© 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. Apollo Research · Taylor · Identifying Functionally Important Features with End-to-End Sparse Dictionary Learning
    5. GPT2-Small
    7. Residual Similar CE Local
    9. 6-RES_SCL-AJT
    11. 13637
    Prev
    Next
    INDEX
    Explanations

    phrases related to gaining or losing positive outcomes or benefits

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    neuronpedia/gpt2-small__res_scl-ajt/6-res_scl-ajt
    Prompts (Dashboard)
    12,288 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    46,080
    Data Type
    torch.float32
    Hook Point
    blocks.6.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    apollo-research/Skylion007-openwebtext-tokenizer-gpt2
    Hook Point Layer
    6
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    756
    +0.04
    0.5%
    673
    +0.03
    0.5%
    177
    +0.03
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    372
    +0.04
    0.10
    675
    +0.03
    0.12
    756
    +0.03
    -0.14
    Negative Logits
    ulum
    -0.27
    idity
    -0.23
    gap
    -0.23
    conn
    -0.23
    tan
    -0.23
    ollah
    -0.23
    ahs
    -0.23
     bios
    -0.21
    olas
    -0.21
    htar
    -0.20
    POSITIVE LOGITS
    inguished
    0.23
    arer
    0.23
    ©¶æ¥µ
    0.23
    icient
    0.22
    ¿½
    0.22
    ritis
    0.22
     overwhelmed
    0.22
    ango
    0.22
    Crash
    0.21
    itle
    0.21
    Activations Density 9.168%

    No Known Activations