© 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
    3. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    5. GPT2-Small
    7. Residual Stream
    9. 7-RES-JB
    11. 1633
    Prev
    Next
    INDEX
    Explanations

    instances of the word "Still" at the beginning of a sentence

    oai_token-act-pair · gpt-3.5-turbo

    the repeated phrase "Still," indicating a continuation or emphasis in the text

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 7-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    7
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    duct
    -0.71
     è£ıç
    -0.65
    alties
    -0.64
    ultimate
    -0.59
    ASE
    -0.59
    agus
    -0.58
     MAG
    -0.57
    CVE
    -0.57
    gd
    -0.57
    ochond
    -0.56
    POSITIVE LOGITS
    birth
    0.85
    heres
    0.76
    noon
    0.74
     reeling
    0.73
    adays
    0.73
    born
    0.67
     unanswered
    0.66
     Despair
    0.65
     undecided
    0.64
    ipop
    0.63
    Activations Density 0.032%

    No Known Activations