© 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. OpenAI · Sparse Autoencoder for GPT2-Small - v5 2024
    3. GPT2-Small
    4. Attention Out - 32k
    5. 3-ATT_32K-OAI
    6. 3515
    Prev
    Next
    INDEX
    Explanations

    references to technology and its integration with various materials or phenomena

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-att_32k-oai
    Configuration
    jbloom/GPT2-Small-OAI-v5-32k-attn-out-SAEs/v5_32k_layer_3
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    32,768
    Data Type
    torch.float32
    Hook Name
    blocks.3.hook_attn_out
    Hook Layer
    3
    Architecture
    standard
    Context Size
    64
    Dataset
    Skylion007/openwebtext
    Activation Function
    topk
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    215
    +0.11
    0.5%
    548
    +0.10
    0.5%
    620
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    264
    +0.11
    0.13
    620
    +0.10
    0.15
    481
    +0.09
    0.22
    Negative Logits
    icz
    -0.78
    leeve
    -0.78
    scl
    -0.76
    hower
    -0.71
     QC
    -0.70
    forth
    -0.70
    onga
    -0.70
    ledge
    -0.69
    beck
    -0.68
    dash
    -0.67
    POSITIVE LOGITS
    Flor
    0.71
    sylv
    0.70
    ĨĴ
    0.69
    yrs
    0.66
    INFO
    0.66
     Champions
    0.65
    Synopsis
    0.65
    Activity
    0.64
    girls
    0.64
     Kardash
    0.62
    Activations Density 12.974%

    No Known Activations