INDEX

Explanations

repetitive patterns or sequences in textual data

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

external links and references

np_acts-logits-general · gemini-2.5-flash-lite

It responds to long chunks of text (higher activation for longer input spans).

oai_token-act-pair · gpt-5-mini Triggered by @vetterc0

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

="@+

-0.68

Manbalar

-0.66

 ModelExpression

-0.66

MigrationBuilder

-0.62

 typelib

-0.62

 Wikimedijinoj

-0.62

 Roskov

-0.60

InjectAttribute

-0.60

Jereo

-0.60

Хьажоргаш

-0.59

POSITIVE LOGITS

the

0.52

 sini

0.43

↵↵

0.42

urn

0.41

 original

0.40

est

0.40

 here

0.40

 Nutz

0.39

</tr>

0.39

 Robinson

0.39

Activations Density 0.080%

repetitive patterns or sequences in textual data

external links and references

It responds to long chunks of text (higher activation for longer input spans).

No Comments

No Known Activations