INDEX

Explanations

phrases indicating verification or confirmation of information

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 10-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_10/width_16k/average_l0_77

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.10.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

-0.77

Rok

-0.76

Sk

-0.73

Rok

-0.72

tas

-0.65

duction

-0.65

yat

-0.65

Sk

-0.64

enzie

-0.62

Pan

-0.61

POSITIVE LOGITS

 confirmations

1.51

 Confirm

1.47

 Confirmation

1.38

 Confirmed

1.31

 confirmation

1.31

 confirms

1.31

 confirmed

1.27

confirmation

1.26

 confirm

1.26

CONFIRM

1.25

Activations Density 0.117%

phrases indicating verification or confirmation of information

No Comments

No Known Activations