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

    Neuronpedia

    APIAssistant AxisNEWCircuit 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. 13299
    Prev
    Next
    INDEX
    Explanations

    references to age or historical context

    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
    +0.21
    0.8%
    85
    +0.12
    0.5%
    173
    +0.11
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    102
    +0.21
    0.12
    119
    +0.12
    0.09
    173
    +0.11
    0.06
    Negative Logits
    utterstock
    -1.03
    Ĥİ
    -1.02
    rawdownloadcloneembedreportprint
    -0.92
    atography
    -0.92
     athlet
    -0.91
    illation
    -0.90
    amera
    -0.89
    rencies
    -0.89
    aminer
    -0.87
    plementation
    -0.87
    POSITIVE LOGITS
     fashioned
    1.69
    school
    1.12
    rums
    1.10
    est
    1.08
     timers
    1.08
    fashioned
    1.07
     standby
    1.05
    timer
    0.96
     school
    0.95
     fart
    0.93
    Activations Density 8.639%

    No Known Activations