© 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. 40
    Prev
    Next
    INDEX
    Explanations

    mentions of the word "Ram" followed by a number, such as "Ram 9" or "Ram 8"

    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
    275
    +0.18
    0.6%
    288
    +0.18
    0.5%
    351
    +0.17
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.18
    0.07
    291
    +0.18
    0.05
    326
    +0.17
    0.06
    Negative Logits
    RAFT
    -1.02
    LECT
    -0.98
     kHz
    -0.95
    rium
    -0.95
    %%
    -0.94
     Pompe
    -0.93
    KI
    -0.92
    chnology
    -0.92
     Consent
    -0.92
    ternity
    -0.91
    POSITIVE LOGITS
    irez
    1.82
    ming
    1.75
    pling
    1.61
    blers
    1.50
    rod
    1.50
    mers
    1.47
    borgh
    1.40
    pled
    1.40
    pton
    1.40
    med
    1.35
    Activations Density 0.899%

    No Known Activations