INDEX

Explanations

instances of self-reference and verification in discourse

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

correct me if wrong

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

 Numerade

-0.41

Personendaten

-0.41

Портали

-0.40

 SwitchCompat

-0.38

 Савезне

-0.35

 initComponents

-0.34

 ब्रेकडाउन

-0.34

 Wert

-0.33

 proof

-0.33

uesia

-0.33

POSITIVE LOGITS

OGND

0.48

Correct

0.47

 Correct

0.47

correct

0.47

sume

0.45

remember

0.42

Resident

0.42

torta

0.42

INCT

0.42

CORRECT

0.42

Activations Density 0.090%

instances of self-reference and verification in discourse

correct me if wrong

No Comments

No Known Activations