INDEX

Explanations

terms related to training and educational programs

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Osbourne

-0.79

 KeyCode

-0.77

 Diocese

-0.76

 écl

-0.76

 parishi

-0.75

Edd

-0.74

 heapq

-0.73

 Wolfram

-0.73

 Neel

-0.71

 Automobiles

-0.70

POSITIVE LOGITS

 training

2.23

 Training

1.91

training

1.86

Training

1.85

 TRAINING

1.70

 trainings

1.69

TRAINING

1.62

 Trainings

1.50

 trainers

1.47

 trainer

1.42

Activations Density 0.068%

terms related to training and educational programs

No Comments

No Known Activations