INDEX

Explanations

string manipulation functions

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

substrings and indices

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

UserScript

-0.47

onOptions

-0.47

}{*}{}

-0.46

writerow

-0.45

 HasFactory

-0.45

Slf

-0.45

)))),

-0.43

angliski

-0.43

 للاسماء

-0.43

 kasarigan

-0.43

POSITIVE LOGITS

enumi

0.65

 betweenstory

0.55

substr

0.51

substring

0.50

Substring

0.48

Архівовано

0.47

ChildScrollView

0.45

 vanaf

0.45

 dAtA

0.43

ighborhood

0.42

Activations Density 0.007%

string manipulation functions

substrings and indices

No Comments

No Known Activations