INDEX

Explanations

significant instances or keywords related to technical terms or procedural descriptions

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

the beginning of the text

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-65k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_65k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

sidemargin

-0.71

<strong>

-0.62

@"/

-0.59

IVEREF

-0.56

 naselje

-0.53

 مرئيه

-0.52

<em>

-0.49

 surla

-0.49

 resourceCulture

-0.48

Excerpt

-0.47

POSITIVE LOGITS

bitField

0.70

 tslint

0.68

MigrationBuilder

0.62

]--;

0.61

glBindBuffer

0.60

μένος

0.59

mybatisplus

0.58

存于互联网档案馆

0.58

 VERSITY

0.57

LEncoder

0.57

Activations Density 0.002%

significant instances or keywords related to technical terms or procedural descriptions

the beginning of the text

No Comments

No Known Activations