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. 34415
    Prev
    Next
    INDEX
    Explanations

    ing verbs that stretch out for multiple syllables, intensifiers like "very", and discussions involving topics of power or leadership

    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
    +0.18
    0.9%
    102
    +0.12
    0.6%
    240
    +0.12
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    232
    +0.18
    0.07
    240
    +0.12
    0.07
    725
    +0.12
    0.06
    Negative Logits
    ãĤ´ãĥ³
    -0.86
    ãĥ¯
    -0.77
    bilt
    -0.69
    éĹĺ
    -0.67
    fman
    -0.67
    ãĥ´ãĤ¡
    -0.64
    enhagen
    -0.64
    ãĥ¼ãĥĨãĤ£
    -0.63
    EStream
    -0.63
     Clause
    -0.62
    POSITIVE LOGITS
    oused
    1.32
    ousing
    1.20
    acking
    1.16
    ulk
    1.14
    idd
    1.13
    anging
    1.11
    awk
    1.11
    anky
    1.10
    olly
    1.10
    ashing
    1.08
    Activations Density 6.575%

    No Known Activations