© 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
    3. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    5. GPT2-Small
    7. Residual Stream
    9. 10-RES-JB
    11. 17288
    Prev
    Next
    INDEX
    Explanations

    names of individuals followed by a past-tense verb 'went'

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "went" in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    10
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    creen
    -0.77
    ________________
    -0.65
    icio
    -0.62
    adjusted
    -0.60
    ullah
    -0.60
     Allaah
    -0.60
     ---------
    -0.60
    hare
    -0.59
    ocated
    -0.58
    eers
    -0.57
    POSITIVE LOGITS
     forth
    0.97
     overboard
    0.97
     ballistic
    0.95
     undercover
    0.90
     viral
    0.88
     rogue
    0.88
     undrafted
    0.86
     MIA
    0.86
     ahead
    0.84
    vt
    0.82
    Activations Density 0.062%

    No Known Activations