INDEX

Explanations

phrases related to support or assistance in various contexts

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

in support of arguments

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 9-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.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

CppMethod

-0.71

 tartalomajánló

-0.71

parsedMessage

-0.67

 Roskov

-0.66

richTextPanel

-0.64

ⓧ

-0.63

verifyException

-0.62

principalTable

-0.60

ftagPool

-0.59

NameInMap

-0.59

POSITIVE LOGITS

 support

0.39

 behalf

0.36

 destek

0.35

 summary

0.35

 Support

0.35

 Engle

0.34

 Ergänzung

0.34

 furtherance

0.33

 supporting

0.33

li

0.32

Activations Density 0.063%

phrases related to support or assistance in various contexts

in support of arguments

No Comments

No Known Activations