© 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. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    5. GPT2-Small
    7. Residual Stream
    9. 1-RES-JB
    11. 20212
    Prev
    Next
    INDEX
    Explanations

    references to pencils, which may imply a focus on writing or drawing activities

    oai_token-act-pair · gpt-3.5-turbo

    references to pencils

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 1-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.hook_resid_pre
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    Data Type
    torch.float32
    Hook Point
    blocks.1.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    1
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    EEE
    -0.79
    ãĥ´
    -0.78
    Interstitial
    -0.76
    κ
    -0.75
    Examples
    -0.69
    SPONSORED
    -0.68
    thy
    -0.68
    ths
    -0.67
    âķIJâķIJ
    -0.67
    FI
    -0.66
    POSITIVE LOGITS
     pencil
    1.26
    edo
    0.90
    cil
    0.88
    clip
    0.84
     pens
    0.82
    matical
    0.81
    ues
    0.79
     sketches
    0.77
    knife
    0.77
     charcoal
    0.76
    Activations Density 0.003%

    No Known Activations