INDEX

Explanations

terms related to regular expressions and regex functionality

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

regex usage

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

Negative Logits

-0.42

/*---

-0.40

spender

-0.38

 Participant

-0.38

in

-0.37

PII

-0.36

CDP

-0.36

 Sino

-0.36

POSITIVE LOGITS

 Regex

2.23

Regex

2.16

 regex

1.77

regex

1.41

REGEX

1.40

Regexp

1.09

arangay

0.88

 regexp

0.82

 desmotivaciones

0.79

 ainfi

0.76

Activations Density 0.005%

terms related to regular expressions and regex functionality

regex usage

No Known Activations