INDEX

Explanations

phrases related to health communication and discussion

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.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

 Picchu

-0.48

 amet

-0.47

 желе

-0.46

 Pfeiffer

-0.45

Appearances

-0.45

 officiellement

-0.44

GenerationType

-0.44

cyn

-0.44

 relâche

-0.43

riezmann

-0.43

POSITIVE LOGITS

 conversation

2.86

 discussion

2.82

 discussions

2.54

 conversations

2.50

discussion

2.35

 Discussion

2.31

 Conversation

2.27

conversation

2.25

Discussion

2.21

 Discussions

2.17

Activations Density 0.416%

phrases related to health communication and discussion

No Comments

No Known Activations