INDEX

Explanations

expressions of emotional states and experiences, particularly related to anxiety and burden

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

negative states and consequences

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

COUVER

-0.46

aarrggbb

-0.46

ymce

-0.42

AddTagHelper

-0.40

DMETHOD

-0.39

èses

-0.39

ftagPool

-0.39

Panamoan

-0.38

ソッド

-0.38

PreferredItem

-0.37

POSITIVE LOGITS

Jeografia

0.42

 BioLib

0.41

 ujednoznacz

0.41

 hoeft

0.40

 CallOverrides

0.40

Kaynakça

0.39

 EdgeInsets

0.39

請繼續往下閱讀

0.39

jenost

0.38

argout

0.38

Activations Density 0.082%

expressions of emotional states and experiences, particularly related to anxiety and burden

negative states and consequences

No Comments

No Known Activations