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. 5888
    Prev
    Next
    INDEX
    Explanations

    occurrences of the word "avail" and various forms of the word "exhaust."

    oai_token-act-pair · gpt-3.5-turbo

    words and phrases related to availability and presence

    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
    acid
    -0.80
     Lonely
    -0.77
    essor
    -0.77
     Samurai
    -0.70
    ocracy
    -0.68
    OLOGY
    -0.66
     Balanced
    -0.65
    throp
    -0.65
     Mobility
    -0.64
    nery
    -0.64
    POSITIVE LOGITS
     avail
    3.93
     exhaust
    1.07
     dispos
    1.00
     dispose
    0.96
     delight
    0.94
     includ
    0.93
     COVER
    0.91
     accomp
    0.91
     baff
    0.90
     rom
    0.84
    Activations Density 0.058%

    No Known Activations