INDEX

Explanations

references to location and accessibility information

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

 sollen

-0.44

 spread

-0.44

peto

-0.41

macher

-0.40

ret

-0.39

’

-0.38

 forme

-0.38

 process

-0.38

 счет

-0.37

POSITIVE LOGITS

StructEnd

0.80

 kasarigan

0.77

 дописавши

0.72

 defStyleAttr

0.72

 الدولى

0.69

 мәкал

0.69

सन्दर्भ

0.69

expandindo

0.69

 tartalomajánló

0.68

 Paglinawan

0.66

Activations Density 0.089%

references to location and accessibility information

No Comments

No Known Activations