INDEX

Explanations

expressions related to peaks and declines in trends or states

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 7-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_7/width_16k/average_l0_69

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.7.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

matchCondition

-0.87

 незавершена

-0.77

 pleaſure

-0.73

WebElementEntity

-0.71

 Anſ

-0.69

MigrationBuilder

-0.69

ásra

-0.68

 jää

-0.65

IVEREF

-0.63

 <=",

-0.63

POSITIVE LOGITS

()];

0.56

Bahkan

0.52

 peaking

0.52

 forState

0.50

pinephrine

0.49

postfix

0.48

()]);

0.48

 peak

0.47

estation

0.47

 persino

0.47

Activations Density 0.256%

expressions related to peaks and declines in trends or states

No Comments

No Known Activations