INDEX

Explanations

actions and processes described in a narrative context

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

while or meanwhile

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

 ſol

-0.45

 leſs

-0.43

 ſen

-0.43

 plufieurs

-0.43

Tikang

-0.42

 ſtate

-0.42

 desmotiv

-0.42

Practices

-0.42

 anſ

-0.41

 ſche

-0.41

POSITIVE LOGITS

IVEREF

0.49

CloseOperation

0.47

énario

0.47

HasIndex

0.47

errHandler

0.45

AnchorStyles

0.43

aarrggbb

0.43

MigrationBuilder

0.43

 EconPapers

0.42

новништво

0.41

Activations Density 0.010%

actions and processes described in a narrative context

while or meanwhile

No Comments

No Known Activations