INDEX

Explanations

concepts related to significance and difference between entities or ideas

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

before "lies" or similar verbs

np_acts-logits-general · gemini-2.0-flash

value lies in

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 22-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_22/width_16k/average_l0_72

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.22.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

 AssemblyCompany

-0.70

tagHelper

-0.51

Насе

-0.48

ără

-0.48

quillas

-0.47

 affiche

-0.46

ybė

-0.45

CWE

-0.44

ariant

-0.44

lässlich

-0.44

POSITIVE LOGITS

 lies

2.40

lie

1.98

 lying

1.61

 Lies

1.61

Lies

1.51

lay

1.45

Lying

1.44

 Lying

1.41

 resides

1.41

 terletak

1.40

Activations Density 0.503%

concepts related to significance and difference between entities or ideas

before "lies" or similar verbs

value lies in

No Comments

No Known Activations