INDEX

Explanations

references to programming constructs and specifications

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

user turn indicator

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

aarrggbb

-0.57

 '\\;'

-0.52

 OFDb

-0.50

nologue

-0.50

ess

-0.48

nuc

-0.47

rrggbb

-0.47

detectChanges

-0.46

gar

-0.46

 Ferry

-0.46

POSITIVE LOGITS

TagMode

0.52

 nakalista

0.45

kháu

0.39

MENAFN

0.38

 assister

0.36

webElementXpaths

0.36

حياتها

0.34

SequentialGroup

0.34

حياته

0.34

 externi

0.34

Activations Density 0.000%

references to programming constructs and specifications

user turn indicator

No Comments

No Known Activations