© 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. 45752
    Prev
    Next
    INDEX
    Explanations

    technical specifications related to computer hardware components, particularly CPUs

    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
    342
    +0.09
    0.7%
    447
    +0.08
    0.6%
    623
    +0.08
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    6
    +0.09
    0.13
    640
    +0.08
    0.18
    255
    +0.08
    0.08
    Negative Logits
    ospital
    -0.52
     sidel
    -0.50
    fits
    -0.46
    pez
    -0.46
    ights
    -0.45
    ards
    -0.45
    pload
    -0.45
    etheless
    -0.45
    selves
    -0.44
    atile
    -0.43
    POSITIVE LOGITS
    xit
    0.59
    gart
    0.54
    erity
    0.52
    opsis
    0.51
    lli
    0.49
    nces
    0.47
    lington
    0.46
    Found
    0.46
     Core
    0.46
     Catalyst
    0.45
    Activations Density 8.135%

    No Known Activations