INDEX

Explanations

concentrations and their effects in scientific contexts

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

dose-dependent relationship

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

 feroit

-0.63

 ainfi

-0.57

édrale

-0.54

 propOrder

-0.54

 transfieras

-0.54

IContainer

-0.53

 plufieurs

-0.53

 auroit

-0.52

 laſſen

-0.52

لينكات

-0.52

POSITIVE LOGITS

 appropriate

0.48

 indeed

0.48

 proper

0.45

 appropriately

0.43

 correct

0.43

 properly

0.38

 specific

0.38

 consistent

0.37

Ver

0.36

Fre

0.35

Activations Density 0.303%

concentrations and their effects in scientific contexts

dose-dependent relationship

No Comments

No Known Activations