INDEX

Explanations

references to additional content or items

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

more and Moore

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

enderror

-0.72

tanleria

-0.66

 للمعارف

-0.59

ſammen

-0.59

Autoritní

-0.58

<unused0>

-0.55

 للاسماء

-0.54

 ſte

-0.54

 Italijani

-0.54

ſchen

-0.53

POSITIVE LOGITS

 more

1.36

 MORE

1.17

more

1.14

 More

1.09

More

1.09

MORE

1.00

 lebih

0.92

更多

0.91

더

0.87

 więcej

0.82

Activations Density 0.018%

references to additional content or items

more and Moore

No Comments

No Known Activations