© 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. 0-RES-JB
    11. 16426
    Prev
    Next
    INDEX
    Explanations

    words related to the surface of objects or locations

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    \\\\\\\\
    -0.66
    cens
    -0.66
    ards
    -0.65
    mad
    -0.64
     Dek
    -0.64
    beck
    -0.64
    crazy
    -0.63
    dad
    -0.62
    edy
    -0.62
    acha
    -0.62
    POSITIVE LOGITS
     surface
    3.86
     surfaces
    2.69
    surface
    2.62
     Surface
    2.02
     exterior
    1.30
     resur
    1.26
     substrate
    1.12
     surfaced
    1.12
     radar
    1.11
     underside
    1.11
    Activations Density 0.022%

    No Known Activations