INDEX

Explanations

elements related to inconsistencies or distractions in storytelling

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

problems and negative qualities

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

 ligiloj

-0.55

CrossRef

-0.49

\{\\

-0.48

StoryboardSegue

-0.48

fonos

-0.47

RTEX

-0.45

 snippetHide

-0.44

rungsseite

-0.44

rénées

-0.44

 tentative

-0.43

POSITIVE LOGITS

 horribly

0.45

 lack

0.43

 disastrous

0.41

 banget

0.41

LookAnd

0.40

 badly

0.38

 glaring

0.38

 ужа

0.37

 hideous

0.37

 terribly

0.36

Activations Density 0.019%

elements related to inconsistencies or distractions in storytelling

problems and negative qualities

No Comments

No Known Activations