© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    5. GPT2-Small
    7. Residual Stream
    9. 2-RES-JB
    11. 337
    Prev
    Next
    INDEX
    Explanations

    the suffix "ival" in words

    oai_token-act-pair · gpt-4-turbo

    mentions of actions or processes related to shedding skin or removing body parts

    oai_token-act-pair · gpt-3.5-turbo

    terms related to evaluation or assessment

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 2-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    2
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    built
    -0.69
    zyme
    -0.66
    haar
    -0.61
     blazing
    -0.61
     clipboard
    -0.61
    lessly
    -0.60
    digit
    -0.58
    upon
    -0.56
     uploading
    -0.56
    ipple
    -0.55
    POSITIVE LOGITS
    ival
    1.35
    iott
    0.82
    ife
    0.81
    cia
    0.75
    ÃįÃį
    0.73
    ive
    0.71
    ency
    0.71
    ocity
    0.70
    rl
    0.69
    inal
    0.69
    Activations Density 0.008%

    No Known Activations