INDEX

Explanations

the word "neither" and its variations, indicating a focus on negation or the absence of options

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 2-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_2/width_16k/average_l0_67

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.2.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

 Meksiku

-0.46

 substance

-0.44

onStop

-0.42

Zuz

-0.42

substance

-0.42

">//

-0.38

 Substance

-0.37

Ingredients

-0.37

 المجلس

-0.37

pedia

-0.37

POSITIVE LOGITS

 both

1.48

Both

1.43

both

1.41

 Both

1.41

BOTH

1.19

Neither

1.17

 BOTH

1.16

 neither

1.16

neither

1.11

Ambos

1.10

Activations Density 0.110%

the word "neither" and its variations, indicating a focus on negation or the absence of options

No Comments

No Known Activations