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. 4-RES-JB
    6. 3345
    Prev
    Next
    INDEX
    Explanations

    specific phrases related to organized groups or structures, specifically involving a hierarchical structure with "and file" being the most prominent

    oai_token-act-pair · gpt-3.5-turbo

    terms related to evaluations or assessments of groups or activities

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 4-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    4
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    skill
    -0.77
    Chat
    -0.68
    Tag
    -0.67
     Scan
    -0.67
     Username
    -0.66
    say
    -0.66
    Spot
    -0.63
    idel
    -0.63
     Osw
    -0.63
    SN
    -0.62
    POSITIVE LOGITS
     iceberg
    0.82
     fallacy
    0.69
    eer
    0.66
    rollers
    0.65
     boosters
    0.65
    âĦ¢:
    0.62
    legates
    0.62
     supervisors
    0.62
     hindsight
    0.62
     debacle
    0.61
    Activations Density 0.486%

    No Known Activations