© 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. 0-RES-JB
    6. 22710
    Prev
    Next
    INDEX
    Explanations

    specific verbs related to memory and recall

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "remember."

    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
     Orders
    -0.62
     District
    -0.62
    ities
    -0.60
     Corporation
    -0.59
     encro
    -0.59
    ism
    -0.58
    ially
    -0.57
    usc
    -0.57
     sub
    -0.56
     gulf
    -0.56
    POSITIVE LOGITS
     remember
    3.63
    remember
    2.61
     remembers
    2.40
     recall
    2.36
     remembered
    2.00
     remembering
    1.93
    Remember
    1.91
     Remember
    1.91
     remem
    1.71
     recalls
    1.58
    Activations Density 0.015%

    No Known Activations