© 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. 2-RES-JB
    11. 23058
    Prev
    Next
    INDEX
    Explanations

    words containing the substring "unt"

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "unt" or variations of it, indicating a focus on themes of untrustworthiness or negatives surrounding a subject

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 2-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    2
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    anwhile
    -1.02
    */(
    -0.89
    hyde
    -0.78
    oples
    -0.78
     Mercury
    -0.71
     briefs
    -0.70
    ĺħ
    -0.69
    SHIP
    -0.68
    Reviewer
    -0.68
    姫
    -0.67
    POSITIVE LOGITS
    ainted
    1.01
    itled
    1.00
    rave
    0.98
    ruly
    0.96
    ouch
    0.96
    oward
    0.94
    ribut
    0.88
    uner
    0.87
    ied
    0.86
    agged
    0.86
    Activations Density 0.006%

    No Known Activations