INDEX

Explanations

emphatic phrases that emphasize strong assertions or beliefs

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.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

Bronnen

-0.73

 bershka

-0.69

 myſelf

-0.68

 theſe

-0.68

ſelves

-0.67

 Theſe

-0.67

parsedMessage

-0.66

ſelf

-0.66

 ashore

-0.65

ſy

-0.64

POSITIVE LOGITS

 disambiguazione

0.68

<bos>

0.54

rest

0.51

 membuat

0.51

madas

0.51

 سكانية

0.50

 davon

0.49

(||

0.49

InjectAttribute

0.48

enumi

0.47

Activations Density 0.079%

emphatic phrases that emphasize strong assertions or beliefs

No Comments

No Known Activations