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. 0-RES-JB
    6. 8615
    Prev
    Next
    INDEX
    Explanations

    phrases related to the concept of gathering and collecting

    oai_token-act-pair · gpt-3.5-turbo

    instances of the term "accumulate" or variations thereof

    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
     Moons
    -0.77
     Milo
    -0.71
     Juliet
    -0.71
     pizz
    -0.70
     pizza
    -0.69
     Harmony
    -0.68
    Rob
    -0.67
     moons
    -0.67
    Mars
    -0.67
     Wel
    -0.66
    POSITIVE LOGITS
     accumulated
    3.35
     accumulate
    3.08
     accumulating
    2.92
     accum
    2.66
     accumulation
    2.64
     amassed
    2.19
     accrued
    2.17
     cumulative
    1.66
     buildup
    1.56
     piled
    1.42
    Activations Density 0.022%

    No Known Activations