INDEX

Explanations

verbs indicating actions or changes that affect objects or subjects in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 conclusion

-0.32

-0.29

 Russ

-0.29

 disant

-0.29

 fase

-0.29

WithEmail

-0.29

 Hopefully

-0.28

 Shipping

-0.28

sos

-0.28

POSITIVE LOGITS

 ExecuteAsync

0.77

addCriterion

0.72

setVerticalGroup

0.71

⤒

0.71

تقاوى

0.70

 ویکی‌پدی

0.69

LookAnd

0.69

majánló

0.68

pushFollow

0.64

CloseOperation

0.64

Activations Density 1.119%

verbs indicating actions or changes that affect objects or subjects in various contexts

No Comments

No Known Activations