INDEX

Explanations

references to objects or things

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

もの・人間

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

 EconPapers

-0.52

TintMode

-0.49

лтамалар

-0.47

oa̍t

-0.44

 violación

-0.43

ьажоргаш

-0.42

__':

-0.42

 Paglinawan

-0.41

 forState

-0.41

__':

-0.41

POSITIVE LOGITS

もの

2.09

 もの

1.61

するもの

1.52

なもの

1.41

したもの

1.37

もので

1.35

ものが

1.33

ものは

1.27

ものです

1.25

ものを

1.23

Activations Density 0.002%

references to objects or things

もの・人間

No Comments

No Known Activations

references to objects or things

もの・人間

No Comments

No Known Activations