INDEX

Explanations

phrases that indicate routes, connections, or relationships between locations or points

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

hasError

-0.35

">)</

-0.33

-0.30

-0.29

 negativity

-0.29

 ratio

-0.29

uParam

-0.29

-0.28

POSITIVE LOGITS

 ModelExpression

0.73

 للاسماء

0.71

 فريبيس

0.69

parsedMessage

0.67

RTSC

0.65

complexContent

0.64

 arşivlendi

0.63

 gynhyrchwyd

0.63



0.61

 zwiſchen

0.60

Activations Density 0.029%

phrases that indicate routes, connections, or relationships between locations or points

No Comments

No Known Activations