INDEX

Explanations

geographical locations and specific place names

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

 total

-0.45

total

-0.44

میزی

-0.41

pão

-0.41

non

-0.41

bow

-0.40

no

-0.40

 onAnimation

-0.39

 cyclo

-0.39

He

-0.39

POSITIVE LOGITS

 Lagos

1.13

Lagos

0.96

 getCity

0.81

 lagos

0.78

 Mumbai

0.74

Mumbai

0.65

 Amsterdam

0.64

stanbul

0.64

 Moscú

0.63

 lago

0.61

Activations Density 0.001%

geographical locations and specific place names

No Comments

No Known Activations