© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Apollo Research · Taylor · Identifying Functionally Important Features with End-to-End Sparse Dictionary Learning
    5. GPT2-Small
    7. Residual Similar CE Local
    9. 6-RES_SCL-AJT
    11. 40423
    Prev
    Next
    INDEX
    Explanations

    references to specific dates and events in news articles

    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
    640
    +0.21
    0.8%
    447
    +0.16
    0.6%
    393
    +0.13
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    640
    +0.21
    0.13
    745
    +0.16
    -0.06
    373
    +0.13
    -0.09
    Negative Logits
     accordingly
    -0.95
     disgu
    -0.91
     advert
    -0.85
     proced
    -0.85
     prey
    -0.85
     aside
    -0.83
     harmless
    -0.82
     dividends
    -0.82
     preced
    -0.82
     immunity
    -0.81
    POSITIVE LOGITS
    Getty
    1.31
    AFP
    1.30
    Courtesy
    1.28
    ËĪ
    1.27
    Flickr
    1.26
    CBC
    1.21
    Bloomberg
    1.20
    HT
    1.09
    Staff
    1.05
    Digital
    1.04
    Activations Density 4.303%

    No Known Activations