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. 40272
    Prev
    Next
    INDEX
    Explanations

    phrases related to detailed events or descriptions

    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
    +0.16
    1.2%
    393
    +0.07
    0.5%
    484
    +0.07
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    393
    +0.16
    0.38
    608
    +0.07
    0.28
    385
    +0.07
    0.20
    Negative Logits
    soDeliveryDate
    -0.46
    hap
    -0.45
    soType
    -0.44
    ety
    -0.42
    SPONSORED
    -0.41
    heed
    -0.40
    hid
    -0.39
     minds
    -0.39
    ationally
    -0.39
    arta
    -0.39
    POSITIVE LOGITS
     tears
    0.48
     laughter
    0.46
    OTOS
    0.45
     applause
    0.44
    ãĥ©ãĥ³
    0.44
     gunfire
    0.43
     tragedy
    0.42
     disappointment
    0.41
    OOOOOOOO
    0.41
     goodbye
    0.41
    Activations Density 17.637%

    No Known Activations