INDEX

Explanations

words and phrases associated with movement or position changes

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

 Huerta

-0.55

WriteLiteral

-0.53

unauthorized

-0.52

DeleteMapping

-0.52

 الرياضيه

-0.51

飽

-0.50

 geslacht

-0.49

SerializeField

-0.48

 Marquette

-0.48

стно

-0.48

POSITIVE LOGITS

 retreat

1.04

 retreated

1.00

 retreating

1.00

 backward

0.99

 backwards

0.98

 recul

0.94

 retreats

0.93

 Retreat

0.92

 regress

0.91

 Backing

0.89

Activations Density 0.277%

words and phrases associated with movement or position changes

No Comments

No Known Activations