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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsBlogSlackPrivacy & 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. 4281
    Prev
    Next
    INDEX
    Explanations

    words related to medical conditions, specifically focusing on psychiatric disorders like schizophrenia, anxiety, psychopathy, and epilepsy

    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
    714
    +0.09
    0.5%
    68
    +0.08
    0.4%
    472
    +0.08
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.09
    0.18
    588
    +0.08
    0.09
    343
    +0.08
    0.08
    Negative Logits
    htaking
    -0.74
    boro
    -0.69
    ACTED
    -0.66
    steen
    -0.62
    20439
    -0.62
    IRD
    -0.61
     ç¥ŀ
    -0.60
    hole
    -0.56
    PLAY
    -0.56
    uggets
    -0.56
    POSITIVE LOGITS
    ī
    0.78
    ĩ
    0.73
    ģ
    0.71
    ohydrate
    0.71
    ians
    0.70
    ines
    0.69
    ic
    0.69
    Ĺ
    0.68
    ļ
    0.66
    iator
    0.66
    Activations Density 4.766%

    No Known Activations