INDEX

Explanations

method calls and object references in a programming context

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 2-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_2/width_16k/average_l0_67

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.2.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

 feroit

-0.86

 păr

-0.71

 auroit

-0.69

 pouvoit

-0.69

 înc

-0.69

 Qualquer

-0.66

 avoient

-0.65

 kuiten

-0.64

 femininas

-0.64

 للمعارف

-0.63

POSITIVE LOGITS

0.59

0.47

self

0.45

0.44

0.41

<b>

0.41

0.40

0.39

0.38

Activations Density 0.058%

method calls and object references in a programming context

No Known Activations