.73
     Ashes
    -0.70
     Harden
    -0.69
     Economist
    -0.69
     compounded
    -0.68
     Reed
    -0.68
     Himself
    -0.68
     Rhodes
    -0.68
    POSITIVE LOGITS
    /)
    1.23
    /,
    1.19
    /.
    1.11
    ascript
    1.09
    /#
    1.08
    /?
    1.05
    _
    1.03
    /_
    0.99
    ournals
    0.99
    utorial
    0.95
    Activations Density 10.959%

    No Known Activations