INDEX

Explanations

the word "in" across various contexts, especially related to conditions or characteristics

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 entirely

-0.42

 fully

-0.41

 completely

-0.38

 Morin

-0.36

an

-0.33

 totally

-0.33

 nécess

-0.33

 wholly

-0.33

 hanem

-0.32

 benefits

-0.32

POSITIVE LOGITS

 autorytatywna

0.71

BagLayout

0.71

----</

0.65

aarrggbb

0.63

rungsseite

0.63

下载附件

0.61

pexpr

0.58

ScopeManager

0.57

BufferException

0.56

TargetException

0.56

Activations Density 0.085%

the word "in" across various contexts, especially related to conditions or characteristics

No Comments

No Known Activations