INDEX

Explanations

instances of the word "to" followed by non-zero actions or activities

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-2b-pt-res/layer_2/width_16k/average_l0_141

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.2.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

Onion

-0.48

навли

-0.47

empatan

-0.47

 mixte

-0.46

 căr

-0.45

 diikuti

-0.45

 braço

-0.44

thru

-0.43

льник

-0.43

idän

-0.42

POSITIVE LOGITS

 NSCoder

0.85

 Италијани

0.83

BufferException

0.69

istisches

0.69

CastException

0.67

 nichts

0.67

Keine

0.65

LEncoder

0.64

not

0.64

 שוליים

0.63

Activations Density 0.164%

instances of the word "to" followed by non-zero actions or activities

No Comments

No Known Activations