INDEX

Explanations

interrogative phrases and responses, indicating inquiries and affirmations

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

 pleaſure

-0.94

 Reſ

-0.92

 itſelf

-0.90

 iſt

-0.90

 Houſe

-0.88

ſelves

-0.86

 ―――――

-0.86

 faſt

-0.84

 houſe

-0.84

 ſche

-0.83

POSITIVE LOGITS

0.54

0.52

0.51

原始内容存档于

0.50

0.48

ValueStyle

0.47

ngilizce

0.47

decorators

0.46

ceros

0.46

it

0.45

Activations Density 0.089%

interrogative phrases and responses, indicating inquiries and affirmations

No Comments

No Known Activations