INDEX

Explanations

words indicating misinterpretations or misunderstandings

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

mistaken identity

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

WebVitals

-0.48

 surla

-0.41

 désolés

-0.39

 wür

-0.39

 eslint

-0.38

 gawas

-0.38

OfYear

-0.38

Alive

-0.38

autorest

-0.38

Filmographie

-0.38

POSITIVE LOGITS

 wrong

0.81

 misunderstood

0.81

 misconception

0.77

 mistaken

0.76

wrong

0.75

 yanlış

0.75

 misinterpreted

0.74

 falsche

0.74

 WRONG

0.73

 incorrect

0.72

Activations Density 0.056%

words indicating misinterpretations or misunderstandings

mistaken identity

No Comments

No Known Activations