INDEX

Explanations

instances of text indicating a continuation or reference to a previous section

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

continued from

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

 Ciencia

-0.38

leih

-0.35

 circulación

-0.35

 spectateurs

-0.35

horabuena

-0.34

 espiral

-0.34

 Ciencias

-0.33

 España

-0.33

 écoles

-0.32

 Generales

-0.32

POSITIVE LOGITS

 '\\;'

0.68

rawDesc

0.62

+#+#

0.62

-------------</

0.61

Tikang

0.59

цездатний

0.57

rungsseite

0.56

PerformLayout

0.56

WriteBarrier

0.55

 continúas

0.55

Activations Density 0.501%

instances of text indicating a continuation or reference to a previous section

continued from

No Comments

No Known Activations

instances of text indicating a continuation or reference to a previous section

continued from

No Comments

No Known Activations