© 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. 34758
    Prev
    Next
    INDEX
    Explanations

    descriptions of outdoor scenes and activities

    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
    570
    +0.12
    0.5%
    497
    +0.12
    0.5%
    308
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    640
    +0.12
    0.37
    393
    +0.12
    0.40
    745
    +0.11
    -0.27
    Negative Logits
    actionDate
    -0.96
    FN
    -0.86
    thood
    -0.85
     Cosponsors
    -0.79
    ħĭ
    -0.78
     portfolios
    -0.76
    ortment
    -0.73
    tein
    -0.72
    folios
    -0.71
    oyal
    -0.71
    POSITIVE LOGITS
     haze
    1.07
     humidity
    0.96
     showers
    0.89
     gravel
    0.85
    absor
    0.84
     precipitation
    0.82
     Glac
    0.82
     rains
    0.81
     intermitt
    0.80
     humid
    0.79
    Activations Density 27.712%

    No Known Activations