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

    Neuronpedia

    APIAssistant AxisNEWCircuit 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. 15901
    Prev
    Next
    INDEX
    Explanations

    proper nouns ending in 'els'

    oai_token-act-pair · gpt-3.5-turbo

    mentions of "els" or similar patterns in various contexts

    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
     insecure
    -0.65
     reproduction
    -0.63
     misuse
    -0.62
     concurrent
    -0.61
     replay
    -0.59
     misc
    -0.59
     Mis
    -0.59
     grievance
    -0.59
     benefit
    -0.59
     compromised
    -0.59
    POSITIVE LOGITS
    els
    4.59
    ELS
    2.85
    eled
    2.57
    eling
    2.46
    el
    2.18
    elled
    1.85
    EL
    1.82
    elson
    1.81
    eln
    1.80
    elsen
    1.77
    Activations Density 0.008%

    No Known Activations