INDEX

Explanations

mentions of excessive numerical or technical details related to processes or elements in a structured format, often associated with programming or documentation

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

mathematical equations and technical text

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

 ſind

-0.91

 CreateTagHelper

-0.90

WriteTagHelper

-0.84

sizeCache

-0.83

ftagPool

-0.82

 unknownFields

-0.81

tvguidetime

-0.80

parsedMessage

-0.79

BufferException

-0.75

 disambiguazione

-0.75

POSITIVE LOGITS

//

0.41

<strong>

0.40

<blockquote>

0.38

FORMANCE

0.36

↵↵

0.36

 above

0.35

POINTER

0.34

<h2>

0.33

<h3>

0.33

Activations Density 0.028%

mentions of excessive numerical or technical details related to processes or elements in a structured format, often associated with programming or documentation

mathematical equations and technical text

No Comments

No Known Activations