INDEX

Explanations

punctuation marks, particularly periods and commas, suggesting it is focused on sentence structure or transitions in text

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

position or index

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Configuration

google/gemma-scope-9b-pt-res/layer_24/width_131k/average_l0_97

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.24.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

Autoritní

-0.57

VYMaps

-0.44

 hochzeit

-0.42

roves

-0.41

 RESULTS

-0.41

AppCompat

-0.40

 esternos

-0.40

 Salamanca

-0.39

SafeAreaView

-0.39

 derived

-0.38

POSITIVE LOGITS

Према

0.41

 pleaſure

0.40

stateProvider

0.39

Thanking

0.38

°;

0.38

 StringTokenizer

0.38

ześnie

0.38

AccessorTable

0.38

]--;

0.35

 صوتيه

0.35

Activations Density 0.001%

punctuation marks, particularly periods and commas, suggesting it is focused on sentence structure or transitions in text

position or index

No Comments

No Known Activations