INDEX

Explanations

references to personal experiences and timelines

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

 artificial

-0.32

κης

-0.31

artig

-0.29

StrictEqual

-0.29

 Flü

-0.29

 residents

-0.28

̲

-0.28

 resident

-0.28

“

-0.27

ushima

-0.27

POSITIVE LOGITS

 ब्रेकडाउन

0.77

 resumed

0.69

 للاسماء

0.66

enterOuterAlt

0.66

 reinstated

0.63

principalTable

0.61

 reappeared

0.60

⟬

0.59

Efq

0.59

Pranala

0.59

Activations Density 0.032%

references to personal experiences and timelines

No Comments

No Known Activations