INDEX

Explanations

comparisons and variations in data across different studies or experiments

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

compare different studies, gels, batches

np_acts-logits-general · gemini-2.5-flash-lite

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Nieder

-0.29

en

-0.28

 called

-0.28

 rule

-0.27

-0.26

 appeal

-0.25

bet

-0.25

al

-0.25

 Nieder

-0.25

 fuse

-0.25

POSITIVE LOGITS

parsedMessage

0.99

 betweenstory

0.76

verifyException

0.64

ThroughAttribute

0.63

stylers

0.63

GraphicsUnit

0.62

 jScrollPane

0.61

 otomatig

0.61

 referrerpolicy

0.61

évaluateur

0.60

Activations Density 0.208%

comparisons and variations in data across different studies or experiments

compare different studies, gels, batches

No Comments

No Known Activations