INDEX

Explanations

specific locations or geographic references

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

place names

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

 смотри

-0.43

хьтан

-0.42

Instruction

-0.42

Roman

-0.41

fiez

-0.40

味

-0.40

BIÉN

-0.40

 maks

-0.40

cektir

-0.40

ură

-0.40

POSITIVE LOGITS

ніципалі

0.49

 <=",

0.43

PointerException

0.41

addContainerGap

0.40

⟬

0.40

StoryboardSegue

0.39

 houſe

0.38

viewDidLoad

0.37

 fubject

0.37

 defaultstate

0.37

Activations Density 0.086%

specific locations or geographic references

place names

No Comments

No Known Activations