INDEX

Explanations

scientific notation and values with associated variability

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

mean ± standard deviation

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

 NSCoder

-0.49

]")]

-0.47

endmodule

-0.45

 Houſe

-0.44

 Außerdem

-0.44

 preſent

-0.43

 césped

-0.43

 виправивши

-0.43

 paſſ

-0.43

 alſo

-0.42

POSITIVE LOGITS

±

0.60

CloseOperation

0.47

allax

0.46

EDEFAULT

0.46

±

0.46

+/-

0.45

makeConstraints

0.43

 average

0.43

setminus

0.42

+#+

0.41

Activations Density 0.079%

scientific notation and values with associated variability

mean ± standard deviation

No Comments

No Known Activations