INDEX

Explanations

references to the word "name" in various contexts

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

Followed by names and colons

np_acts-logits-general · gemini-2.0-flash

Name followed by colon

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.18.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

RegressionTest

-0.73

 /\.(

-0.69

]));

-0.67

دانشنامهٔ

-0.67

 Официальный

-0.62

CompleteListener

-0.59

kheim

-0.58

UnknownFields

-0.57

/\.

-0.57

Tikang

-0.56

POSITIVE LOGITS

plate

0.88

plates

0.78

lessness

0.67

akes

0.65

Surname

0.63

úrate

0.61

ake

0.57

plated

0.57

PLATE

0.56

infall

0.56

Activations Density 0.183%

references to the word "name" in various contexts

Followed by names and colons

Name followed by colon

No Comments

No Known Activations