INDEX

Explanations

intensifiers or modifiers that express a degree of emphasis

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

AutoScaleMode

-0.85

 '\\;'

-0.80

protoimpl

-0.76

Personensuche

-0.74

 CascadeType

-0.74

 CreateTagHelper

-0.72

følgelig

-0.69

OGND

-0.69

 Baillargeon

-0.68

SharedDtor

-0.68

POSITIVE LOGITS

 less

0.56

 weniger

0.53

 than

0.51

 Fewer

0.50

ecznie

0.48

 looser

0.48

 menos

0.47

ủng

0.47

THAN

0.47

不太

0.46

Activations Density 0.091%

intensifiers or modifiers that express a degree of emphasis

No Comments

No Known Activations