© 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. 4-RES-JB
    11. 8558
    Prev
    Next
    INDEX
    Explanations

    words related to death or strong negative outcomes

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "dead" and related phrases indicating death or being lifeless

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

    No Comments

    Negative Logits
    orney
    -0.76
    ãĥ´ãĤ¡
    -0.73
    ECA
    -0.72
     CLSID
    -0.68
    ATIONS
    -0.67
    ickr
    -0.66
    ãĥķãĤ©
    -0.65
    fty
    -0.64
    zzle
    -0.64
    doi
    -0.63
    POSITIVE LOGITS
    beat
    1.00
    liest
    0.92
    locked
    0.89
    lier
    0.84
    weight
    0.83
    dead
    0.83
    heading
    0.83
    lift
    0.83
    spin
    0.82
    hound
    0.80
    Activations Density 0.018%

    No Known Activations