INDEX

Explanations

elements related to studies, reviews, and detailed examinations of topics or conditions

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

 betweenstory

-0.61

 NSCoder

-0.61

########.

-0.60

StoryboardSegue

-0.58

 فريبيس

-0.57

SequentialGroup

-0.55

帖最后由

-0.52

RTEX

-0.50

buttonShape

-0.50

 propOrder

-0.47

POSITIVE LOGITS

 surla

0.43

Kaynakça

0.41

Source

0.39

lucene

0.35

Referencoj

0.34

title

0.34

gulp

0.34

 Савезне

0.33

ieh

0.33

Title

0.33

Activations Density 0.034%

elements related to studies, reviews, and detailed examinations of topics or conditions

No Comments

No Known Activations