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

    Neuronpedia

    APICircuit 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. 1-RES-JB
    6. 19786
    Prev
    Next
    INDEX
    Explanations

    phrases or sentences indicating a direction, movement, or leadership towards a specific goal or place

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "headed" indicating direction or movement

    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
     Corpus
    -0.76
    EVA
    -0.72
    rehens
    -0.70
    urity
    -0.70
    Premium
    -0.64
    æŃ¦
    -0.63
    oret
    -0.60
    ressor
    -0.60
    ional
    -0.60
    Mit
    -0.59
    POSITIVE LOGITS
     headed
    0.95
     nodd
    0.89
     adolesc
    0.81
    heading
    0.76
     heading
    0.75
    pins
    0.74
    canon
    0.73
     nodding
    0.73
    wcs
    0.73
    lander
    0.72
    Activations Density 0.006%

    No Known Activations