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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 0-RES-JB
    6. 4714
    Prev
    Next
    INDEX
    Explanations

    phrases related to the front of physical spaces or objects

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "front."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    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
    uria
    -0.67
    leness
    -0.67
    seed
    -0.65
    achu
    -0.64
    ľ
    -0.64
    lys
    -0.63
    anian
    -0.62
     Romantic
    -0.62
    idium
    -0.61
     Species
    -0.60
    POSITIVE LOGITS
     front
    3.73
     fronts
    2.43
     Front
    2.34
    front
    2.31
    Front
    2.14
     rear
    1.68
     frontline
    1.49
     forefront
    1.28
     frontal
    1.14
     backend
    1.14
    Activations Density 0.016%

    No Known Activations