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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & 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. 45096
    Prev
    Next
    INDEX
    Explanations

    dollar amounts and pricing information

    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
    +0.24
    1.0%
    385
    +0.13
    0.6%
    37
    +0.12
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    640
    +0.24
    0.22
    442
    +0.13
    0.08
    160
    +0.12
    0.14
    Negative Logits
     Gors
    -0.81
     exerc
    -0.73
     Boone
    -0.71
     Sed
    -0.70
     Tall
    -0.70
     fuzz
    -0.70
     pron
    -0.69
     quot
    -0.68
     Slam
    -0.68
     Ups
    -0.67
    POSITIVE LOGITS
    HOME
    1.05
    ©¶æ¥µ
    1.04
    $$$$
    1.03
    6000
    0.97
    100
    0.95
    5
    0.95
    250
    0.94
    iris
    0.94
    1
    0.94
    1600
    0.93
    Activations Density 6.093%

    No Known Activations