© 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
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 3-RES-JB
    6. 11169
    Prev
    Next
    INDEX
    Explanations

    instances of the word "too" in sentences

    oai_token-act-pair · gpt-3.5-turbo

    the word "too" and its variations, indicating a focus on excess or repetition

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    3
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    inates
    -0.67
    è¦ļéĨĴ
    -0.64
    uate
    -0.64
    inarily
    -0.61
    riots
    -0.61
    icipated
    -0.60
    inators
    -0.60
    eer
    -0.60
    Rus
    -0.60
    craft
    -0.60
    POSITIVE LOGITS
    ħĭ
    0.67
    oths
    0.67
     bulky
    0.65
     abst
    0.62
    akening
    0.61
    pes
    0.61
     much
    0.60
    ls
    0.60
     stren
    0.60
     tempting
    0.59
    Activations Density 0.038%

    No Known Activations