© 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 E2E Recon
    9. 6-RES_SCEFR-AJT
    11. 13749
    Prev
    Next
    INDEX
    Explanations

    descriptions of constructing and using a raft

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    neuronpedia/gpt2-small__res_scefr-ajt/6-res_scefr-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
    316
    +0.06
    0.5%
    575
    +0.06
    0.5%
    343
    +0.06
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.06
    0.06
    326
    +0.06
    0.04
    575
    +0.06
    0.05
    Negative Logits
    eyes
    -0.44
    places
    -0.44
    arial
    -0.44
    uate
    -0.39
     Lamar
    -0.38
    uality
    -0.38
    mary
    -0.36
    phis
    -0.36
    heimer
    -0.36
    pta
    -0.35
    POSITIVE LOGITS
    loads
    0.52
    ikk
    0.45
    boat
    0.45
    asso
    0.43
    anas
    0.43
    wright
    0.42
     boats
    0.41
     boat
    0.40
    load
    0.40
    boats
    0.39
    Activations Density 0.385%

    No Known Activations