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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 8-RES-JB
    6. 4929
    Prev
    Next
    INDEX
    Explanations

    phrases indicating a requirement or instruction for certain actions to be taken

    oai_token-act-pair · gpt-3.5-turbo

    the word "all" or variations of it in different contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    rir
    -0.77
     nowhere
    -0.72
    IND
    -0.64
    robe
    -0.63
    aminer
    -0.63
    potion
    -0.62
     Few
    -0.62
    shed
    -0.62
    Few
    -0.60
    dl
    -0.60
    POSITIVE LOGITS
    ocating
    1.35
    owing
    1.20
    ocate
    1.12
    ocated
    1.10
    ocation
    1.08
    iances
    1.07
    ocations
    1.06
     kinds
    1.05
    owed
    1.05
    igators
    1.04
    Activations Density 0.130%

    No Known Activations